MORTAL!“ D‘lFFERENTIALS 3N M‘CHIGAN Thesis for the Degree of Ph. D. MICHIGAN STATE COLLEGE Paul Michael Heuser 1948 This is to certify that the thesis entitled has been accepted towards fulfillment of the requirements for " - W __deql‘ee in 309101073 21ml AHULTOIWIO j; . - .__. fl.‘ 1 Major professor ‘ \" _‘ 1“ l ' C) Date ’ "-795 (‘Hfhof _ ‘ r By Paul Michael Renee:- ' J.- “c!" ...“ ' A mere '— ' .’ . 7- . " .i:~‘1\l“”‘ ‘3‘?“ I to the School at Grednete emu” or % flute College of A521 culture and Applied ‘ "" ' ' C. n pestle]. fulfillment of the requirements {*1 __ ... for the degree at C A ‘ - .Ej") ..:'t . D O ’ ‘ . . ‘ “ l ‘ e ‘ l , en - - . ' ' .2 - r ‘5 - - <- _ a . ~ TABLE OF C ON'L‘INTS PART I Intro duo t i on m!” i Page 1'. Intredu cti on La Importance of Health Studies Present Studies in Michigan D. - Problems in Studying Vital Statistics Mortality Data as Important Social Phenomena Opinion of Authorities Regarding the Relative 15 Merits of Certain Kinds of Rates in the Measurement of Mortality ’ Mgt$ty Differentials by Sea: a Problem for 16 Significance of Racial Mortality Differences 19 in Michigan Problems in the Treatment of Mortality Data 80 The Development of Vital Statistics Reporting 28 omen-J P m Death Registration System 25 Problems Relating to the Reporting of the 89 Causes of Death 116 Methede and Procedures 35 Importance of Mortality Studies 35. Objectives 37 Hypotheses 38 Measures of Mortality hployed 40 Procedures Used in Analysis of County Data 41 Order of Presentation 43 Sources of Data 44 Table and Figure Numbering System 4.5 Ii'he Appendices 46 Part II Present Health Situation in Michigan Mortality and Health in Michigan: A General View as Crude Resident and Age-Adjusted Death Rates 47 Age-specific Death Rates 49 _ and Sex Differences 58 Rural-Urban Mortality Differentials 53 Racial Mortality Differences 8d realty-five Selected Causes of Death 56 Infant and Maternal Mortality Rates 68 Cent 216909 _,,- T M”! Pulse 1!. Rural-Urban Mortality Differentials 66 Crude Resident and Age-adjusted Death Rates 66 Age and Sex Mortality Differentials '70 Mortality Differentials by Age in Various Size 81 Cities and Rural Areas Twenty-five Principal Causes, in Michigan by 83 Residence Age-adjusted Death Rates for Selected Causes 95 Twenty-five Principal Causes of Death in 9'] Michigan by Race and Residence Age-adjusted Death Rates for Selected Causes 100 by Race and Residence Infant Death Rates in Rural and Urban Areas 102 Infant Death Rates by Selected Causes 103 Maternal Mortality Rates in Rural and Urban 106 Areas Va Mortality Differentials by Race 109 Race-specific Crude and Age-adjusted 110 Mortality Rates Age-specific Death Rates 111 Box Mortality Differentials by Race 114 Rural-Urban Differentials 118 Specific Causes of Death by Race 123 Mortality Rates of Infants and Mothers 128 VI. Mortality Differentials in Michigan Compared 132 With the Nation Mortality Differentials by Age 154 Mortality Differentials by Sex 137 Rural-Urban Mortality Differentials 140 Mortality Differentials by Race 145 Principal Causes of Death 155 Infant Mortality Differentials 162 Maternal Mortality Differentials 170 Part III INTERNAL VARIATION VII. Mortality Data by county ' 171 Crude and Age-adjusted Death Rates: All Causes 1'72 he Principal Causes of Death: Cause-specific 1'78 Death Rates Infant Mortality Rates 20]. .5. I S {r ' Chapter vii Page VIII. Mortality Data by County: Residential Variation 205 Crude Resident Death Rates 205 Causes of Death in Rural Areas-Non-communicablezlo Causes of Death in Urban Areas- Non-communicable Infant Mortality Rates by Residence Part IV TRENDS 0F MORTALITY II. Mortality Changes in Michigan Mortality Trends by Age Mortality Differentials by Sex Mortality Changes by Race Mortality Change by Cause Change of Infant Mortality Changes in Maternal Mortality 1. Summary and Conclusions Objectives and Methods Findings Implications Recommendati one Bibliography Appendix I. Mortality Rates: Definitions, Computations, and Evaluations IIe 1'.th III. International List of Causes of Death IV. Mortality Differentials and Sex Roles: An Explanatory Hypothesis 224: 232 238 244 248 259 269 275 280 284 284 284 289 300 308 308 368 3'76 ILLUSTRA'I‘I ONS Hare Number 7.1 7.2 7.8 7.4 7.5 7.6 7.7 7.8 7.9 7.10 7.11 7.12 8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 8.9 8.10 8.11 Death Rates From.All Causes Per 1,000 Population, Michigan, Ayerage 1935-1944, by County. Age-Adjusted Death Rates From.All Causes Per 1,000 Popula- tion, Michigan, Average 1935-1944, by County. Death Rates Prom Heart Diseases per 1,000 Papulation, iMichigan, Average 1935-1944, by County. . Death Rates From.Cancer Per 1,000 Population, Michigan, Average 1935-1944, by County. Death Rates From.Apop1exy Per 1,000 Population, Michigan, AEEIOBO 1935-1944, by County. Death.Rates From Accidents Per 1,000 Papulation, Michigan, Average 1935-1944, by County. Death Rates From.Pneumonia Per 1,000 Pepulation, Michigan, Average 1935-1944, by County. Death Bates From.Nephritis Per 1,000 Pepulation, Michigan, Average 1935-1944, by County. Death Rates From Tuberculosis Per 1,000 Pepulation, Michigan, Average 1935-1944, by County. Death Rates From Premature Births Per 1,000 Pepulation, luchigan, Average 1935-1944, by County. Death Rates From Diabetes Per 1,000 Population, Michigan, Average 1935-1944, by County. Infant Mbrtality Rates Per 1,000 Live Births, Michigan, Average 1935-1944, by County. Rural Death Rates All Causes Per 1,000 POpulation, Michigan, Average 1935-1944, by County. Urban Death Rates All Causes Per 1,000 Pepulation, Michigan, Average 1935-1944, by County. Rural Death Rates From Heart Diseases Per 10,000 Pepula- tion, Mchim, Average 1935-1944, by County. Rural Death Rates From Cancer Per 10,000 Papulation, Michigan, Average 1935-1944, by County. Rural Death Rates From Apoplexy Per 10,000 Population, lflchigan, Average 1935-1944, by County. Rural Death Rates From Accidents Per 10,000 Population, Michigan, Average 1935-1944, by County. Rural Death Rates Iron Nephritis Per 10,000 Pepulation, Michigan, Average 1935-1944, by County. ‘Urban Death Rates From.Heart Diseases Per 10,000 Popula- tion, Michigan, Average 1935-1944, by County. Urban Death Rates From Cancer Per 10,000 Population, Muchigan, Average 1935-1944, by County. Urban Death Rates From Apoplexy Per 10,000 Population, Michigan, Average 1935-1944, by County. Urban Death Rates From Accidents Per 10,000 Pepulation, Michigan, Average 1935-1944, by County. 8.18 Urban Death Rates From Nephritie Per 10,000 Population, Michigan, Average 1935-1944, by County. 8.13 Rural Infant Mortality Rates Per 1,000 Live Births, Michigan, Average 1935-1944, by County. 8.14 Urban Infant Mortality Rates Per 1,000 Live Births, Michigan, Average 1935-1944, by County. P\ r 111 II s1- Q! .4 Cl l 1 \ PREFACE his dissertation is a systematic analysis of the general mortality characteristics of the pepulation of Michigan, and is designed to supply background data for ether health studies now underway in the State. Dr. C. P. Leonie originally called my attention to the need for such a study. Mortality studies by their nature require many hours of tedious effort and involve many procedural technicali- ties. I am particularly indebted to Dr. John F. Thaden and to Dr. J. Allan Beagle, not only for the time which they have spent with me in consultation which has been a source of constant inspiration, but also for the additional ser- vice of reading the entire manuscript and making valuable suggestions for its improvement. To Dr. Duane L. Gibson and Robert T. Hyde, I am also indebted for reading certain chapters and for making many helpful comments. The vriter appreciates suggestions and criticisms given by: Dr. Margaret Jamal: Hagood, Dr. Robert D. Grove, Dr. John K. Ryan, Dr. Bernard G. Mulvaney, and Dr. C. J. Nucsse. I wish to acknowledge the cooPeration of the Michigan State Department of Health in supplying me with certain needed tabulations, and the valuable assistance of the De- partment of Sociology and Anthropology of Michigan State College Ihich granted me the use of necessary equipment and aided in the drawing of several maps. & xiii Special acknowledgments are due the typist, Mrs. Glarence Semans, whose patience and painstaking care in typing and assembling the material have been invaluable. I wish to express my appreciation also to Gladys Harvey, Blaine Balaton, Ruth Davidson, Anne Hyde, and Angall Bavosa 'ihe supplied valuable clerical assistance. And lastly, for the constant encouragement and practical assistance of my wife, who inspired me to make my aspirations a reality, I an grateful. Paul M. Houeer University of Maryland November, 1948 'wr‘ _ _— — ' 'W"_"Tm. "‘v-fi 4' I CHAPTER I INTRODUOTI ON 1. Importance 2; Health Studies One of the challenging problems facing the United States today is that of providing adequate health educa- tion and health care for its citizens. Vance says, 'High among the traits determining the cultural adequacy of any people must be listed their vitality."l Action programs in this field, however, presuppose considerable knowledge about the incidence of diseases, mortality rates, and the cost and availability of medical care. Although in the process of every day living a variety of data about these problems have been accumulated, our scientific infor- nation about morbidity and mortality data is too meager, without added research, to serve adequately as a base for any long-time program. As we realized this, more and more interest in research proJeots developed. The people of Michigan were interested in health problems, even at a very early date. In 1852 Dr. Zine Pitcher of Detroit listed by months the forms of diseases most prevalent in his practice for that year.2 These people 1. Vance, Rupert 3., and Danilevsky, Nadia, All These la, The University of North Carolina 5588, Chapel , North Carolina, 1945, p. 335. 8. Kleinschmidt, Earl 3., aPrevailing Diseases and Hygienic conditions in Early Michigan," Michigan History maniac, v01. 85, NO. 1, 1941, PO 580 l had one cause for concern because, at that time, Michigan was considered one of the most unhealthful States in the Union} Infectious and contagious diseases were taking a tremendous tell. Dr. Earl R. Kleinschmidt, who made a study of public health in Michigan has this to say, "Some diseases such as malarious fevers, dysentery, rheumatism, pneumonia, pleurisy, cholera infantum, and croup were present during most of the year. There was a constant and “varying succession of inflamatory diseases in winter, and malarieus fevers of varying severity during the remaining three seasons of the year which, with dysentery and infantile smer cmplaint, constituted most of the diseases."2 me environmental factors contributing to epidemics were carefully noted over a period of years. Swamps and swales were contributing much to the health hazards of early Michigan residents. As late as 1881, physicians estimated that malaria accounted for over fifty percent of all illness in the state. Thus Michigan came to be known as the home of malaria. As the drainage of the land for agricultural pur- poses progressed, conditions improved. In this regard, Dr. Kleinschmidt remarks, "No less an authority than Dr. H. F. Lyster of Detroit declared that, because of the drainage of land, all forms of malaria diminished fully seventybfive per cent in and around Detroit over a period of twenty years 1" min'OMId‘, Earl Ea, 1b1&' Pe 57e s. Klein-91min, Earl 3., ibidp,p. so. I P.“ fren 18“ to 1874.»1 During the same period, sanitary conditions in cities threuaout Michigan were contributing their share of casual- ties. Tho nature of these conditions was set forth clearly by the Rt. Rev. George D. Gillespie of Grand Rapids in a speech to the Sanitary Convention held in that city in 1881. He spoke as follows: “But when man got the cobwebs of ignorance out of their eyes when they began to look abait and inquire into things what did they find?'°°° They found sewers, under prominent residence and business streets, where the amount of drainage they ought to carry off is immense, entirely filled up, so that even the slim rats themselves could not get through them: for years had they been in that condition, and the filth, and slope, and disease-breeding refuse, fran hundreds of houses and out-buildings had to be left on the surface of the ground to faster and poison the atmosphere. They found streets and alleys, in the very heart of the city, filled with rotting vegetables and animal debris. They found eess ools and privy vaults in all parts of the city fair y running over with rottenness. All these things they found and much more of a like sickening nature, and they no longer wondered that the death rate wasaalmost up to that of New York and even Lomone' While most physicians in this early period associated the occurrence of disease with environmental factors, many of their beliefs about causal relationships we now know to be erroneous. Dr. Kleinschmidt continues, "In the absence of scientific explanations for the spread of prevailing diseases, many theories were evolved to explain their cause. 1. Kleinschmidt, Earl 1a., ibid, pp. 60-61. 3. mimm... Earl 3., ibid., p. 91 from An. Rep. S.B.H., H.237“. Rev. Gillespie was President of the Grand Rapids Banitary Association. v—Vvv‘f ‘——'—7_ T— In general, disease was ascribed to the presence of filth of one kind or another and to poisonous gases coming from such aeeunulations.'1 Out of such observations came the impetus for these early research efforts. Although an examination of the public health history of liehigan shows that great inroads have been made on the control of conmmnicable diseases, ‘ little more than a beginning has been made on the control of chronic illnesses. This seems to be true also for the nation. This apathy toward the control of chronic diseases is partly a.matter of attitude. Stieglitz, in his note- worthy collaction of works on Geriatric Medicine, deplores the semen attitudes of the medical fraternity, the public health officials, and lay leaders, which assume that the chronic diseases, which take such a large toll from the middle and upper age groups of our population, are of such a nature that little can be done to prevent or stay their course. He observes, “Biologists and physicians have been strangely content to take the phenomena of aging as a :lntter of course.'2 He continues, “It is impossible in our ignorance to say what aging is, what it does, or why and how it does it. Very little is known; much of what we think we know is mrely suspicion. Some is more purely wishful thinking. But the consciousness of ignorance is 1. Ileinschmidt, Earl 3., ibid, p. 80. 8.. Stieglits Edward ed. Geriatric Medicine, W.B. Saunders (’30., Philadelphia, FennsyIvanIa, 1944, p. 3. distinctly encouraging."1 As a result of the observations of people interest- ed in the problems of health, the demand for scientific knowledge is growing. Consequently, both public and pri- vate agencies today are encouraging research projects in ._ this field. M Studies _i_n Michigan. Michigan has been among the leading states in the field of health research for many years. There are a number of influences accounting for her present position of leadership. An important factor has been her membership among the ten original death registration States of 1900. Attempts of the State to record the number of deaths occurring within her borders began as early as 1868 but further emphasis was given to this work with the creation of a State Board of Health in 18-75. lIhrough the influence of this Board and the Honorable Jonas MacGowan, member of Congress from Michigan, the National Board of Health was authorized a few years later.2 Because of the nature of early health problems in Michigan, it is not surprising that leaders in this field became aware not only of the importance of accurately recording deaths and the incidence of disease in the population, but also of the value of using such data as a basis for planning 1. Stieglitz, Edward, ed., M" p. 5. s. neinsohmidt, Earl 13., "Major Problems in Sanitation and ' Hygiene .in Michigan, 1850-1900,” M ELL-€9.92: M azine, Nels 28, No. 5, 1944, pp. 443-444. A ‘ I L .M ‘l \| I . . 1 fl . | 1 l I 1 6 subsequent health pregrams. Consequently, many of the early annual reports of the State Board of Health contain extreme- ly interesting statistics and keen analyses of health con- ditiens.1 In addition, the prestige gained for the State by the University of Michigan Medical School established in 1850 and the now very active School of Public Health have been among the important factors Operating to keep the State aware of its emerging health problems. Thus they have pro- vided stimulation for continuous research and planning in the'health field. Nor can one overlook the influence of two privately endowed foundations, in promoting and fi- nancing both research and action programs in child health and welfare, particularly in rural areas. These foundations are the Childrens' Fund of Michigan established in 1929, and the us. Kellogg Foundation established in 1930. And finally, the leaders of the 71 County Health Departments2 operating in Michigan have also been a constant inspira- tion, encouraging adjustments and improvements based on facts derived from scientific reports, surveys, and studies. There is no doubt that the healthy social climate created by these forees was partially responsible for current studies ‘ 1. See early volumes of the Annual Re orts of the State De- HE. mat fifDHealth, available In the Michigan State rary, sing, Michigan. 2. In 1943, 69 counties having County Health Departments ’ comprised 77.1 per cent of the rural population and 24.5 per cent of the urban population. Since that time Lenawee and Mach counties have organized County Health Depart- um. See Savant -first mug geport of the W W «Tatum—T943, ootnote, p.115”. 0 J! “——" i .v ‘ fl . 1':'.- 4—- g, . 1 ' .— 0 . -O'-‘ 4'. ' 4".- .I} V ’glseted or new in the process of completion. , sample of rural families. of threg Such studies A study of hospital resources and needs in Mich- igan conducted by the Michigan Hospital Survey Comittee appointed by Governor Kelly. Accord- ing to the Committee, they were to (a) study present health facilities and personnel in Michigan, (b) study related socio-economic factors bearing upon need for hospital service, and to) determine hospital needs and draft a plan which would provide an adequate system of coordinating hospital and public hialth facilities in all parts of the State. A survey of the costs of mdical care among recipients of Old Age Assistance, initiated by the State Department of Social Welfare in cooperation with the Bureau of Public Assis- tance of the Federal Security Agency. Their objectives were to study the adequacy, the availability, and the costs of medical care to these recipients of minimum income.3 A study of the incidence of disease in selected counties of Michigan, initiated by the Depart- ment of Sociology and Anthropology of Michigan State College and the United States Department of Agriculture. This study was a house to house survey of the incidence of disease in a counties, Shiawassee, Kent, and Cheboygan. Under the supervision of a State Committee, a similar study is being made of Michigan. A study of nursing education needs in Michigan which was prepared for the Michigan Council on Report of the Michigan Hospital Survey, Hos ital §_e_- W e s, WK. Kellogg Foundation, Battle ' eh, chigan Gesterhof, lillis W., "Medical Care Under Public Assis- 946. tance in Michigan " Michi *an Welfare Review Vol. 3 _ ' 1...; 1946, pp. 114; Continue n 01. I, m3. 4, 1947, ' DP. 1'3 0 ,_ :35:- Eoffor, Charles R., "Health and Health Services for 72-1—- liehigu ram Families," Department of Sociology and ‘13-. -' ,,":'threp_ology, Michigan State College, East Lansing, 533‘ "{«fif .,; ,"an, 1947 (Mimeographed). See Also, Hoffer, Charles dical Needs of the Rural Population in Michigan,“ :_ W: vol. 12, No. 2, June 1947, pp. 162-188. A Community Nursing. It was designed to supple- ment data on nursing needs and facilities collected by the Michigan Hospital Survey.1 3. Problems _i_1_i_ Studying 13.9}. Statistics Moggalitz gale... _a_s_ Impggtant Social Phenomena. Mortality data masure certain aspects of health. They represent the climax, er the culmination, of the incidence of morbidity. If mor- tality rates are high for a given disease in a particular area, one can assume a high incidence of the disease, a lack of control of the disease, or both. Such lack of control. may result from inadequate knowledge of the nature and cause of disease, or from lack of medical personnel and health facilities, or from low standards of living. If the death rate is low, however, the implications are not so obvious. Such a death rate may imply a decrease in incidence, an increase in preventive measures, or no change in incidence but an advance in treatment techniques. for example, our decreasing death rates from typhoid fever resulted largely from preventive measures which we initiated, but decreasing death rates from diabetes have resulted from an advance in treatment techniques. Obviously, then, while mortality data throw light on the health of a people, they should never be considered as a substitute for morbidity data. It is morbidity data plus mortality data that make the whole. However, to date, morbidity data have been so limited 1. lixler, Genevieve Knight, ”Nursing Resources and Needs in Michigan," Michigan Council on Community Nursing, Lansing, Michigan, 1946 . 3’, e l ' a '0 _.9 hit has been necessary to rely on mortality figures. v 0 ‘ .‘c 4 ' l_;\-. .; -~ :4 c 1r - France devotes two chapters in All. gl_1_e_§_e_ 322113 to _- *lglfil rates as measures of health and vitality of the :people. He laments the absence of figures on illness and {the-necessity of depending upon death rates to describe health. However, he scans positive that death rates can he used properly to measure certain aspects of health. He writes: 'In the United States to a large extent we have been compelled to render our judgments on the state of the people's health not frcm figures efdillness but from figures in death. This can hes-*done only by relating rates of death to at least two conditions: £1) the age and sex make- up of the people, and 2) the cause of death. Thus death, which can be regarded as a biologically normal phenomenon at the and of the life span, met in the absence of morbidity statistics be related to the age curve of the population if it is to be accepted as a valuable index. Implicit in this procedure is the idea of the average incidence of death on a standard ' population under a given environmental complex. Over a period of time the changing incidence of death on a population gives a measure or the improvement in health canditicns attendant upon improvement in environment and progress in medical knowledge and practice. To relate this to cause sf death and thus to type of illness we need to determine standard ratios of the incidence and length of morbidity to death from each cause. Such figures, if obtained, however, would last but a few years before the progress of medical services rendered them obsolete. Surgeon General Thomas Parran well stated the situation when he said that the sickness and death rates of previous years are inadequate yardsticks for the present and are useless as goals for the future. To think otherwise is to regard medical science and pullic - health as static rather than dynamic forces.“ A -‘g.;_..vse¢e, Rupert, and Danilevsky, Nadia; pp. ci§., p. etc. "3:1,; ,_.J 5' J flw'c—mw—‘V 10 Walter lillcox, one of the foremost demographers in the United States, claims that death rates are the best measure of the incidence of morbidity that we have at the present time. He assumes that morbidity is correlated with mortality, but that the correlation will decline as the death rate declines. His position is summed up in the following statement: "Since there is no direct and satisfactory measure of health or sickness, the problem can best be attacked by assuming that health is correlated with longevity and sickness with death, and measuring the death rate and length or life. Now that we have country-wide registration, the health of the people can be measured approximately by the amount of death. To be sure, a pOpulation might have no sickness and no premature death; everyone might remain healthy to the end of life and death come with no premonitory warning. This may be an unattainable ideal, but the pro- longation of life and reduction of certain dis- eases suggest that the movemant is in that direction; if so, the correlation between sickness and death is decreasing. But it has not yet dis- appeared; indeed it is still strong enough to make the statistics of death and their interpreta- tion in various ways not only the Rest but almcst the only way of measuring health." Newsholme writing in 1924 comments: " .... under present conditions deaths as a rule give more reliable statistical information than cases of sickness.'2 Sydenstricker, discussing the necessity for relying cm.mortslity statistics in analyzing health conditions, says: 'Changes in the health of the population can be l. lilleox Halter 1'. Studies in or can Demo a h , Corne’ll’University’Pres's, Ithaca, aw or , 94 , p. 222. 8. Newshelme, Arthur The Elements 9; Vipal Statistics, D. Appleton a co., NewTor—k,—l—_—924, p. 596. """"'“"'""" ‘; Q 11 measured in only one way because of the absence of suitable records. No data comparable from one period to another exist on the prevalence of physical and mental defects and impairment, and statistics of illness for representative population groups for any period of time are entirely lack- 138- '0 are perforce restricted to statistics of mortality."1 Sydenstricker continues wdth this caution: "Mortality statistics as an index of the trend in the health of a popu- lation possess serious limitations and should be used with caution .... The lower the death rate of a population be- comes, the more nearly does it approach the rate that re- presents mortality at the end of the life span and the less sensitive it is as an index of ill health at earlier ages."2 In a study of health and mortality in Louisiana published by Louise Kemp in collaboration with '1‘. Lynn Smith, the writers indicate that "The index to health and mortality used in this study is the death rate."3 Although Professor Smith based his study of health in Louisiana on mortality rates, in his book 2h; _s_9_g_i_c_:_l_9_gy g m 31.33, he points out that mortality data do not new health in the same sense as do morbidity data. He separates his discussion of the two topics into different _.__ l. Sydenstricker, Edgar, Health and Environment, McGraw-Hill Book 60., Inc., New York, 1533, p. 4 . 2. Sydenstricker, Edgar, ibid, p. 147. 5. Kemp, Louise, and Smith, ‘1‘. Lynn, Health d Mortalit % s , Agricultural Bxperimen Sta on, ou s ana "_a‘ varsity, Bulletin 390, 1945, p. 9. 12 chapters. His true position is indicated by the follOw- ing statement: "A sidespread practice among writers on this subject (incidence of sickness) has been to sub- stitute data on mortality for data on morbidity, although it has been pointed mt repeatedly that the pictures re- sulting from the two types of data are by no means identical. Accordingly, data on mortality will be presented in another chapter, and this section will be devoted entirely to materials dealing with health itself."1 It should be pointed out, however, that Professor Smith finds it necessary to employ mortality data even in his chapter on incidence of sickness. The maps of tuber- culosis death rates on page 115 of 1h; Sociology 91 M $3.11 are referred to as incidence rates on page 116. Pearl never believed that mortality rates should be substituted for morbidity rates, but each should be employed to study the particular aspects of health which it measures. He says: ”There is ever present in vital statistics, and from the beginning always has been, an attempt to make the incidence of mortality a measure or index of the incidence of morbidity, generally speaking. What actually is done is to weaken and impair the value of the statistics for the study of mortality in the hepe to make them a little better indices of morbidity.“ 1 Smith '1‘. Lynn The Sociology 2; Rural Life, Harper and . Brothers, New lork, Rev sed Edition 1947, p. 106. S harl Raymond lntroducti on _t_g Medical Biometr and ’. Etatig’tigs, W.’B. Saunders Company, EilideIphfia, Fenn., ' , ’0 650 13 m 2; Authorities Regarding _t_h_e_ Relative Merits 9; m; grid; 3; M 3.3 the Measuremep_t 2; Mortality. According to Raymond Pearl, death rates, properly calculated, measure the force of mortality. Likewise, properly calcu- lated rates of illness or sickness measure the force of morbidity. He defines the “force of mortality" as the proportion of those exposed to the risk of death who actually die. 'Thus, if 100 persons are truly exposed to the risk of dying within a given year, and 3 die, the force of mortality within the time limit of that year is 3 per cent."1 I h The exposed cases must be at risk before the result- ing rates can be called measures of the true force of mortality. Hales obviously could not be exposed to the risk of death from puerperal causes, and neither could females who are not within reproduction ages. Thus death rates, when properly calculated, measure the force of mortality, or the probability mat death will occur to a given class. Each different death rate states a different measure of probability of death. Pearl preferred to define the meaning of each rate and let it go at that, rather than to blanket the meanings of several rates under the concept of a general term such as 'health.‘ Such will be the posi- tion of the author in this study, because it seemsreason- able to state the meaning of each mortality rate and proceed to employ the rates with those particular meanings as a ___. 10 ””1, Raymond, 1b16, pe 173e 14 nasure of health. Host authorities have expressed themselves regarding the relative value of certain rates as a measure of health and mortality. Pearl very clearly states those death rates which, in his opinion, most nearly approach "scientifically accurate records of natural phenomena: 1. Deaths from all causes (either for all ages to- gether or for separate age groups, as for example, 'infant mortality' (deaths under one year of age). 2. Traumatism (Rubrics 169 to 198 inclusive). 3. Homicide (Rubrics 166 to 168 1nclusive)."l In an earlier edition of Medical Biometry _a__n_<_i_ Statistics, Pearl had included suicide in the above list, but excluded it later because "suicide is sometimes concealed by surviv- ing members of the family; how often no one knows."2 Eany authorities believe that infant mortality and maternity death rates are the best measures of the health situation in so far as mortality rates are concerned. Newsholmo views the following types of death rates as the best measures of public health work, that is, when con- sidering only death rates:3 1. A statemnt for the current year and for a series of years of the death-rate under 1 year of age per 1,000 births. 2. A statemnt of the death-rate for the first 1.. Pearl, Raymond, ibid., pp. 65-66. '. “”1, amond’ ibidP’ p. 660 5.. lewshelme, Arthur, gp. cit., pp. 595-596. 15 week and for the first month after birth. 5. A statement of the death-rate per 1,000 births from diarrhea and enteritis of children under 2 years of age. 4. A statement of the number of deaths from septic and other diseases and accidents associated with child-bearing per 1,000 births, if possible for a series of years. 5. A statement of the death-rate from tuberculosis (All forms and Phthisis) for a series of years. 6. A statement of the death-rate from: Typhoid fever Diphtheria Scarlet fever Measles Whooping cough Malaria Cerebro spinal fever and of any other prevalent infectious disease. iHe claims that the "value of these is increased when the rates for a series of years can be given."1 Professor Kumlien, in his study of health facilities in South Dakota, writes with respect to mortality rates: 'Orude, maternal, and infant death rates are probably a fairly good index of general health conditions for any given area."2 Dr. Leland B. Tate, Rural Sociologist at the Virginia ___ A ' 4 1. NO'Qhom, nth”, 1b1¢, pe 595s 8. Bullion l. I. Basic Trends of Social Chan 9 in South Mz’l’ublic’fiealth Facilities, Angcu ural'pr-Er'fient t-ation, fiut‘fi Bfiota State Eollege, Bulletin 334, 1940, pp. “7e 16 Polytechnic Institute, lists the following rates as measures of the health status of a.p0pu1ation: (1) the infant death rate, (2) the maternal death rate, and (5) the general death rate. under "Reflectors of the Present Health Situation," he writes, 'The health status of a people is reflected by a number of things, such as infant, maternal, and general death rates, and the prevalence of certain diseases, defects, and deficiencies.'1' He quotes Dr. George M. Lawson, Pro- fessor of Preventive medicine and Bacteriology, University ef‘Virginia School of medicine, as saying that "the infant mortality of any area is the best single indicator of health." Thus there seems to be consensus to the effect that crude death rates indicate the trouble spots in the health situation, while the infant and maternity death rates, and death rates from communicable diseases, particularly if taken.over a period of years, seem to indicate the effective- ness of the public health program. lortality Differentials 2y fig; _a_ Problem £93.; m. Stu- dents of demography and vital statistics have known for many years that males have higher death rates than females, but, to date, few studies have gone any further than to show that there are differences in the death rates by sex. The English statistician John Graunt (1620-1674) observed that nudes hare a.higher sex ratio at birth but that ultimately ' 1 late Leland B. The Health and medical-Care Situation ’ ' 21%], 71 ini , Buiietin 333, Virginia Agricuiturai apex- nt at on, Blacksburg, Virginia, 1944, p. 10. A 1'7 females constitute about half of the population.1 Professor Willcox showed in 1966 that female death rates have improved more than male rates from 1900-1960. He pointed out that fact, and observed that "it is a point inviting analysis."‘2 However, he offered no explanation. In another study, however, Willcox attributed most of the excess of male over female deaths to biological forces.‘3 Sydenstricker points. out the increase in death rates from certain degenerative diseases, and then summarizing from Iiehl4 he says, 'Wiohl, after a careful study of the sext-age-specific rates for the principal causes of mortal- ity among adults, made the interesting observation that the upward trend in these diseases (except diabetes) was more pronounced among males than females, especially in the ages 35 to 64 years, and pointed out that in‘the short period considered (1921-1927) the greatest relative in- creases were in heart diseases and accidents, especially among males.“5 Again Sydenstricker offered no explanation l. Graunt, John, Natural and Political Observations Made fig the Bi%s-_3?-Mor-tETit—T—, edited‘F—y wa‘it' “Tar .-'iTlcox) a 3‘5 op ins Press, B's-itimore, Md., 1939, p. '71. See Also Duncan, Otis Durant, Social Research 93 Health, Social Science Research Councii, New York, 1546, p. 22. 2. Iillcox, Walter F., Introduction to Vital Statistics 3; United States 1905 _t_g 9%, Bureau of the Eensus, ‘ h Deton, D. CO. 1533, PO 4 O Iillcox Walter F. Studies in American DemoEraghy, Cornell’University’Press, Itfica, New Tor , 9 , p. 120. Iiehl, Dorothy 0., “Some Recent Changes in the Mortality ng Adults,a Journg, 3; Preventive Medicine, Vol. IV, .0 ‘3' my, 1930. . m‘n‘tr10ker, Edgar, _O_Be 0115c, Pa 1680 18 ed? the sex differential. Pearl presents data showing that M... the age specific death-rates are normally lower for females than for males at all ages except possibly at the extreme ends of the life span, where certain difficulties prevent, the accurate collection of the data necessary to establish specific rates with the highest degrees of reliability."l Pearl found that not only were age-Specific rates generally lower for females but also that from birth to age20 and from age 50 up, the declines in death rates age-specific were relatively greater. He offers a biological explanation. He says, "Now it is seen that also the female death-rates, within the stated age limits, decline relatively more than the male in a secular period when mortality is generally improving. This is another indication of the innate biological or constitutional superiority of the female sex. For it is plain that public health efforts during the period were not directed more towards females than males. Furthermore, females were not less exposed to general environmental hazards in 1930 than in 191.0."2 There is one outstanding exception to this trend. In diabetes mellitus, the males are the favored sex. Borts points out that in 1938 in the United States, there were 9,967 male deaths from diabetes mellitus among those 50-89 1. Pearl, Raymond, 92. _c_i_t_., p. 181. S; Pearl, Raymond, M p. 183. W'I'JV ".".n ' ' Le" -. 19 “ge as compared with 20,740 female deaths.l He _' _w’: ,ri’! ' i M .191th ion. 4 '( , a a ‘ :‘u in the opinion of the writer, mortality-sex .1 '~’£,S,F'4nt1als are not adequately explained by the bio- ’ 410-1 causation hypotheses of Pearl and Willcox.2 _a‘ are too many variations in the differentials for pick there are no satisfactory biological explanations, ‘ Aid for ilich socio-economic explanations are more con- . ‘.fioing. (It is conceded that biological explanations L more acceptable for certain phenomena.) Many of the 5" ’ferentials by sex may be explained in terms of the a ' Mal roles of the sexes and their relation to the total lial structure. The testing of this socio-cultural ° {1.— . J _ithesis is outside the scOpe of this thesis but in “was“: Iv, it is elaborated as a partial explanation of ’ :Ceaesu or death. :_3' . e co gg________ Racial Mortality Differences _i__n Michigan s‘earch in racial differences to date has not revealed any "ishorifid biological differences which might explain varia-z ' 1 ,gems in mortality data among races. By and large, dif- Mile” in mortality for racial groups do not exist because 1 thinlerent racial differences, but rather because of «a one hidden classifications, such as economic differences XL. 1n, Edvard. adores: _E“ n PP‘ 23"“228' M1, Raymond, 22. _c___it., p. 183 and Willcox, Halter , C 1nd, P. 1200 A mgr—3.1..- 3"». .2} 3" ' 2 . ‘ 30 m classes, within racial categories. Linder and Grove explains: “It is easy to overemphasize the value of a racial break-down for mortality and natality statistics. The mortality rate is obviously and necessarily a function of ago. Very little evidence, however, exists to show that observed differences in mor- tality for various racial groups are due to inherent racial differences. Race is not independent of other factors, and the economic, social, and medi- cal circumstances of one racial group may be quite different from those of another. When deaths are classified by race there is created also, for example, a hidden classification of economic status to the extent that economic status is correlated with race. An observed difference in mortality between races may in actuality be no more than a difference of mortality for different economic classes. This observation applies to the inter- pretation of data classified by any other characteristic, but it is particularly important in the case of racial classification." Since the non-white population in Michigan referred - to as “all other races" numbered but 216,463 in 1940 and comprised only 4.1 per cent of the total population, race as a variable offers little by way of explaining the total mortality health situation in Michigan. However, one chapter is devoted to race mortality differentials in lichlgan e 2 Egblems in 3h; Treatment 3; Mortality Q3133. _ (a) Adjust- ing for Age and Sex. In most populations, differences in the age of the population group are so great as to make it necessary to adjust for this factor. As Linder and ' 1.. binder, Porrest E. and Grove, Robert Diéo‘m 0 .‘_'. ° . t tic ggteg in the United States - , ‘.‘ ~ . ~ % . an of Censusf'Washingion, D. 6., 1943, p. 12. .E ‘: - Q ‘. .,_ r-ztl‘ * s- sip-“Sec chapter V, “Race Mortality Differentials." \Y'_ - Iv- . m .m- iii-i 21 (trove1 point out, it is usually also desirable to adjust for so: although it ordinarily makes little difference in the result. (b) Treatment of Unknown Ages. One of the problems which confront investigators interested in mortality data is the number of cases which cannot be assigned to an age frequency because the age is unknown. In some states this number is very high. However, in handling Michigan data it is not a serious problem since the number of "unknown ages“ is seldom over 25 in any one year. This small number can be disregarded without affecting the results. Linder and drove“3 do not attempt to distribute unknown ages, but include them in their "all ages" class in each table. According to their figures, in 1940 there was only one state in which the percentage for unknown ages at death exceeded 1.00 per cent, and in all other states the figure was less than 0.40.3 (c) Use of Age Limited Rates. The Committee on Perms andllethods of Statistical Practice,‘ an authority on the use of ”age limited rates," recommends limiting age 1. Linder, Forrest 1!. and Grove, Robert D., 22. 21.1., p. 68. 2. hinder, Iorrest E. and Grove, Robert D., $11.95 p. 11. 3. See discussion “Problems Relating to the Reporting of the Sauces of Death," p. 29. 4. Linder, Porrest R. and Grove, Robert D., _qp. cit., Poet note, p. 85. Pan 4‘ .81 .II ‘I‘ "1 01 I‘ A. 28 group populations for a specific disease to those age groups in which the deaths occur. Discussing "age limited rates" they’state: "These are simply broad age-specific rates designed to:neasure the mortality from.a single disease or group of related diseases. For example, instead of finding the death rate fran communicable diseases of childhood based upon the total population, it is found specific for the pepulation under 15 years of age. For heart disease a broad age group from.the middle to the end of life may be used. This practice helps to reduce the distortion in cause-of-death rates that is due to varying age distribution.” 2!; Development‘ggflxgggl Statistics Reporting. Vital sta- tistics have been defined as "the numerical study of groups of human beings politically organized,"1 and in this sense the term is synonymous with demography. The founding of vital statistics is attributed to Captain John Graunt (1620-1674) by most writers. Graunt made an analytical study of the Bills of Mortality of London and in 1662 pub- lished.his findings in a book entitled Natural ggd Political gpgervations ypntioned‘ipng Following Index and Made upon the Bill! g Mortality. Known Bills of Mortality, that is lists 'ef burials, baptisms, and.marriages date back as far as 1532, but not until Graunt did anyone attempt to analyze them: statistically with the idea of deriving generalisations about the nature of vital statistics phenomena. He is credited with having demonstrated at least four important facts. These have been summarized by Pearl as follows: l. 'lillcox,‘la1ter 3.,‘92. cit., p. 195. 23 "Pirst, he made clear the regularity of certain vital phenomena which appear to be merely the play of chance 11 their individual occurrence. Second, he pointed out the excess of male over female births, and the approximately equal numbers of the sexes in the pop- ulation. Third, he demonstrated the relatively high rate of mortality in the earliest years of life, and finally he discovered that the urban is higher than the rural death-rate normally.”- lo small contribution, also, was the interest that Graunt stimulated in others to examine vital statistics data. Perhaps, the next most significant event in the growth of vital statistics was the preparation of the first life table based on what might be considered really sound principles. This work was done by the English astronomer ldmund Halley (1656-1742) and was published in assay form in 1693 under the title "An Estimate of the Degrees of Mortality of Mankind drawn from curious Tables of the Births and Punorals of the City of Breslaw.“a In passing, one should also mention the work of Adolph Jacques Quetelet (inc-1s“), who applied the skill of a trained mathematician to the task of developing a more adequate methodology for the analysis of vital statistics. with the development of an adequate theoretical back- ground for the analysis of vital phenomena, it remained for Dr. lillian Farr to insist that such theory be applied toward the solution of practical problems and the attainment of definite goals in the field of public health. Conse- quently, he adapted statistical procedure to practical ends. 1. lhe Ronald Press Company, New York, 1933, pp. 42-43. Pearl, Raymond, 93. cit., pp. 24-25. :" ‘- 3,: (notes from Greenwood as follows: s...But in the .1; . . '. st forest of symbols Farr never loses sight of, and a new” allows his companion to lose sight of, some definite . and. concrete end which he proposes to reach. I... The greatest mathematical statisticians of the first 11qu of the ninteenth century were not Englishmen; we have not to our credit any theoretical work of that ._ date which will compare with the researches ofgLaplace I and of Poisson in France or Gauss in Germany; but of no .. civilised country can a record of fatal disease be con- structed with the precision which appertains to the medioc- 'statistical'history of England and Wales since 1840.71 in the new world, Lemuel Shattuck, born in 1793, hotel. interested in the problems of public health, perhaps because of the loss of four members of his family from ‘ *- ‘ tuberculosis, and through this interest introduced several major reforms of a practical nature into vital statistics procedures. Although a layman, Shattuck is credited by _ Iillcox with the following achievements: 'Shattuck was the leader in founding the American Statistical Association. 'Re was the main agent in carrying. to a successful conclusion the prolonged campaign for that effective system of registering births, marriages, and deaths in Massachuesetts, which has now expanded to include every state in the Union. .' _ "Through his influence and that of his Boston ' '- " ' ' census of 1845 upon the Federal census of 1850, the “ A -rv‘.4 Raymond pp. cit. p. 32 from Greenwood, Major, ' fitatistics, most, May 7, 1921. "-e . t.‘ ., ‘ - . 'O -— . t , _‘ 5. .-'- I s. 2' -HH. .... ,. \m . 3’ e ‘5 I - ‘z-O L I a. . V -_ . 7 . - ‘5 .. " 25 3 _“ee‘nsus practice of the United States was modernized) ' 'hrough his Plan for a Sanitary Survey of Mass- achusetts to be executed by a projected State Board ' of Health, he contributed more than any man of his generation, professional or lay, to the improvement of Amrioan public health and preventive medicine. Such a State Board of Health, was established in lassachusetts 20 years after his report recommended it and then imitated in State after State. . “through the 50 recommendations embodied in the flan he anticipated nearly all the public health measures not based on bacteriology which were intro- ‘dueed in the following two generations.'3 John lbw Billings (1838-1913) while attached to the ruin of Gensus was the first man to suggest machine tabu- ' lation of vital data. These machines were employed success- fully for the first time in 1890 and greatly increased both the accuracy and volume of work.3 ,mm Qgistration m. is first used the term vital statistics applied chiefly to mortality and morbidity data, aid in the United States was identified withregistration gtetisties rather than census figures. This separation was acmu since the legal basis of the two reporting systems pried. the taking of decennial censuses. was required in _ our constitution and was under Federal control... The first federal census in the United States was in 1790 but it was 5 r. not until 1998 that the Bureau of Census wasmade a perma- " _. ‘ . _L .. ' a“ one valuable-innovation was the substitution of the 3. --~* individual for the family as a census unit. 51.27.311.001. Walter F., 22. cit., p. 4.79.. gnu... Walter 1., ibid., p. 488. — s .._d5. 4. '. - .3- 26 neat governmental body.1 Begistration statistics, on the other hand, were the responsibility of the state and local governments and had a more spotty development. State figures were then pooled in a Federal report. The first states to report for a national total were Hassachusetts, New Jersey, and Washington, D. 0., in 1880.8 After the Bureau of Census became a permanent group in 1908, it sponsored a national registration system for both deaths and births. Perhaps because American citizens were more interested in the public health movement than in sane other aspects, the registration of deaths progressed more rapidly than that of births. The first step was to collect all the laws, ordinances, and statistical forms pertaining to the registration of deaths in use by the states and local units from which a model form was drafted. The American Public Health Association pioneered. in this work and later it was endorsed by the American Medical association and similar bodies. This became the basis for a uniform Federal Reporting System. a model state law was also drafted and submitted to Congress for approval with the reconnnendation that it be adeptcd by the states. Most of the states passed the model law or one similar to it. Michigan was among the _ten states included in the original death-registration States of 1900. 1. Pearl, Raymond, _op. 333., pp. 22-24. a. Linder,l'orrest B. and Grove, Robert D., 22. g_i_t_., p. 96. .- . . I ,‘1._.I.:H. ktihflgsnty,states were in the death-registrat1on ' lg£§m1980 the number leaped to thirty-four, and bit": 1;! 1933, for the first time, all of them were ’égd to both the birth-and death-registration.systems.1 9- {shin features of the death-registration system.provide: Ex :a91‘31. .1 state law controlling the reporting system. ‘1 j, Natig‘ Provisions in the state law which prohibit the , t .‘ disposal of a corpse until a permit has been r if”) ”Cuere . ._‘ ., . .s datum S. Provision that a permit cannot be given until .;;-__‘ a certificate of death has been completed on ‘ ' a standard form. This form.includes a sta- tistical description of the deceased and-a professional statement of the cause of death. d. completion of the certificate of death by the undertaker with the necessary assistance of friends, physicians, etc. . 3.5.. Provision for a standard nomenclature and 7 ‘“ classification of causes of death excluding vague statements of cause. These have been revised periodically. Provision.that the local registrar file with . the State registrar each month the certificates which he has received. “5 9. Provision for state supervision and follow-up _ by the State registrar. .8. Provision for the purchase of copies of the ‘ ' certificates from.the state and local units by .the Federal Government, to be turned over to the Bureau of Census for the preparation of .- -data to be made available to the registration 03‘ States. fidmdsr, Ferrest B. and Grove, Robert D.,‘_p. cit., pp. ”.”" e t t.‘ . ;‘-“ “a, 'dth‘ Fe, Je Cite, Ppe 202-303. 99;: :'- 'u‘; 9 9“} 9L9 3'3" 7 pi e 3 ~» 13:; .e ‘4“ I t | l .2" _ . fidnittance to the national registration system.is achieved (1) by'demonstrating that at least 90 per cent of all deatis or births were registered, and (2) by passing a state law controlling registration. It is believed by most authorities that death regis- trations are more complete than birth registrations although no Ration-wide test of death registration has ever been made. Some comparison may be made by observing the results of the Nationewide test of the completeness of birth registration made by the Bureau of Census in 1940. According to this test, only 7 of the 48 states had a.higher percentage return_ for birth registration than Michigan, Which averaged 97.8 per cent. Returns from.Michigan were more complete than from all other States in the East North Central Region.1 Iillcoxa presents data to show that the nine states with the lowest death rates in 1935 averaged only about half as long a period in the death-registration area as the other forty states including the District of Columbia. It is evident that some of these States with death rates under 10 per 1,000 population, for example Arkansas, have low rates partly because of incomplete registration returns. It is estimated 'that only 75.9 per cent of the births in Arkansas were re- gistered in 1940.' ‘lillcox applied the same sort of analysis to death rates by rural and urban areas in 1930, and showed.that -__As A .A‘ 139: tinder, Forrest B. and Grove, Robert D.,‘gp. cit., p. 99. , -. -- Lass-(I4 .‘...,I1vll°°3: Walter F., pp. cit., p. 211. ., of" , n , "‘ . . . 29 I "~ fifipal rates were more equal to urban rates in those states 119}; T where the registration system.was initiated earliest. In the Iest South Central States, the rural death rate was only 8.3, and in the west Nerth Central 9.1, while the urban death rates for these regions were 15.6 and 13.1 per 1,000 pepulation respectively.1 This seems to be acceptable evidence for Iilleox's contention that death rates tend to be trustworthy only after a state has had ample time to perfect its registration system. However, it does not follow that the longer a state has been in the registration system.the more complete will be its registration returns. lhcre there is reason to believe that the number of deaths are incommlete, death rates should be interpreted with great caution and where possible adjusted for under-regis- -tration. Per example, there is reason to believe that lhcuisiana.has a fairly high amount of under-registration of deaths because of its low percentage of birth registra- ticns (85.1), much of which is likely to be in rural areas. Thus, the low death rates in this State for rural areas (8.7 per 1,000 in the rural areas for all causes as compared with 13.7 in the urban districts)3 may be due to failure to report all deaths that occur. m gelating to the Reporting _9_f the Causes .9; Death. There are several sources of error which influence the original data on which specific-cause death rates are based, ,-11cox, Ialter 1., ibid, p. 213. I . — . J u q ' V —. 7 ; ', Louise and Smith, T. Lynn, 22. cit., ‘p. 14. .-.- 3:3 _. ghen they are studied over a period of time. 3 \‘t; i , fifths important ones are listed here and are followed .4 3..';;;: -‘ . ' .* ‘3, .3 - pydiscussion. a '§..'«cs 1.3"-‘11'1‘1. Periodic revisions of the International List ' . of Causes of Death. , p s :am 3... 7' 8. Proportion of death certificates with only s wide. ;-fi . , one cause of death given by the physician. in... 5,3. Variation in the accuracy of the diagnosis ' . _ , 3. of causes of death by physicians among states, 10mg, and in rural versus urban areas. a. Huber of deaths classified as “ill-defined and unknown causes." . fl. Under-registration of deaths. . ”33.331? in the development of the statistics of mortality, it as” obvious that if data were to be comparable from fidelity to locality a uniform system of classification was fihsury. In 1853 the First Statistical Congress, held in eels, appointed William Parr and Marc d'Bspine to draw #flcb a classification. Their system, subsequently pre- ierel' -was adopted in Paris in 1855. It was revised in 18‘, 1896, 1880, and 1886. In 1893, arrangements were made fii’ decennial revisions and the last one was completed in 4.3” obviously data collected prior to 1938 must be ad- fistcd for this factor before it can be considered compar- 3‘3 insofar as possible the Bureau of Census indicates Kfierahe causes within specified dates. another factor ecmplicating mortality data is the lag number of death certificates being returned with 4.. A ,Baymond, _ep. cit., pp. 57-58. - e: -‘I’ -) 1;, run-1, Raymond, ma. pp. 60-61. ,I). 31 ' i (O ’jfieit than one cause of death listed. Pearl makes this . starvation, ‘It is interesting to note that in the 20 year‘s from 1917 to 1936 inclusive the percentage of such certificates in the United States Registration Area rose from about 35 to almost 80.... -- a reflection of a more widespread desire to report causes of death as accurately, intelligently, and honestly as possible."1 In 1929 the Ccnission for Revision of the International List of Causes of Death requested the United States Government to study the problem, and Dr. Timothy P. Murphy, then Chief Statistician of the Division of Vital Statistics, United States Bureau of the Census, initiated the study. He cupiled a set of 1032 cases containing from two to five causes of death and submitted them to the statistical offices of 18 countries. Seventeen countries replied. The largest consensus regarding the primary cause of death was for cancer and other tumors with an average of 77.4 per cent.‘ At present, the offices of vital statistics operate under a set of arbitrary rules. But, as the study mentioned above indicates, considerable work must be done in revising and interpreting these before international data can be presumed comparable. Variation in the accuracy of the diagnosis of causes of death by the physician is another factor which introduces lg Icarl, Raymond, ibidc, pp. 59-60. rr " .‘ :37?- - . " . .sfgprrel‘ into mortality“ statistics. Pearl observes, . ‘ *ihysiciens all of the time and all physicians some of .t 3 fine, will use their own terminology instead of that of ; grnational Classification in reporting the cause of fight)! on the original death certificate .... The only son- ‘1_ gimwable remedy is the slow and tedious' one of education ’ W to ever closer and more precise cooperation ...."1 ' p. as» errors tend to be greater in less advanced and isolat- It“. at areas of the nation because they are largely dependent *i“ apes the efficiency and cooperation of physicians and '3; administrators of the registration system. '3. 4,. Ierhaps one of the major sources of error is the "J: me:- of deaths classified as "ill-defined and unknown ‘1 .u gains." There is great variation among the states in this .; 't papaet. In 1940 New Mexico had more than 20 percent of ; its total deaths so classified as compared with 9.01 per 4 I l i 4; | ,,‘ ! ..‘ “east in lhede Island. Obviously, a comparative study of 3 3" .6: the causes of mortality between the two States would have 3 little value. In trend studies, it is important to note 1 1' ._.the proportion of deaths classified as “ills-defined and It'appewnfleauses' because the increase in some. of the specific :_ equaes play result from a decrease in deaths assigned to ill- 3-3 game, and unknown causes. In states and geographic areas 3:. “flare it is known that a high preportion of deaths are unre- page er are classified as ill-defined and unknown causes, -v.‘F. _ ‘- _- . ~ .. l .— -\ b i.4llilliiao 53 there is no really satisfactory way to adjust data for a study of causes of death. Obviously, mortality studies of causes in these areas should be interpreted with great caution. llany physicians not only report inaccurately but in some instances fail to report at all. This results in under-registration of both births and deaths. For some localities, these emissions are quite large, if results of the Census Bureau Nationrwide test of birth registration completeness is any indication. Thus, it appears that the best known indices of the influence of these various sources of error on mortality data are (l) the preportion of deaths classified as ill- defined and unknown causes, and (2) results of the Bureau ef Census Ration-wide test of birth registration complete- ness. If the state ranks low in completeness of birth registration.and high in ill-defined and unknown causes, then the inaccuracies resulting from all of the other sources of error listed above are likely to be present also. Inlflichigan, according to the Nationrwide test, 97.8 per cent of all births were reported for the year 1940, while only 4.1 per cent of the deaths were classified as ill-defined and unknown causes. We are assuming, therefore, that lichigen is among those states having the more accurate Iertality data. Specific-cause death rates are among those included in this study. Ttentybfive causes of death listed by the Bureau of Census as being comparable over a period of time are presented MW!” at the population.1 In an effort to ;~ fiscal-aw, the change in the death rates tree is shown for the period 1910-1940. All rates Coal; 'mpeaths allocated to place of residence unless . '3. 3' . ated. I '2.“ r .7 ‘. "’ a “'4, . . P» e ’- )‘l g . . ' l V . 4 _or {1.1 i‘. A: J‘- ‘ 0 .ll. ' I . , ,_ ,o 'Iw . . . , . a ‘ a hd J . S 3 I his a). ' ‘ I ' l .- 1 I . . C an I w. a, th . .. . 3:." hath-1 '3 - ‘ =3 «.n'g": ~: . 3 - .'. t l‘mr ,. ,5. e? ,. , ‘.-. as . ,-___,_fi.,, , . ah\ 1' 3 :_p ‘eylrS. ‘f‘ 33‘ hit-25 fi'«:~ "‘ 81.15331“: ‘ -.¢u _‘V .3,,.‘"._ ’ 7‘ 7 "x A _‘ ‘- 4 .k f #1.“ .ATA_ A._,..__V 3.4 a, a; 9* lt It and Grove, Robert De, n) my ‘( .' ski '1 01‘ {3—5- ‘74 t ‘ - CHAPTER II METHODS AND PROCEDURES We 31; Mgtality Studies In Chapter I, the importance of health studiesl vas discussed briefly from a general view, but in this section a specific emphasis is placed upon the value of analyses of the incidence of death. The ability of indi- viduals, or groups of individuals, to live long and to 'stay the hand of death" has not been an object of scien- tific study until relatively recent years.2 However, there has probably always been an interest in the subject. In the words of Dublin and Lotka, "From time immemorial, the l. , 8. 3e length of the human life span has been a topic of the .keenest interest. The poetic reference to it by the Psalmist is in the mind of every one. That this in- terest in longevity continues to date is seen in the frequent references in the daily news to the span of life, and to the gains which have been made in the average duration of life, especially within the last century.“ Aside from the curiosity and interest manifested by the ordinary citizen in such questions as, "How long will I live?” or "he dies in my community?“ systematic analyses of mortality rates in communities and states may be Justified F in part, at least, by the following: 5.. pp. 1’5e See pp. 22-25 for abrief discussion of the history of Vital Statistics. Dublin, Louis I., and Lotka, Alfred J., _o_p. cit., p. 111. 35 in 1.. I 111 l 36 1. Hortality data are more completely and accurately recorded than morbidity, and for this reason they are more ccnonly employed in analyses of the health of a pepulation. Reporting of deaths is enforced by law, and the rules and reg- ulations of the death-registration systanl are now in force in all of the States. Michigan was included in the original ten death-registration States of 1900, and with this long history of participation in the national registration system this State may be considered among those having the more \ accurate mortality data. * 2. Hospital planning now under way in many States, makes use of the number of deaths taking place in the com- munity in estimating facilities which should be provided. } This is especially important because deaths occurring in hospitals are increasing. The Michigan Hospital Survey, for example, related the number of deaths in the hospitals of a commity to the length of stay and computed an index which they used in determining the need for hospital beds.2 3. Health campaigns having financial and educational goals t.ond to loan heavily upon the arguments supplied by the increasing number and rate of deaths from pertinent causes. One needs only to mention such research and educa- 1. Bee pp. 25-89 for a history of the death-registration systu in the United States. I. Report of the Michigan Hospital Survey, 0 . cit., Chap- ter VII, p. 101 for a discussion of the ed-Eeath ratio and its use in estimating the number of hospital beds } needed. "' A. lhat is the present mortality situation tienal programs as those presented in behalf of cancer and heart discuss control. Questing The general purpose of this study is to describe systematically the mortality characteristics of the popula- tion of Michigan. To achieve this end, the analysis will be centered around the following general questions: 1’in Michigan? B. How does present mortality in Michigan compare with the Nation? 0. Are there great variations among the counties? D. lhat are the principal changes in Michigan com- pared with the United States? The specific items to be analyzed are listed as follows: 1. The principal causes of death and their relative rank in Michigan, for rural and urban areas. s. Age, sex, and race mortality differentials in Michigan. 3. Mortality in Michigan compared with the Nation classified by age, sex, race, residence, and principal causes 0: dOlthe 4. The principal causes of death in counties of Mich- igan by. rural and urban residence. .1; The present situation is defined as the year 1940 or averages centering around this year. . a $ Le ' .I.‘ scle‘ 38 5. Changes in the principal causes of death in Mich- igan and the Nation, 1910-1940. 6. Changes in age-and sex-specific death rates of Michigan and the Nation, 1910-1940. 7. Changes in.mortality differentials by race, unch- igan and the Nation, 1910-1940. Hypothegg; .Although death rates tend to follow the same general patterns throughout the United States, there are sufficient variations in social, economic, and other environmental con- ditions to Justify the testing of several widely accepted generalizations of mortality phenomena. Proven propositions based on social phenomena do not necessarily remain true from one period to another. Changes in mortality rates are in part reflections of variation in the pepulation composi- tion. Mblyneaux probably expresses a rather widespread re- action when he writes concerning tho generalization that the urban is higher than the rural death rate, five have reached a time when it becomes necessary to qualify this proposi- tion.'1 ‘.As time progresses, some mortality differentials are becoming less and others greater, and further these changes are not likely to be identical in all local areas and in all l. Mblyneaux, Lambert, Gilliam, Sara K., and Florent, L.0., 'Differencos in.Virginia Death Rates by Color, Sex, Age, and Rural or Urban Residence,“ American Sociological Review, Vol. 12, No. 5, 1947, pp. 552-533. 39 parts of the country. If we know that, in general, non- whites do have higher death rates than the white population, this knowledge provokes equally important questions as to whether the magnitude of differences is greater in urban or in rural areas, or in one region as compared with another. Thus, this dissertation will be concerned with such natural inquiries as: Do males have higher death rates than females? Is Michigan an exception? Do sex diff- entials vary among segments of the pepulation? Does a particular racial, sex, or resident group have higher rates from a certain disease than another? Are the mortality characteristics of rural residents in highly urban counties the same as those living in predominantly rural counties? These, and many other questions considered in this disser- tation are listed below as prepositions and are considered the hypotheses of this study. 1. Urban death rates are higher than the rural. 2. Colored death rates are higher than the white. 3. Males have higher death rates than females. 4.‘ Death rates have declined in all segments of the population. 5. Small cities have higher rates than large cities. 6. Il'hc Nation has higher death rates than Michigan. '7. Mortality change is greater for the Nation than for Michigan. 8. Sex-mortality differentials are increasing. 9. Mortality declines for younger ages are greater than for older ages. 4O 10. Race-mortality differentials are decreasing. 11. Infant and maternal death rates are decreasing. 12. JMortality rates for chronic diseases are increas- ing. 13. Sex-mortality differentials are greater in urban than in rural areas. 14. Rural death rates tend to be high in counties where urban rates are high, and low Where urban rates are low. 15. There is an inverse relationship between size of city and mortality rates. The implications of a.mortality generalization are usually matters of interpretation or deduction. However, inferences as to general health conditions may be more accurately indicated by certain rates than others. For ex- ample, infant death rates, tuberculosis, and mortality rates for other communicable or preventable diseases are frequently employed as measures of the health level of a population. Such implications, or possible explanations, are indicated frequently in this dissertation. However, no attempt is made to test hypotheses involving causes or explanations of high rates. A social-causation hypothesis of mortality differ- entials is briefly elaborated in Appendix IV. Measures gt; Mortality Employed The number of deaths per 1,000 or 100,000 population is the basic measure employed, although it is usually refined by (1) computing the rate for specific classes of the popular tion, such as, age, sex, race, and residence; or (2) adjust- ing, or standardizing, the rate for age. These measures are 41 commonly referred to as crude death rates, specific death rates, and adjusted death rates.1 In general, the more re- fined the rate, the more specific the meaning and thus the less chance there is of misinterpreting the rate. With the exception of infant and maternal death rates, deaths are related to the population as enumerated by the Federal Census in April, 1940, rather than the estimated mideyear population as is customary. The population of Mich- igan was fluctuating and growing quite rapidly3 during this period and it is believed that the conventional practice of employing estimated populations as of July 1 would result in an appreciable error. It is assumed, therefore, that the errors of mid-year pepulation estimates for Michigan in 1940 are greater than those introduced into the rates by employ- ing enumerated populations of a date only a few months from the mid-year.3 Pyocedgres Used'ip,Analyse§‘g§ County Data' The mortality rates presented for the counties include 1. See Appendix I for a list of all formulas used in this study. The direct method of adjusting death rates is used wherever possible, and the standard population used in.most instances is the United States, 1940. This in- formation is given in the tables where the adjusted rates are presented. The indirect method is employed for ad- justing rates by principal causes in counties. Examples of these methods are given in Appendix I. 2. See Beagle, J. Allen, Michigan Population, Michigan State College, Agricultural Experiment Station, East Lansing, IMichigan, Special Bulletin 342, p. 11 and 21. 3. See Linder, Forrest E., and Grove, Robert D.,.gp. cit., p. 37 where such an assumption is recommended under stated conditions. tza'gn't‘v‘!‘ 42 the following: (1) Total crude and age-adjusted rates (2) Crude rates by residence (3) Ten principal causes and suicide (4) live principal causes by residence (5) Infant death rates for total and rural and urban areas. These data are given both in tables and maps. The major problem.with which one is confronted in attempting to analyze death rates for counties is the small number of deaths on which some of the rates are‘based. This problem also occurs for some of the State data separated into racial groups. In order to provide a basis for evaluating death rates, it is necessary to'indicate those rates which are based on small numbers of deaths. Therefore, following a procedure employed by the United States Bureau of the Census and the National Office of Vital Statistics,1 all death rates which are based on less than 20 deaths are indicated by underscoring. Differences,and especially small ones, between two such rates should be interpreted with caution.2 14 The Vital Statistics Division of the United States Bureau of Census was transferred to the United States Public Health Service in July 1946 and is now known as the Nae tional Office of Vital Statistics. 2. According to studies made by the U.S. Bureau of Census and reported by Linder and Grove, a death rate based on 20 deaths '...has a standard error of roughly 22 percent of the magnitude of the rate. Rates based on smaller frequencies have larger standard errors, and those based on larger frequencies have smaller standard errors. The standard errors of many'cf the rates based on frequencies greater than 20 should also be calculated before a con- elusive interpretation of the differences between the rates 18 MdOe' _ (Sec Linder, Forrest E., and Grove, Robert D., 2p. p_i_1:_. Po 94c Ill 3-; .r J r I'- .4 ‘1 43 In view of the foregoing consideration another technique was employed to increase the accuracy of rates in local areas. This method consisted of finding the total number of deaths in each county for ten years, 1935-1944, and relating the average deaths to the corresponding enumerated populations of 1940. This procedure produces a much more stable death rate than one for any single year.1 Thus the county rates which are underscored indicate the number of deaths to be under 200 rather than 20. Oscoda, the small- est county in the State, a population of 2,543 in 1940, had 227 deaths during the ten years or an average of 22.7. All of the counties had more than 200 deaths and only four counties had less than 500 during the ten years. Although many of the rates for specific causes are based on less than 200 deaths, the stability introduced by the ten-year average appears to give them a relatively high degree of consis- tency. This consistency is indicated by the regular occurr- ence of counties either in high or low rate columns as expected. 93.95 91 ngntation . Part I of this study presents in this and the fore- going chapter the introductory materials pertaining to the problems, methods, procedures, and objectives of mortality 1. Michigan began allocating deaths to place of residence in 1933, and mortality data for rural and urban residents in counties are available in the Annual Re orts of the Michigan Department of Health since that date. Prior to 1933 only deaths by place of occurrence are available. 44 studies and at apprcpriate places indicates their bearing on this particular analysis. Part II portrays in four chapters the mortality-- health situation in.Michigan as of 1940, presenting data for the State as a whole covering rural-urban differentials, racial differences, and finally comparisons of Michigan with the United States. Part III, a discussion of internal variation, consists of two chapters devoted to the analysis of county data. Part IV, a section on trends, shows in one chapter some of the changes in mortality that have occurred during a thirty year period, 1910-1940, for Michigan in comparison with the United States. ' Part V. Summary and Synthesis. The pattern of presentation within each chapter is as follows: (A) Data, (B) Interpretation, (0) Summary. The tables and graphic materials are placed on the same page with the discussion when possible; otherwise they follow the page after which the reference is made. Souycec _o_f_ 93:93., ' The primary sources required for an analysis such as the present one are the official tabulations published by the State and National Bureaus of Vital Statistics. The State data has been obtained from the following two sources: Vital Statistics of the United States, 1940, Part II, BSsidonce, United—States Government Printing Office, Washington, D.C., 1943- 45 (Data in this volume covering the calendar year 1940 are compiled from.transcripts of original death certificates received from registration officials of States, cities, and territory and possessions of the United States. Data are classified by geo- graphic area, cause of death, age, sex, race, and institution. Tabulations were prepared under the immediate supervision of Dr. Robert D. Grove, Social Science Analyst, and Dr. Iwao M5 Mbriyama, Associate Biometrician. See page one.) Lindor, Forrest 3., and Grove, Robert D., Vital Statistics gateg‘iglyhg,United States 1900-1940, United States Gov- ernment Printing Office, washington, D.C., 1943. This volume contains detailed time-trend tables of mortality. (Death rates are based on official tabulations of deaths and include specific rates for cause of death, age, sex, race,and geographic area. No adjusted rates are included. The age- specific rates in this volume were employed in ad- justing rates for the dissertation.) The county data presented in Chapters VII and VIII are from the following source: Annual RoEort of the Michi an Department of Health, Num- bars as t rough-7T, chigan Department of‘HTE‘e t , Lansing, Michigan , 19 36-1945. The bibliography at the end of the dissertation lists the complete reference to all publications cited. Citations also occur on the page there the reference is made. Table pad. Iigge Numberipg System Tables and Figures are numbered consecutively within each Chapter preceded by the chapter number as follows: Chapter VII _ Table 7.1 Figure 7.1 Table 7.2 11811330 7.2 Chapter VIII Table 8.1 Figure 8.1 Table 8.2 Figure 8.2 The first table and the first figure in Chapter VII, for example, are referred to as “Table 7.1“ and "Figure 7.1." (I 1251 I {P ‘9‘ 1 Mn ”9. ad 46 A text reference to the second table listed in Chapter VIII is referred to as ”Table 8.2". The Appgndices The appendices are given Roman numerals and tables in them are numbered consecutively by Arabic numerals within each Appendix. Several large tables are placed in Appendix II. PART II PRESENT HEALTH SITUATION IN MICHIGAN . 0mm III nosrmrr AND HEALTH IN MICHIGAN: A GENERAL VIEW Crude Begident and Agg-Adjgstod Death Rates In 1940 the crude resident death rate1 for all causes in Michigan was 9.9 per 1,000 of the total popu- lation. This includes deaths from.unknown causes. This rate comprises all residents of Michigan whether they died in or out of Michigan.1 When.the period 1935-1944 is taken as representative of the present situation, the crude resident death rate1 for Michigan is slightly higher, namely 10.1 per 1,000 pepulation. This ten-year average, perhaps, more accurately measures the force of mortality8 on.the Michigan population. When the deaths on which the 1940 rate is based are adjusted to the age distribution of the population of the United States in 1940, the age-adjusted death rate3 is approximately the same as the ten-year average rate, namely, 10.2 per 1,000 1. 2. 3. For a statement of the method of computation, meaning as a.measure of health, advantages and disadvantages of the crude resident death rate, see pp.309ff. A11 death rates employed in this study are based on deaths allocated to the place of residence of the person unless otherwise indicated. Deaths were allocated to place of residence in M1chigan.for the first time in 1935, and this practice had been continued in all subsequent Vital Statistics Reports. Mortality data prior to 1935 are not available for computing the crude resident death rate, and it is necessary to employ deaths by place of occurrence. For a discussion of the force of mortality, see p. 13. See Appendix I for'methcd of computation, meaning, and uses of the adjusted death rate. Here the 1940 deaths were adjusted by both the direct and indirect methods and essentially the same result was obtained using the pepu- lation of the United States in 1940 as a standard. 48 population. The crude resident death rate for the year 1945 was 9.91per 1,000 population, or the same as for 1940; how- ever, this does not mean necessarily that no changes occurr- ed in any of the varying mortality risk groups composing the respective populations. Although Michigan's population fluctuated a great deal from.1930-1940 and since 1940 during the war years, it is likely that most of this change has been between the rural and urban areas within the State.2 However, there has been a slow, steady flow of negro and white mi- grants from the Southern States to the industrial centers of Nuchigan, which probably has influenced the mortality patt- erns more than.the rapid population movements in Michigan occurring during the war. In view of the close similarity among the four rates presented, it is reasonable to assume that the relative standing of classes (specific mortality risk groups) in 1940 remained approximately the same in 1945. (The 1940 crude rate is 9.9, Which is identical with the 1945 crude death rate; and the ten-year average crude death rate is 10.1 which is almost the same as the age-adjusted death rate for 1940, namely, 10.2.) Since 1940 is a.more satis- factory year for a detailed analysis of data inasmuch as 1. "Deaths and Death Rates for Selected Causes United States, Each Division and State, 1945," Vital Statistics-~Special Re ort , 1947, Volume 27, Number 3, Tablez, p. 41. 2. See Commission on Hospital Care, "State Hospital Study manual," Book IV, (maps and tables) pp. 10-12, (pre- liminary draft). l1 ‘ 0'" :0. ma“ ‘I fist-a a! in" Ii . a I v unii 0! "D.‘ u . . iv, ‘ I O :u‘a,” I u :.---E "In“ “ "I‘a. I 49 pcpulation estimates for some of the classes are not avail- able for other years, the year 1940 will be used. fig-flecific Death gates There are greater variations among the various age- group mortality rates than among any other classes of the pepulation for the obvious reason that the risk of death is very much greater at certain ages than at others (see Table 3.1). The age-specific death ratesl for a given pop- ulation have a characteristic form. It is a picture of high death rates for those under age 1 year, the rates dropping to the lowest point at ages 5-14, and then rising steadily until the end of the life span. Table 5.1 AGE-SPECIFIC DEATH RATES, MICHIGAN, 1940 M Age Group Rate p_e_r 1,000 Population All Agssl 9.9 Under 1 48o]. 1 " ‘ 2.2 5 " 14 1.0 15 " 24 1e7 85 *- 34 2.5 35 " M 4e6 55 -" 64 21.2 65 " 74 48.2 '75 - 84 113.6 85 Over s42; Includes deaths of unknown ages. Source: Computed from Vital Statistics of _t_h_g United tat s, I910, Vqume II, Fosidence, Tab e , p. 190. 1. For a statement of the method of computation, advantages, and disadvantages of the age-specific death rates, see Appendix Io 50 In Table 3.1 are shown the age-specific death rates for Michigan and the rate for all ages. At least three similarities can be pointed out. (1) The age-specific death rate that most nearly approximates the crude resident death rate for the State is 9.6 for age group 45-54. This is interesting since it indicates that an individual living in Michigan is subject to approximately the same average chance of death as are the people living there who are of age. 45-54. (2) The age-specific rate for persons under 1 year of age is approximately the same as it is for the old aged group, 65-74, the rates being 48.1 and 48.2 respec- tively,--though persons in these two classes die from.quite different causes. (3) The chance of death occurring to people aged 1-4 is 2.2 per 1,000 population, which is quite close to the probability of death for the population aged 25-34, namely, 2.5. The greatest difference between age-specific death rates occurs between age group 5-14 and those 85 years and over. The lowest rate is that of age group 5-14, 1.0 per 1,000 population; and the highest rate, 245.1, that of age group 85 years and over. .All of the age groups 55 years of age and up have death rates ranging from 21.2 to 245.1, which are two or more times the rates of all other age groups except those under 1 year of age. The low death rate age groups are 1-4, 5-14, 15-24, 25-34, 35-44, and 45-54. Thus, the age groups from.which a toll of more than 10 deaths per 1,000 population in each age group is taken 51 Table 302 ! NUMBER or DEATHS PER 10,000 OF TOTAL DEATHS or KNOWN AGE, MICHIGAN, BY AGE GROUPS, 1940 Deaths of Known Age Age Number Per 10,000 Deaths Under 1 4,032 773 l - 4 760 146 5 - 14 860 165 15 - 24 1,631 313 25 - 34 2,105 404 35 - 44 3,505 672 45 - 54 6,289 1,206 55 - 64 8,677 1,664 65 - 74 11,084 2,125 75 - 84 9,974 1,913 85 Over 3 230 619 Total 52,147 10,000 Unknown age 36 Source: Computed fromVital Statistics of the United States, 1940, Vol. II,—:Residence, Table—5,—p. I'T‘g . ""_"' Table 3 e 3 NUMBER OF PEOPLE PER 10,000 TOTAL POPULATION, MICHIGAN, BY AGE, 1940 (Enumerated as of Apr. 1) Total Population of Known Age' Total Age Population Per 10,000 Population Under 1 83,807 159 1 - 4 347,577 661 5 - 14 882,772 1,680 15 - 24 935,371 1,780 25 - 34 848,848 1,615 35 - 44 763,633 1,453 45 - 54 654,347 1,245 55 - 64 408,897 778 65 - 74 229,842 437 75 - 84 87,831 1:; 85 Over 13 181 Total '5", 2‘5" 6","10'6 $031560 4 Source: Computed from Linder, Forrest E. and Grove, Robert D., Vital Statistics Rates in the United States, 1900-1940, U. 8. Government Printing Office, Washington, D. 0., 1943, Table II, p. 898. y‘- |h ‘\ are those under 1 year of age and all age groups above 55 years of age (see Table 3.1). In Tables 3.2 and 3.3 the proportions of deaths and of populations in each age group are given. Age and Sex Differences Sex mortality differentials by age groups are given in Table 3040 Table 304 AGE-SEX SPECIFIC DEATH RATES, MICHIGAN, 1940 Rates er 1 000 P0pul ation Age Groups fiEIe Female Excess Male All 13.31 11.0 8.9 2.1 Uthr 1 5501 4007 1404 1 ' 1 203 200 03 5 - 14 101 08 03 15 - 84 201 105 06 25 ' 34 207 202 05 35 - 44 500 401 09 45 - 54 11.0 7.9 3.1 55 ' 64 3403 1709 603 65 - 74 53.5 42.8 10.7 75 - 84 122.2 105.2 17.0 85 07.3 24900 24107 703 r 1 Includes deaths of unknown ages. Source: Computed from.Vital Statistics of the United States, 1940, Volume II, Residence, Table 9, p. These data.mey be summarized as follows: (1) Death rates for.ma1es are higher than rates for females in all age groups, although some of the differences are small. (2) The excess of death rates for males over death rates for females vary from..3 per 1,000 population for those aged 1-14 years to 17.0, the greatest excesses being for those under 1 year of 53 age and those over 45. In general, the excess of rates for nmles over the rates for females steadily increases from age group 5-14 through age group 75-84. (3) Certain pairs of age groups have approximately equal death rates for. males, and also approximately equal rates for females. They are summarized below: Pairs of Male Female Age Groups Rate Ratefl (1) A11 A898 1100 809 45-54 Years 11.0 7.9 Difference .0 I.0 (2) 1 - 4 Years 2.3 2.0 15-24 Years 2.1 1.5 Difference .2 .5 (3) Under 1 Year 55.1 40.7 65-74 Years 53.5 42.8 Difference I06 2.1 appal-Urban Mortality Differentials The crude and age-adjusted death rates for rural and urban areas are presented in the table below: Table 5.5 PERCENTAGE DISTRIBUTION OF POPULATION, CRUDE RESIDENT AND AGE-ADJUSTED DEATH RATES, RURAL AND URBAN AREAS OF MICHIGAN, 1940 Crude Death Age-Adjustedl Residence Percent Rate Death Rate Area Pepulation Per 1,000 Pop. ngpl,000 Pop; State 10000~ 909 1002 Urban 65.7 9.4 10.4 Rural 34.3 11.0 10.2 IAdjusted by direct method using the population of the United States for 1940. Source: Vital Statistipglg§.phg United States, 1940, Volume II, Residence, Table 9, p. I9 . 54 Crude resident death rates reveal that each year approxi- mately 11.0 persons per 1,000 population living in rural areas die from.all causes in comparison with 9.4 in the cities and towns. These two death rates indicate that urban mortality data tend to determine the death rate for the State. 'The variation of the rural rate from that of the State is 1.1 while the urban rate varies from.the State rate only .5. This is to be expected since 65.7 percent of the total population of the State is urban and only 34.3 percent is rural. Because the urban pepulation is young in comparison with the rural pepulation, urban deaths re- duce the crude death rate for the State. This rate is 9.9 per 1,000 population. When.the rural crude_rate is adjust- ed to the age distribution of the United States, it is lowered from.1l.0 to 10.2; and when.the urban rate is ad- Justed in.the same manner, it is raised from.9.4 to 10.4. m Mortality pupal-angel]- Although Michigan has relatively few colored people and many of their death rates are based on small numbers of deaths, it is considered desirab1e_to present some of the mortality differentials by race. The colored population of Michigan has increased rapidly during the past three decades,3 and especially since the war. In all probability, 1. See Chapter‘v for a more detailed treatment of racial mortality differentials. 2. See Table 5.1 in Chapter V. 55 it will continue to increase for the next few years. In 1940 there were 216,463 colored people in Nflchigan of whom 96.2 percent were negro. The remaining 3.8 percent were primarily Indians. These groups combined had a crude death rate for that year of 12.6 per 1,000 pepulation as compared with 9.8 for the white race (see Table 3.6). Thus, colored persons in.Nuchigen die at a rate almost 3 per 1,000 population greater than that at which white peeple die, the excess of the rate for the colored group over the white being 29 percent. Table 306 CRUDE RESIDENT AND AGE-ADJUSTED DEATH RATES FOR THE WHITE RACE AND ALL OTHER RACES, MICHIGAN, 1940 Rates Pep:1,000 Population Crude Death Rate Age-Adjusted DeatH Racg Rate1 White Race 9.8 10.0 All Other Races 12.6 15.8 Difference 2.5 5.3 Illdjusted to the age distribution of the population of‘ the United States, 1940, by the direct method. Source: Crude data from.Linder, Forrest E. and Grove, Robert D., Vital Statistics Rates in the United States l90§~I940, United States Government PrInt- T‘Tng ffice, Washington D. c., 1943, Table 10, p. 193. In spite of the fact that the colored group has a younger age distribution than the white pepulation in.Michi- gan, the crude death rate of the non-white segment is 29 percent higher than that of the white. Furthermore, when these two death rates are adjusted to the age distribution of the population of the United States for 1940, the 56 differential becomes even greater. The actual difference between the age-adjusted death rates is 5.8, which means that if the two racial groups had the same age distribution the colored people would die at a rate 58 percent higher than that of the white group. Thus, the excess of the death rate of the colored population over that of the white pOpulation cannot be explained in terms of their age dis- tribution because they do not have larger preportions of their number among the upper ages (45 years and over) where the chronic diseases take such a large toll of human lives. It is not the object of this chapter to explain such dif- ferences but rather to show that they exist as a character- istic of Michigan's population. Several explanations for the high mortality rate among the colored races are offered in Chapter V. Twenty-five Selected Causes 93 26332.41 To know that 52,183 persons died in Michigan in 1940 is of little value unless such information is supplemented by an adequate analysis of the causes of death. It is the constantly shifting emphasis among the causes of death that introdudes much of the uncertainty into the planning of health programs. .An inspection of Table 3.7 indicates that in 1900 the two chief causes of death were communicable diseases--namely, pneumonia (including influenza); and 1. For a treatment of various types of errors in reporting causes of death and their influences on the cause-specific rates, see pp. 29-34. 57 diarnhea (including enteritis, etc.). By 1910 heart dis- eases occupied first place but pneumonia again took the lead in 1920 following the influenza epidemic. By 1940 both of these canmunicable diseases had been displaced by two degenerative ones-~namely, heart diseases and cancer. Table 3.7 CAUSE-SPECIFIC DEATH RATES,1 BY Pugs or OCCURRENCE, MICHIGAN, 1900-1940 Year Cause 'T‘coo pffio 1920 I950 T540— Pneumonia including 12607 10907 23209 8204 5208 influenza Diarrhea including 124.1 103.0 58.1 20.6 4.6 enteritis, etc. Heart Diseases 119.3 167.0 176.2 260.1 293.6 Cancer 61.2 74.6 83.2 92.5 119.4 Y Source: Linder, Forrest E. and Grove, Rebert D., Vital Statistics Rates in the United States, 1900-1940, 9. 8. Government PPintIng Office, Washington D. 0., 1943, Table 20, Ppe 344-3470 Heart diseases, the Chief cause of death in 1940, proved fatal to 15,602 persons, or about 297 of every 100,000 of the State's population. More than.two and one- half times as_many people died from heart diseases as from any other cause. The second most fatal disease in.Michigan in 1940 was cancer, which accounted for 12 percent of all deaths l. The formula for computing cause-specific rates is given on P0 3380 2. Trends and changes in.mortality rates are presented in detail in Part IV of this study, Chapter IX. 58 Table 308 IMORTALITY RATES FROM THE COMMON CAUSES OF DEATH IN MICHIGAN, 1940, AND THE RELATIVE RANK 0]? EACH causal Rate Per 100,000 Rank Cause of Death POpulation All causes 992.8 1 Diseases of the heart 296.8 2 Cancer and other malignant tumors 119.4 3 Intracranial lesions of vascular origin 89.5 4 Nephritis 56.5 5 Pneumonia and Influenza 52.8 6 .Accidents other than motor vehicle 41.7 7 Tuberculosis 34.3 8 motor-vehicle accidents 33.1 9 Diabetes 26.6 10 Premature births 24.0 11 Suicide 13.9 12 Congenital malformation 13.8 13 Syphilis 11.5 14 Appendicitis 9.5 15 Hernia and intestinal obstruction 9.3 16 Cirrhosis of the liver 9.2 17 'Ulcer of the stomach 6.1 18 Diseases of the prostate 5.7 19 Puerperal causes 5.5 20 Exophthalmic goiter 3.6 21 Diarnhea, entegitis, ulceration of intestines 3.3 22 Bronchitis 3.2 23 Homicide 3.0 24 Alcoholism. 1.1 25 Whooping cough . 1.1 26 Scarlet fever .8 27 Diphtheria .4 28 Measles .3 29 Cerebrospinal meningitis .3 Ill defined causes 4.1 .All other causes 112.3 Source: Vital Statistics of,£hg United‘épgtgg, 1940, Part II, RSSidence, TabIe‘IO, pp. 226-7. 1. Based on population enumerated as of April, 1940. 2. Includes only deaths from this cause under 2 years of age. Note: Deaths of residents occurring outside of Michigan are included in this table. 3.0 ’1‘ I "a I l m.“ . . a .1 .9, I Ii. -A u, 0‘ I 'ln' A I uhsu hi‘ L... '9‘. "Ii. HP. 5: :a' 0 '0‘ t": .‘ 59 for that year, or about 120 people out of 100,000 of the population. Ranking second among the ten principal causes of death, cancer took a toll of 6,275 lives, which is 30 per 100,000 population more than its next nearest competitor, intraoranial lesions of vascularorigin. Intracranial lesions of vascular origin occupied third place among the ten principal causes of death and was responsible for 90 deaths per 100,000 of the population. These disorders caused 4,702 fatalities or 9 percent of the total number of deaths in 1940. Nephritis, another degenerative disease, ranked fourth. This disease killed 2,971 people, or 57 out of Ovary 100,000 of the State's population in 1940. Approxi- 1llctely 6 percent of all deaths for that year were due to naphritis. The four most common causes of death in 1940 (heart disease, cancer, intracranial lesions of vascular origin, and nephritis) were degenerative in nature. These four Muses were responsible for 56.7 percent of the total deaths in Michigan. Only one other degenerative disease, diabetes mellitus, was among the ten principal causes. The five de- generative disorders listed among the ten principal causes of death accounted for over half, or 59.4 percent, of all the deaths in the State in 1940. Pneumonia and tuberculosis, two communicable diseases, ranked fifth and seventh respectively. Pneumonia proved fatal to 2,777 persons, with a rate of 52.8 per 100,000 popu- 60 lation, while tuberculosis took the lives of 1,803 or 34 per 1L00,000 peeple. Together these diseases were responsible for approximately 9 percent of all deaths in Michigan in 1940. Accidental deaths other than.motor vehicle accidents ranked sixth and resulted in the death of 450 individuals. This is 26 percent more fatalities than occurred from motor vehicle accidents. The rates were 33 and 42 per 100,000 population respectively. When the two types are combined, accidents from all causes ranked fourth in the list of principal causes of death in Michigan, killing 3,930 people or 75 out of every 100,000 of the State's population. Slightly over 7 percent of all deaths in 1940 were caused by accidents. Ninth and tenth in rank on the list of common causes were diabetes mellitus and premature births. About 27 Michigan people per 100,000 population died from diabetes and 24 from.premature births. Slightly over 5 percent of all deaths were due to these two causes. The foregoing ten leading causes of death in Michigan comprised 78 percent of the total number for the year. Each of them took an annual toll of 20 or more people out of every 100,000 of the State's population. In addition, there were three principal causes in which each killed more than.10 per 100,000 people in Nuchigan. They were as follows: suicide (13.9), congenital malformation (13.8), and syphilis (11.5). Each of these re “-01 4 1" ‘ .ypdl A“ I t... "Isl . s .I U c \ ‘0. I 61 proved fatal to more than 600 people in 1940, and together accounted for almost 2,000 deaths that year. In addition to the ten principal causes, there were some other diseases with rates so high as to merit the con- sideration of everyone concerned with health conditions in the State. Among these are appendicitis (9.5), cirrhosis of the liver (9.2), hernia (9.3), ulcer of the stomach (6.1), and diseases of the prostate (5.7). Each took an annual toll of from 300 to 500 lives, and combined, about 2,000 lives. One other classified cause, puerperal causes, had an annual death rate of more than 5 per,100,000 popula- tion. There were 287 deaths in 1940. Goiter and.homicide were responsible for the death of from 100 to 200 people in 1940, or from 3 to 4 out of each 100,000 of the State's population. The rates for the remaining diseases in Table 3.8 indicate that few people died from infectious diseases in 1940 though they were common causes of death not many years ago. 'Referring to Table 3.8, we find that degenerative diseases caused more deaths in Michigan in 1940 than com- municable diseases. Among the ten principal causes of death for the year, the five degenerative diseases caused 59.4 percent of all the fatalities. At the bottom of the list fell many of the infectious diseases--such as, whoop- ing cough, measles, and diphtheria. As will be shown later, Michigan.has performed her task of combating these com- municable diseases better than the rest of the United States. 62 But she has a growing problem in the control of such dis- eases as heart disease, cancer, and intracranial lesions of vascular origin. Infant ;a_n__d_ Maternal Mortality 321221 The rapid decline of the infant death rate during the last few decades perhaps has signaled the end of that his- torical phase of our public health programs which might be called the "Communicable Disease Period," and has initiated the new phase, the "Chronic Disease Period." Although the infant rate is not as sensitive an index of public health progress as has been claimed,2 the rate is the best known and most widely accepted single measure of the success of the public health movement in the control of communicable 1. See Appendix I, p. 339, for formulas of these death rates. 2. Pearl studied the decrease of infant death rates in 21 countries during a period of approximately 17 years and found that seven countries had larger decreases than the United States. The average annual decrease of the infant death rate in the United States was 2.19, while those countries with larger decreases than.the United States were: England and Wales 2.26, Switzerland 2056, Japan 2087, Finland 3001, 08371011 3005, Italy 3089, and Germany 5.19. Pearl concludes: "Evidently this relatively huge, rapid, and consistent drop cannot have been wholly due to public health efforts directed to this end. One reason for believing that other causal factors were involved is that the health services of the 20 countries in the table differ widely in the ex- tent and effectiveness of their efforts. Another is the position of particular countries in the table. Thus the United States stands 8th in.the table. It would take a lot of proving to demonstrate that each and all of the seven countries above it have health services more active and efficient in the infant mor- tality field.” (See Pearl, Raymond, Introductigp 29 Medical Biometr ‘gpg Statistics, W. B. Saunders Company, Miranda"; Pennsylvania, 1941, pp. 191-192.) {‘I «\ .s's' , In. , .. a we.» . I-w: A“; d O. .. v-.‘. III a." In 63 diseases.l Although communicable diseases are no longer the predominant health problems as measured by mortality data, their prevention and control will remain, without question, an important public health problem for Which the infant death rate will continue to be a useful index. The infant death rate in Michigan was 40.7 in 1940 and is an acceptable index of the present situation. It is identical with the average rate computed for the years 1935-1944. A four-year average, 1938-1941, was slightly higher, 41.4, a difference of less than 1 from the other rates. In 1940 there were 4,032 deaths under 1 year of age. The causes from Which these deaths occurred and their relative rank are given in Table 3.9. Among the ten leading causes of infant deaths is premature births, which is also among the leading causes when related to the total population. Premature birth ranks first as a cause of infant deaths. The number of pre- mature births per 100,000 live births for the period, 1938- 1941, was 1,352.1. This is twice as high as_the rate for congenital malformation (570.0) which ranks second. Infant death rates from.pneumonia, injury at birth, other diseases peculiar to the first year, and diarrhea, were all higher than 225. Two other causes listed among the ten principal ones, mechanical suffocation and diseases of the thymus gland, took between 90 and 100 lives each, per 100,000 live 1. See Introduction, pp. l3ff, for Opinions of various aumhcrities relative to the use of mortality rates as measures of health. The infant death rate is included by most of the authorities. Table 3.9 INFANT MORTALITY RATES BY SELECTED CAUSES, MICHIGAN, 19323-1941 64 Number of deaths under 1 year of age per 100,000 Selected Causes live births (3-yr. ave.) All Causes 4,137.3 Premature birth 1,352.1 Congenital malformations 570.0 Pneumonia (all forms) 557.9 Injury at birth 401.8 Other diseases peculiar to first year 259.7 Diarrhea, enteritis, etc. 229.5 Mechanical suffocation & other accidents 98.1 Diseases of thymus gland 96.0 Congenital debility 72.7 Whooping cough 56.8 Other diseases of respiratory system. (55.1 Influenza 48.3 Diseases of nervous system. 39.7 Intestinal obstruction 35.4 Diseases of ear and mastoid process 22.9 Syphilis 1904 Diseases of circulatory system. 18.6 Ill-defined 2 Unknown causes 18.1 Dysentery 17.3 Diseases of genito-urinary system. 14.3 Measles 12.8 All other causes 140.8 Source: Commission on Hospital Care, State Hospital anual, Book IV, (MimeographedT, 1946, TabIe 3, Study p. 28. Data computed from.Vita1 Statistics of the - United States, 1938, Part I‘T‘B‘, a le 24; 19—39,"" Part II, TabIe 13; 1940, Part II, Table 14; 1941, Part II, Table 19. births. The last two among the ten leading causes are con- genital debility (72.7) and whooping cough (56.8). All of the other causes listed had infant death rates of 55.1 or less per 100,000 live births. 65 The maternal mortality rate has declined to such a low point that one wonders if maternal deaths will not practically disappear in the future. It is another example of the success of health education, public health programs, and medical care and treatment by physicians. In 1940 only 2.9 mothers per 1,000 live births died in Michigan, and this figure had dropped to 2.1 by 1942.1 The decreases in the maternal mortality rate over a period of years are presented in Chapter Ix. l. Linder, Forrest E. and Grove, Robert D., gp. cit., Table 36, P0 6210 CHAPTER IV RURLLPURBAN'MDRTALITY DIFFERENTIALS .Hom does the mortalityhhealth situation of Michigan's people living in cities and towns compare with that of those living in the country? To the degree that the health of a population is measured by the rate at Which people fail to overcome sickness, this question may be examined by compar- ing: (1) crude resident and age-adjusted death rates for rural and urban areas, (2) age and sex mortality differentials for urban.and rural groups, (3) selected cause-specific death rates, (4) race-specific rates, and (5) infant and maternal mortality'rates. 95%; Resident 32g, Age-Adjusted 29.9.3}; 33351 Each year approximately 11 persons per 1,000 people living in the rural areas die in comparison with 9.4 in the cities and towns. (See Table 4.1.) .a partial explanation of this difference was indicated in the previous chapter ‘ where it was pointed out that a large proportion of the rural pOpulaticn falls into the older age groups which.have greater risks of death. The crude resident rate for rural areas is also higher than the rate for all of the various pepulation size cities with the exception of those cities of 2,500 to 10,000 population. 1. See Appendix I for method of computing crude resident and age-adjusted death rates. 66 67 When.a.more refined death rate is computed for the rural and urban areas, by which differences in the age dis- tribution of the two areas are eliminated, the total rural death rate is lower than the total urban rate.1 Further, the rural death rate is lower than the rate for various populationvsize urban places. (Table 4.1) Therefore, when an age-adjusted death rate is employed to measure health, 1. This proposition has been demonstrated both in Europe and the United States. The writer could find no study or reference which claimed to have proven the reverse proposition that the rural death rate is higher than the urban. Willcox, Walter,.gp..git., p. 213, gives data showing the States in the New England division of the United States had a crude death rate in 1930 Which was higher for the rural than.the urban population (11.7 compared with 12.0), but these rates adjusted for age would without doubt show the reverse for this area. Landis writes, "Rural communities have long had a lower death rate than urban communities." Thompson presents data showing that the rural districts of England and Wales have lower death rates than urban, and he writes “... urban life in the United States and in England and Wales is less favorable to a low death rate than rural life, in spite of the fact that a disproportionately large part af all expenditures for health, and particularly for public health, takes place in the cities." (See refer- ences for Lendis and Thompson below.) For some of the recent studies Which show the death rates of the urban to be higher than the rural area see: Warren S. Thompson and P. K. Whelpton, P0p_u1ation Trends _i__n th__e_ United States, MbGraw Hill ook Company, Inc., New York, I933, p. 241; Warren S. Thompson, Pepulation Problems, moGraw- Hill Book Company, Inc., New York, 1942, p. 229; Paul H. Landis, Po ulation Problems, American Book Company, New Yerk, 1943, p. 191; Bureau of the Census, "Age-Adjusted Death Rates in the United States, 1900-1940,“ Vital Sta- tistics---S ecial Re orts, Vol. xxm, No. I, T45, T551. , p. 21; . Lambert Molyneaux, “Differential Mortality in Texas,” American Sociological Review, Vol. I, 1945, p. 17-25; J. Lambert Molyneaux, Sara K. Gilliam, and L. C. Plorant, "Differences in Virginia Death Rates by Color, Sex, Age and Rural or Urban Residence," émggigggugggiglgr 1ca1 Review, Vol. 12, No. 5, October 1947, p. 534; fiEuise Kemp and T. Lynn Smith, Health Egghmortality in Louisiana, Louisiana Experiment Station, Bulletin 3907 Baton Rouge, 1945, p. 14. 68 small cities tend to have the lowest level of health,1 While rural areas occupy the most favorable level. Table 4.1 DEATHS PER 1,000 POPULATION BY RESIDENCE, MICHIGAN, 1940 (by place of residence) Deathspper 1,000 population Residence Crude Death Age-Adjusted Rate (1) Death.Rate (2) Total 9.9 8.93 Urban Total 9.4 9.15 Cities of 100,000 and over 8e5 8e98 Cities of 10,000 to 100,000 10.1 9.21 Cities of 2,500 to 10,000 11.7 9.39 Rural Total 11.0 8.59 1. Linder, Forrest 3., and Grove, Robert D., Vital Statis- tics Rates‘ig the United States 1900-1940, 1943, Table II, p. 202; Ursa-E and rural total computed from Vital Statistics of the United States, 1940, Vol. II, Residence, Table 9, p.190.— 2. .Adjusted to the age distribution of the U. S. pOpulation in 1940 by the direct method. One might naturally expect patterns in death rates for the small American city and for rural areas to be more or less similar since such cities usually occupy the role 1. Small cities also have the highest crude and age-adjusted death rates for the United States as Shown in Chapter VI, Table 6.9. In.Virginia, it is true for the White but not for the colored population, according to Molyneaux, et‘al, _qp. 31.3., pp. 533-34. Molyneaux and colleagues explain the high rates in small cities on the basis of low densities of pepulation and lack of financial ability to purchase expensive sanitary devices and-medical services. Molyneaux found the small cities of Texas also to have the highest death rates of various size cities. (Mblyneaux, J. Lambert "Differential Mortality in Texas," gp.lgi§., pp. 17-25.) According to Kemp, Louise, and Smith, T. Lynn, small cities (2,500 to 10,000) also have the high- est crude death rates in Louisiana, but no age-adjusted rates were given, gp.‘gip.,pp. 14-15. 69 of service and trade center for the surrounding rural terri- tory. They are frequently the place where many rural-farm people retire after their active farming days are over. Mblyneaux offered the explanation that small cities have a low density of population and are financially unable to purchase expensive sanitary equipment such as certain types of modern sewage diSposal, water supply, and medical ser- vice.1 The concentration of the aged in the small cities, however, seems to explain the high crude rates satisfactorily. For example, compare the crude rates of the small with the large size cities. Their differential is 3.2, compared with the differential of the corresponding age-adjusted rates of only .41. The above factors undoubtedly account for most of the excess of death rates in small cities compared with larger ones. Because of the apparent relationship between small American cities and the rural areas, cities of 2,500 to 10,000 are frequently referred to as "rural cities"2 in publications of the National Office of Vital Statistics. Prior to 1930 the Vital Statistics Reports of the Bureau of Census included all places of less than 10,000 pOpulation l. Mblyneaux, J. Lambert, gfi‘g;»"Differences in Virginia Death Rates by Color, Sex, Age, and Rural or Urban Residence:'gp. cit., p. 535. 2. See volumes of Vital Statistics 2; £22 Uhited States, prior to 1930, published by the Bureau of Census. See 'Vital Statistics.g§ Egg United States, 1940, Part II, Residence, p. 5, for special reference to "rural cities." 70 as "rural.'1 In 1930 for the first time, such "rural cities," that is, those with a population of 2,500 to 10,000, were reported separately from.the rural population. For purposes of this study, "rural" refers only to all places and population outside of incorporated cities of less than 2,500 papulation, according to the last census. Agg- 9311 pp; Mortality Differentials The age-and sexvspecific mortality rates are pre- sented in Table 4.2. This Table allows one to make at least four important comparisons of death rates: (1) the urban and rural pOpulation for specific age groups, (2) the males in urban with males in rural areas by age groups, (3) the females in urban with females in rural areas by age groups, (4) males with females in.urban and rural areas by age groups. The facts obtained by these comparisons are summarized as follows: (1) In rural areas the death rates are lower than those of urban areas in six of the eleven age groups, While 1. From.1902 when the permanent Bureau of the Census was created until 1930, records of deaths were collected for ”cities of 10,000 pepulation or over" and "rural," which meant the remainder of the county. "Rural” for purposes of tabulating vital statistics referred to all places of less than 10,000 population. In 1930 incorporated places of 2,500 to 10,000 were reported separately, and the remainder of the class "under 10,000 pOpulation' was reported as "rural," Which in- cluded deaths occurring in all places outside of in- corporated places of 2,500 to 10,000 population. These classifications of deaths by size of city and rural were 'de facto' in Michigan until 1935 when the reports were made for the first time on a "de Jure' basis. ll- ‘ 1. I, Ifl‘ I a uh I -,.(I EI‘I.«".!‘.I.N.E'. I ‘HL‘I‘ quufluv‘ ‘THIU‘ .§.w.‘w ~ s \- h..<.r. A Q ] ‘.~§ it . 71 .pmw.hm..o sense .ooeoesmom .HH assoc» .oemfl .mopspm eapsep we» no mospmspspm Hepfi>.sonu eopseeoo ”cannon .mowo seesaw: no mousse nocsaosH H m.snm m.oem «.mnm a.mem n.emm o.nmm a.aem o.oem a.mem nope mm «.eoa H.ema H.»HH . a.moa n.0ma H.¢HH a.moa «.mua o.naa «m u as n.ee n.0m a.me H.Ne n.0m m.oe a.me m.nn a.me «a . no a.ma e.nm a.mm e.oH a.mm m.ma o.ea ~.«m m.em so . mm e.m m.HH H.0H 9.0 o.oH s.m m.e o.aa o.o «a u we «.¢ H.n m.e a.» e.¢ .m.e H.« o.m o.e as u on a.» m.« m.» o.a . a.» a.» a.» e.~ m.m en . mm m.a o.m e.a n.a a.» m.H n.H a.» e.a em . as o. .H.H o.a m. H.a o.” m. H.H o.a ea - n a.» 9.“ a.» o.m «.m m.~ o.» a.» a.” e - a m.ae o.sm a.me a.mn o.om H.ne e.oe H.no a.me a House n.m. \~.oa e.o n.o a.ma o.ad. a.» o.aH «.9 some HHHd E g .512 HE E amiss scene spasm apnea Hence ow< u) .eesouamon no ocean hp ..nom ooo.a you one nepsm. oeaa .zaeHmUHs .Mozmonnm ens .num .me4 em amass madam oHaHoemm Now Danna g a I V n v ’ I Q s. s '. I o s e I . ¢ 5 . r \ . § 9 ' A l- : L r. s g a C c s I- s w 0 ~ “ ‘ I. g 6 1 t a t a O O I «\ O c v . g r. I t s t . a w s a I ‘ ' ' ' rv u. s q, a | I w a 5 IV! I 'n‘. ‘N ‘l ‘.O .E III! “|| 'u: an. an. ‘v... 'Iwgl I lit: \ "m w u: . . ,~ I.' I I b 1" ‘-l «a: . .1 '1 72 urban residents have lower rates in only three of the age groups, 15-24, 75-84, and 85 and over (Table 4.2). In two age groups, 1-4 and 5-14, the death rates in the two areas are identical, namely, 2.2 and 1.0 per 1,000 population respectively. Urban death rates are noticeably lower than rural rates in the age group 85 years and over. The urban rate is 238.2 and the rural rate 253.0. While urban death rates are lower in age groups 75-84 and 15-24, they are only slightly so. The urban death rates are respectively 113.1 and 1.7, While rural rates for the corresponding age groups are 11401 and 1.8. Rural Urban Age Group Death Rate Dgath Rate Difference 25-34 2.3 205 02 35’44 408 4e8 e6 45-54 8.6 10.1 1.5 55-64 19.8 22.1 2.3 Source: Table 4.2. During the ages 25-64, inclusive, as well as under the age of 1 year, urban people succumb to death at higher rates than do rural people, although some of these urban rates are only slightly higher than the rural rates. For example, the difference between the urban and rural rates V at ages 25-34 is only .2, the two rates being respectively 2.5 and 2.3. At ages 55-64, however, there is a difference between rural and urban death rates of 2.3. The differences in.the four age groups comprising those persons 25 to 64 years of age, increase steadily from.the youngest to the 73 oldest age group as follows: .2, .6, 1.5, and 2.3.1 (2) Observation of the columns rural male and urban nmle in Table 4.2 reveals that death rates of males in urban areas exceed those of males in rural areas from age 25 to 84 and also those under 1 year of age. Death rates of males under 1 year of age in the city exceed those in rural regions by approximately 13 percent. There is an actual excess of the urban rate of 6.7 (See Table 4.3). The greatest difference between male rates in the two areas occurs in the age group 85 and over where the dif- ference is 17.3. (The second highest difference is 6.7, for the age group under 1 year of age. Ranking third and fourth are respectively age groups 65-74 with a difference of 5.8 and 75-84 with a difference of 3.6. The latter two age groups are particularly significant since they comprise the 20 years above 65 when most working men are retired and when chronic diseases originating in middle life increasingly 1. This suggests to the writer that there may be selective social factors in urban life which operate disadvantage- ously in man's biological adaptation and which may char- acterize rural life but to a lesser degree. Assuming that death rates reflect the culminating effect of all these factors, an interesting hypothesis becomes apparent, namely, that these factors tend to increase progressively in intensity from age group 25-34 to the oldest age group in this series, 55-64. It is the opinion of the writer that differences among the social roles of the rural and urban environments and their influence on group mores and institutional controls are in part responsible for differences in death rates. It is outside the scepe of this dissertation to attempt a defense of this hypothesis. However, it is formulated in greater detail in.Appendix IV. I ""ev. I .- 5;: _._ “L‘ s .g‘“ re '- O 74 Table 4e3 DIFFERENCES BETWEEN DEATH RATES OF MALES.AND DIFFERENCES BETWEEN DEATH RATES OF FEMALES IN RURAL AND URBAN AREAS, MICHIGAN,BY SELECTED AGE GROUPS, 1940 Differences Differences Age Male Rates Female Rates Group Rural and Urban Areas Rural and Urban Areas Under 1 6.7 excess urban 2.8 excess urban 35 - 44 .4 excess urban .9 excess urban 45 - 54 1.5 excess urban 1.5 excess urban 55 - 64 3.1 excess urban 1.9 excess urban 65 - 74 5.8 excess urban . 1.2 excess urban 75 - 84 3.6 excess urban 2.0 excess rural 85 - Over 17.3 excess rural 11.9 excess rural Source: Table 4.2. take their toll. Urban.rates for the age group 55-64 exceed rural rates by 3.1 deaths per 1,000 pepulation. This is the last decade before old age retirement under the Social Se- curity.Act and again represents a period in life when chronic diseases begin to take their toll. In all of these three age groups, the urban male dies at the rate of 3.1 or more per 1,000 in excess of males of the same age in rural areas. It should be pointed out that in each of these age groups the urban population has a smaller proportion of its total number than does the rural pOpulation and hence the risk of death for the total urban pepulation is less. For example, the three age groups combined amount to 1,245 population for each 10,000 of the total urban population, while the corresponding figure for the rural pepulation is 1,645. (See Table 4.5.) The proportionate distribution of deaths is given in Table 4.4. Deaths at ages 55-74 are also con- ‘7.-- “ur- '0‘.- III- In... .‘,‘ .- w. ‘:M u.“ h, - _ I- 4' 0 MIA. 1 1'- o ‘ . I 4:- 1 II. 75- centrated more heavily for rural than urban, the number per 10,000 being 6,234 and 5,478 respectively. (3) In Table 4.2 the columns showing death rates of females in rural and urban areas demonstrate that rural death rates of females are lower than the corresponding urban rates from age 25 to 74 inclusive, and likewise for those under 1 year of age. The female death rate under 1 year of age in the city is only slightly higher than the corresponding rate in the country, or 41.8 as compared with 39.0. In those age groups comprising females from.1-24 years, the death rates are approximately equal in the two resident groups, While from.age 25 to 74 inclusive, the ten-year age groups of females have death rates only slightly higher in urban places than.in rural areas. The greatest difference between female rates in the two areas occurs for those 85 years and over, Where the rural rate exceeds that of the urban by 11.9 deaths per 1,000 population. The second highest differential between these areas is for those under 1 year of age, Where the urban death rate exceeds the rural by 2.8. The differences between urban and rural rates of females in all other age groups are 2.0 or less. Thus, there is not as much variation between the death rates of females at these important ages in rural and urban areas as there is between corresponding rates of males. These dif- ferences are summarized in Table 4.3. (4) From an examination of the columns in Table 4.2 showing death rates of males and females in rural areas and .OI’I . U-‘ # # 9' 15" // 76 Table 4 e4: NUMBER OF DEATHS PER 10,000 DEATHS or KNOWN AGE, RURAL.AND URBAN AREAS, MICHIGAN, BY AGE, 1940 _Deaths of Known Age 'Rural Urban Per 19,000 Deaths Age Deaths Deaths Rural Urban Under 1 1,422 2,610 720 805 1 - 4 302 458 153 141 5 - 14 328 532 166 165 15 - 24 562 1,069 285 330 25 - 34 598 1,507 303 465 35 - 44 916 2,589 464 799 45 - 54 1,764 4,525 893 1,396 55 - 64 3,109 5,568 1,575 1,718 65 - 74 4,665 6,419 2,363 1,981 75 - 84 4,534 5,440 2,296 1,679 85 0v r 1,543 1,687 782 521 Totall 19,743 32,404 (10,000 10,000 _‘L 1 Does not include deaths of unknown ages. Source: Vital Statistics of the United States, 1940, Part II, Table 4e5 NUMBER OF PEOPLE PER 10,000 POPULATION, RURAL AND URBAN' RESIDENCE, MICHIGAN, BY AGE, 1940 (Enumerated as of Apr. 1) Total Population—Bf Known Age Per 10,900 POp. Age Rural Urban *Rural Urban Under 1 31,522 52,285 175 151 l - 4 137,567 210,010 763 608 5 - 14 337,341 545,431 1,872 1,579 15 - 24 309,138 626,233 1,716 1,813 25 - 34 256,926 591,922 1,426 _ 1,713 35 - 44 220,525 543,108 1,224 1,572 45 - 54 205,582 448,765 1,141 1,299 55 - 64 157,120 251,777 872 729 65 - 74 99,697 130,145 553 377 75 - 84 39,722 48,109 220 139 85 Over 6.09g 7,082 38 20 Total 1,801,239 3,454,867 10,000 10,000 Source: Sixteenth Census g§,the United States, 1940, Population, Fourth Series, filchigan, Table 1, p. 5. 1". IN N‘ p: I‘ 0‘ -l he“: , . 1. s., . u I-sl . In. .r’ 1.34..- 77 the corresponding rates in urban areas for the age groups specified, the generalization that death rates of males are higher than rates of females is true for every age group.1 The magnitude of the differences between the death rates of males and females for each age group in the two areas :I. 8 shown in the following table: Table 4.6 DIFFERENCES BETWEEN DEATH RATES OF MALES AND FEMALES IN RURAL COMPARED WITH DIFFERENCES IN URBAN AREAS, MICHIGAN, 1940 - Age Difference of Male and Female Death Rates Group“ Rural Urban Total 2.8 excess male 1.7 excess male Under 1 11.9 excess male 15.8 excess male 1 - 4 .4 excess male .2 excess male. 15 - 14 .3 excess male .3 excess male 15 - 24 .6 excess male .5 excess male 85 - 34 .8 excess male .6 excess male 35 - 44 1.2 excess male .7 excess male 4:5 - 54 3.1 excess male 3.1 excess male 55 - 64 5.6 excess male 6.8 excess male 65 - 74 8.4 excess male 13.0 excess male '75 - 84 14.1 excess male 19.7 excess male 85 Over 8.8 excess male 3.4 excess male Source: Table 4.2. Three interesting comparisons are presented based on 1. This generalization is well established, and exceptions to it are rare Where death rates from all causes for total pepulations are given. See Willcox, Walter F., ,gp. cit., p. 222; Yerushalmy, J., "The Age-Sex Composi- tion“ of the Population Resulting from.Natality and Mortality Conditions," The Milbank Memorial Fund Quarterly, Vol. xxx, NoT‘I,""—"January"I‘—, 943, p. 37; Mo yneaux, et. al,'gp. cit., p. 531. A few exceptions do occur when death rates are given by age and race; these are discussed in Chapter V of this stumy. a .7! .I . . 1-! Heel- ..-a .l " - ."o . f .5... l D 5 ’a .. A J in I 4‘ ‘H lh‘ i \ :\ a,» O.’ (m 78 the data in Table 4.6: (a) Age groups in which differences between death rates of males and females are exactly the same in rural and urban areas, (b) age groups in which dif- re rences between male and female rates are higher in rural than in urban areas, (c) age groups in which differences be tween male and female rates are higher in urban than in 1:11 ral areas . (a) The age groups which have the same difference between death rates of males and females in rural and urban areas are as follows: (See Table 4.2 for rates) When death rate of male is 100, the Difference (Male - Female) rate of female is; _Agg ur Urban Rural Urban 5’14 e3 03 73 75 45-54 3.1 3.1 69 73 Since age group 5-14 has the lowest death, rates for both sexes in both the rural and urban segments, it may be considered the healthiest period of life as measured by mortality rates. Males have the same chance of dying in either location since the age-specific deathrates for this age group are identical. Likewise, females have, the same chance of dying in either of the segments, since they also ‘ have identical rates. However, males of this age group have a slightly higher chance of dying in either location than do females, since males have a slightly higher age- specific death rate for both urban and rural areas. Although the age group 45-54 is characterized by 1 f on YE. .‘ s- I... I . the same sex mortality differences in the two areas, one cannot conclude that the force of mortality on males as c canpared with females is identical in rural and urban . areas because the two series of death rates on which they are based are not identical. The death rates for males and females in rural areas for age group 45-54 are 10.0 and 6.9 respectively; while the corresponding rates for the urban segment are 11.5 and 8.4 respectively. The female death rate in the urban segment for this age group 1 s 73.0 percent as high as that of males, while in rural areas it is only 69.0 percent as high. In the urban pop- ulation the death rate of females age 45-54 is closer to that of males of the same age than it is in rural popula- 79 1:1 on. Thus, we might conclude that there is a tendency for the force of mortality on males and females to be more near- 137 equal in urban than in rural areas for this age group. (b) The age groups in which differences between the death rates of males and females are higher in rural than in urban areas are as follows: (See Table 4.2 for rates) When death rate of Difference male is 100, the l(Male - Female) rate of female is: £53 ral Ur an Rural Urban 1- 4 .4 .2 83 91 15-24 .6 .5 '71 75 25-34 .8 .6 70 79 35-44 1.2 .7 75 86 85 Over 8.8 3.4 97 99 The age groups comprised of persons 15 to 44 years old are significant ones because they include the greater part of 00‘ ‘ fi 0 uh L". «g. 'u o‘. It. a a II! I] I a H Psi w 5 .a i e\w Al e I .o vru Add I as I s e h III 'I e l w n . u l . 7.. a. u a ran 9 I \l I J T‘ .I ‘6 HIV 80 the working population in both areas and because death rates are normally low. In these age brackets, the differences between rates of males and females are greater in rural than in urban areas.‘ Furthermore, as is indicated by the rela- tive percentages given above, the death rates of both sexes for these age groups are more nearly equal in urban than in rural areas, and we may conclude that the relative force of nmrtality on them is more nearly identical in urban than in rural places . (c) The age groups in Which the differences between the death rates of males and females are greater in urban than in rural areas are as follows: (See Table 4.2 for rates) When death rate of Difference male is 100, the (Male-Female) . rate of female is: Age Rural Urban Rural Urban 0- 1 11.9 15.8 77 73 55-64 5.6 6.8 75 73 75-84 14.1 19.7 88 84 .As the relative percentages show, the death rates of the sexes are more nearly equal in the rural than in the urban ijpulation. The higher these relative figures, the more Inearly equal are the male and female death rates.l * 1. It is believed by the writer that these aspects of urban environment are primarily social in nature and are the combined effects of the tempo of city life and its re- sulting stresses and strains on sex roles as one passes from the younger to the older ages. 81 Mortality Differentials pl Age _ip Various Size Cities and Rural Areas The age-specific death rates for rural areas, and the small, medium, and large cities are presented in Table 4.7. Table 4.8 has been prepared to facilitate the observation of the death rates given in Table 4.7. The principal object is to show the relationship of the four population groupings to each other; namely, the rural population, and small (2500- 10,000), medium (10,000-100,000), and large sized cities (100,000 and over), using the age-specific death rates of the rural pOpulation as a basis of comparison. Accordingly, three observations are presented: (a) an analysis of the respective age-specific death rates of the four population groups to determine which one has predominantly lower rates, (b) a comparison of the age-Specific death rates of large cities with the corresponding rates of the rural population, (c) a comparison of age-specific death rates of small cities With corresponding rates of the rural population. (a) or the 33 age groups for all of the various size Cities, there are only 9 Which have death rates lower than those of the rural areas. The percentages presented in Table 4.8 exaggerate these differences and permit one to ob- serve immediately those age groups in which death rates are higher in cities than in rural areas. The rural population has death rates which are lower than corresponding rates of all the cities for age groups under one year and for age groups 25-74 years. In \\ -41-“... Table 4.7 AGE-SPECIFIC DEATH RATES IN'VARIOUS POPULATION- SIZE CITIES AND RURAL AREAS, MICHIGAN, 1940 Death Rates.p§r 1,000 Population cgpiee Rura Age IUO,555 l0,000 to 2,505 to nder Gpoup f ove£:p 109,000 10,000 2,500g_ All Ages 8.5 10.1 11.7 11.0 Under 1 45.7 55.0 54.4 45.1 1 A- 4 2.0 2.7 1.7 2.2 5 - l4 .8 1.1 1.1 1.0 15 ‘ 2‘ lo? 106 202 1.8 35 " 3‘ 2.6 204 2.8 2e3 35 " 44 4.8 4A 5e5 4.2 45 - 54 10.2 9.8 9.9 8.6 55 '- 64 22.4 21.9 21.6 .19.8 65 .. 74 49.2 49.7 48.7 46.8 75'- 84 113.3 114.3 113.8 114.8 _83 Over 203.5 249.0 245.2 244.5 Source: Linder, Forrest E., and Grove, Robert D., Vital Table ‘e 8 a 0 o P0 2. momma DEVIATION or AGE-SPECIFIC DEATH RATES IN VARIOUS POPULATION-SIZE CITIES FROM THE RURAL Damn: RATES, MICHIGAN, 1940 Percent e DEviation from rural death rates Cities Cities CItIes Rural Death Ago 100,000 10,000 to 2,500 to Rates Per 1,000 5EISEEEL_, or more 100,000 10,000 Population All Ages -23 -c {6 11.0 Under 1 { 1 {22 {21 45.1 1 - 4 - 9 {23 -23 2.2 s - 14 -2c {10 {10 1.0 15 - 2,4 - 5 ~11 {11 1.8 25 - s4 {15 { 4 {22 2.3 35 - 44 {14 { s {31 4.2 45 - s4 {19 {14 {is 3.5 55 - 54 {is {11 { 9 19.8 65 - 74 { s { o { 4 46.8 '75 - s4 - 1 -.4 - 1 114.8 95 Over -17 {.2 {.3 244.5 Note: In this table the rural death rate for each age group equals 100.0. For death rates of cities on which per- centage deviation from.rural rates are based see Table ”Plus“ indicates that the death rate is in excess of the rural rate, and 'minus" that it is less than the 4.7. rural death rate. I". fa I. 83 (b) In the large cities, those of 100,000 popula- tion and over, 5 of the 11 age groups have death rates lower than corresponding rates in the rural population. All of these age groups are among the young or the old, namely, 1-24, and 75 years and over. In two age groups, 45-54 and 55-64, these large cities have higher death rates than the rural areas, the medium, or the small size cities. (c) In the small cities, 2,500-10,000, there are mfly two age groups which have lower death rates than the corresponding rates in rural areas. These age groups are 1-4 and 75-84. There are also only two age groups among the medium sized cities Which have lower rates than those or rural areas, namely, 15-24 and 75-84. Thus as measured by the age-specific death rates, IPural areas have better health than any of the various sized cities. waentyrFive Principal Causes lg Michigan, b Residence Thus far in this Chapter it has been shown that the Inopulations of the Open country, with the exception of a few age groups, have lower death rates than the urban people. The question to be considered in this section is whether lulral people maintain their favorable mortality position over urbanites when compared as to the manner in which they die. Twenty-five of the most common causes of death in 1"ll-'all. and urban areas are ranked according to the rural 84 crude rates in Table 4.9. Of these 25 causes, 14 have higher rates in rural areas, 9 are higher in urban areas, and 2 causes are nearly the same.1 The rural and urban rates are approximately equal for diabetes mellitus and are identical for diphtheria. (See Table 4.10) The four diseases which rank highest in number of doamhs are heart diseases, cancer, intracranial lesions cu'vascular origin, and nephritis. These are the "big four" degenerative diseases in both urban and rural areas. As IMght be expected in view of the age distribution in the two areas, they take a greater proportionate toll of lives in rural than in urban areas (See Table 4.4). Of the total deaths from.all causes in rural areas, 58.5 percent were from.these four diseases, while the corresponding figure in 'urban areas was only 55.6 percent. ZHeart diseases account Zfor 338.9 and 274.9 deaths per 100,000 population respective- Ily'in rural and urban.areas. Furthermore, as indicated be- low, the rate from heart diseases is more than twice the rate _— J.. Few studies of mortality by cause for 1940 here been made except those analyzing death.rates for the United States or Regions. For comparison of mortality by cause in Michigan and the Nation see Chapter VI. Kemp and Smith found that in Louisiana the crude death rates for most of the principal causes (19 out of 25 causes) had higher rates for urban than rural areas. This was true for the chronic diseases among both white and colored people in Spite of the greater prOportion of the rural population among the older age groups. No further explanation of this was given. (Kemp, Louise and Smith, T. Lynn, gp.,g;§., p. 14; the same data reported also in.Hitt, Homer L., and Bertrand, .Alvin L., Social Aspects.g§ZHospital Planning, Louisiana Study Series No. 1, Health and Hospital Division, Office of the Governor, Baton Rouge, Louisiana, 1947, p. 9) 85 Table 4.9 MORTALITY RATES FROM COMMON EAUSES OF DEATH IN MICHIGAN, 1940, BY RESIDENCE Gauge Rates per 100,000 POpulation 0 .Rural Death Rural Urban Rammm Diseases of the heart 338.9 274.9 1.23 Cancer 127 . 6 115 . l l . 11 Intracranial lesions of vascular origin 110.6 78.4 1.41 Nephrltis 63.3 53.0 1.19 Pneumonia and influenza 60.2 49.0 1.23 Accidents (non-vehicle 49.7 37.5 1.33 Accidents (motor-vehicle 38.5 30.3 1.27 mbBrOuIOUi I 30 e 0 36 e 3 e 83 Diabetes mellitus 26.4 26.7 .99 Premature births 2002 25c]. e77 Suicide 16.0 12.8 1.25 Syphilis 15.8 9.3 1.70 Congenital malformation 14.3 13.5 1.06 Hernia 9.8 9e]. 1.08 ‘ppGMlOitl. Vol. 1008 066 Cirrhosis of the liver 7.0 10.3 .68 Ulcer of the stomach 5.8 6.2 .94 Puerperal causes 4.2 6.1 .69 Diarrhea, enteritis, etc.3 3.7 3.1 1.19 EIOphthalmic goiter 2.8 4.0 .70 Whooping 001833 1e6 08 2e00 Homicide 105 3.8 e39 Scarlet fever 149 .7 1.43 Diphtheria .4 .4 1.00 Cerebrospinal meningitis E :2 .25 L11 0811.08 1,097e0 938e4 1017 I Based on pOpulation enumerated as of April, 1940. Includes only deaths under 2 years of age. Source: Computed from Vital Statistics _o_§ the United States, 1940, Part II, Residence, TabIe IO, pp. 226-227. ’1‘ Nth 86 Table 4.10 DEATH RATES PER 100,000 POPULATION'FRCM COMMON CAUSES, BY RESIDENCE, PERCENTAGE DIFFERENCE, MICHIGAN, 19401 Death : Percentage Difference Cause Rate : Where of : Death : Urban Less RuraI Less Rural ‘Urban : Than Rural Than.Urban 3(Rural-loo) (Urban'lOO) Ifleart diseases, 338.9 274.9 18.9 ---- Cancer 127.6 115.1 9.8 ---- Intracranial lesions of vascular origin 110.6 78.4 29.1 ---- Nephritis 6305 53.0 16e3 "" Pneumonia a influenza 60.2 49.0 18.6 ---- Accidents (non-vehic1e49.7 37.5 24.5 ---- .Accidents (vehicle) 38.5 30.3 21.3 ---- Tuberculosis 30.0 36e3 "" 17e4 Diabetes mellitus 26.4 26.7 ---- 1.1 Premature births 20.2 26.1 ---- 22.6 Suicide 16.0 12.8 20.0 ---- Syphilis 15e8 903 41e1 "“ Congenital mal- formation 14 03 13 05 5 e6 "" Herniaz 908 901 7e]. ""‘" Appendicitis 7.1 10.8 ---- 34.3 Cirrhosis of Liver 7.0 10.3 ---- 32.0 ‘Ulcer of the stomach 5.8 6.2 ---- 6.5 Puerperal causes 4.2 6.1 ---- 31.1 Diarrhea .tOea 307 3e]. 16.2 "" Exophthalmic goiter 2.8 4.0 ---- 30.0 Whooping cough 1.6 08 50.0 ""’ Homicide 1.5 3.8 ---- 60.5 Scarlet fever ‘;L% .7 30.0 ---- Diphtheria . _,_4_ ---- "-- Cerebrospinal meningiti—ll _gg ---- 75.0 1 Based on population enumerated April, 1940. 2 Includes intestinal obstructions. 3 Includes only deaths under 2 years of age. Note: Rates based on less than 20 deaths are underscored. Source: Computed from‘Vital Statistics.gg the United States, 1940, Part II, Residence, Tab e , pp. 226-227. 87 of the other three degenerative diseases. The rural rate exceeds the urban rate by 23 percent. cher Diseases Other Diseases Cause of Heart Disease Heart Disease Death Rural Urban Heart diseases 100.0 100.0 Cancer 37.7 41.9 Intracranial lesions of vascular origin 32.6 28.5 Nephritis 18.7 19.3 Source: Based on rates taken from.Tab1e 4.10. The death rate from cancer in rural areas is only 11 percent in excess of the rate in urban places, 127.6 compared with 115.1. Intracranial lesions of vascular origin have a rate in rural areas almost twice the size of the urban rate, or 110.6 and 78.4 respectively. The next highest death rate in both areas is nephritis, which takes a toll of 63.3 per 100,000 in rural compared with 53.0 in urban places, an ex- cess of 19 percent for the rural pOpulace. When.the deaths from.these four degenerative diseases are combined and related to the total pepulation, the respec- tive rates for rural and urban areas are 640.4 and 521.4, giving a difference in rates of 119.0 per 100,000 population, or a rural rate 23 percent in excess of the urban rate. When these same deaths are related to the population over 45 years of age, the rural rate is 2,269.8 and the urban rate is 2,033.4 per 100,000 with a difference of 236.4, or a rural rate only 12 percent higher than the urban rate. These "age-limited" death rates are influenced less by distortions due to varying age distributions of the rural and urban I‘ 88 1 and their relative differences tend to be populations, less than those of the crude rates when expressed as ratios. This is indicated in the following table for the four de- generative diseases under discussion. (See Table 4.11) Table 4.11 AGE-LIMITED AND CRUDE DEATH RATES PER 100,000 POPULATION FOR FOUR DEGENERATIVE DISEASES, MICHIGAN, 1940, BY RESIDENCE 1 Cause Crude Age-Aimited Ratio:Rura1 of Death Rate Death Rate Urban Death Age- Rural Urban Rural Urban Crude Limited Heart disease 338.9 274.9 1201.3 1072.0 1.23 1.12 Cancer 12706 11501 45202 44809 1011 1001 Intracranial lesions of vascular origin 110.6 78.4 392.2 305.8 1.41 1.28 Nephritia 6303 5300 22403 20607 1019 1009 T0tal 64004 52104 226908 203304 1023 1012 (l) The total number of deaths in rural and urban areas from specific causes characteristic of the older population groups are related to the population over 45 years of age in each area. Deaths by age for specific causes are unavailable by rural and urban residence. Source: Computed from Vital Statistics 9; the United States, 1940, Part II, Residence, Table 15, pp. 226-227. Although both the crude and age-limited death rates for these four diseases are higher for the rural than for the urban areas, this is to be expected in view of the age dis- tribution in the two pOpulations. It will be shown later 1. See Chapter I, page 212f.The Committee on Forms and Methods of Statistical Practice of the American Public Health Association recommends that death rates for these degenerative diseases be based on broad age groups from the middle to the end of life and for children's diseases on the population under 15 years of age, in order to eliminate partially the distortions due to differences in age oprOpulations being compared. 89 that these rates are reversed in favor of the rural seg- ment when more refined rates are computed in which allow- ance is made for age differences. Specific causes of death may be classified in terms of their predominance in rural versus urban areas as fol- loss: Causes of Death Causes of Death Predominant in figgal Agredominant in‘Urban 1. Heart disease 1. Tuberculosis 2. Cancer 2. Premature births 3. Intracranial lesions of 3. Appendicitis vascular origin 4. Cirrhosis of the liver 4. Nephritis 5. Ulcer of stomach 5. Pneumonia.and influenza 6. Puerperal causes 6. Accidents {non-vehicle) 7. Goiter 7. Accidents vehicle) 8. Homicide 8. Suicide 9. Cerebrospinal menin- 9 . Congenital malformat ion gitis 10. Hernia ll. Syphilis 12. Diarrhea 13. Whooping cough 14. Scarlet fever Among the causes of death which are classed as pro-1 dominant in urban places are three which are often considered among the better measures of healthful environmental condi- tions in a population. These diseases with their rates are tuberculosis 36.3, premature births 26.1, and puerperal causes 6.1. The percentage excesses of these rates over corresponding rates for rural areas are respectively 17, 23, and 31. Tuberculosis, a transmissible disease, is usually associated with low economic and educational 90 conditions. Detroit tends to dominate the urban death rates of Michigan, and it has a.high rate for tuberculosis.1 On the assumption that the major portion of recent migrants to the industrial areas of Michigan came from the Southern States, particularly'from Kentucky and Tennessee, a partial explanation of the high urban rate may be attributed to the fact that rates from tuberculosis have been high in these Statehz However, it this is true, a more difficult thing to explain is the higher rate for non-Whites in Detroit than exists in these Southern States. 1. The death rate from.tuberculosis in Detroit is 48.5 per 100,000 pOpulation. The urban rate for Michigan is 34.1, but minus Detroit it is only 25.5. For rural areas it is 30.0. Detroit has 73 percent of the State's urban non-white population. Since 91.4 percent of Michigan's colored papulation is classified as urban, it follows that most of them are located in Detroit. The tubercu- losis rate for colored persons in Detroit is 181 whereas in urban areas outside of Detroit it is only 130.2. The next largest city in Michigan, Grand Rapids, has the low~ est death rate from this cause of any city in the United States, namely 15.6. See Byrd, Oliver 3., Health Lg: structiog'Yearbook, 1945, Stanford University PPess, S anfor niversity, California, 1945, p. 138. About 1.1 percent of the population of Kent County (including Grand Rapids) is colored, while 8.1 percent of Wayne County (including Detroit) is colored. See Beagle, J. Allan, _920 M0, Table 2, D. '79s 2. Palmer, Carroll 3., Tuberculosis'ig the United States, ,Medical Research ComfiIttee, fiatIonal‘TEbercqusis Assoc- iation, New York, Volume 4, 1946, pp. 79, 63, and 159. .According to this study, three-year average death rates from tuberculosis for the years 1939-1941 were as follows: State .All Races White Nonewhite Michigan 34.8 ---- ----- Detroit 5007 3605 18900 Kentucky 68.9 61.2 163.2 Tennessee 78.8 61.6 160.1 91 One explanation is this. It is quite possible that latent lesions are reactivated by the poor working and living con- diticns1 which characterize the social environment of these migrants. J. Burns Amberson, Jr., Professor of Medicine, Columbia University, has this to say:2 "If the infection (tuberculosis) is acquired in early childhood and the individual survives the immediate effects, the lesions usually remain latent until adolescence or adult life when reactivation is most likely to occur." He observes further, '.... Poorly fed and housed people have a higher incidence of the disease. .... It is known that laborious occupae tions under poor working conditions, exemplified by un- skilled laborers, involves a death rate from tuberculosis which is five or ten times higher than.that among pro- fessicnal classes. The factors of low wages and living conditions are of course interwoven. It is self-evident that the ultimate conquest of the disease must take into consideration these and.other incidental causes as well as the factor of infection alone.” Premature births rank second and puerperal causes sixth on the list of predominant causes of death in urban l. Freedman found in his study of health differentials among rural-urban.migrants that "in 1936 the health of recent rural-urban.migrants who remained in the city was poorer than that of the general urban population." See Freedman, Ronald, I'Health Differentials for Rural- UrbanIMigration,' American Sociological Review, velume 12. N00 5, 1947’ p. 541. s. Stieglits, Edward J., ed., 22. c t., pp. 356-538. 92 areas. One might have expected these rates to be lower in urban areas because medical care and health facilities are more accessible. However, over-activity on the part of the expectant mother, criminal abortions, and similar sit- uations tend to increase the rates.1 Therefore, it is pro- bable that the stress and strain of urban life encourages habits and attitudes of living which are unfavorable to motherhood and this is reflected in the urban rates. In addition, the concentration in Michigan's industrial areas of Southern migrants whose levels of living are low2 is offered as a further explanation of these rates.3 In rural areas, one might question the presence of high rates for motor vehicle accidents, suicides, and syphilis. They are 38.5, 16.0, 15.8 respectively. Rural residents die from.hoth motor vehicle and all other acci- 1. See Smiley, Dean Franklin, and Gould, Adrian Gordon, Communit H ions, The MacMillan Company, New York, "I946, pp. 2&3-285. See also Dublin, Louis,‘gt.gl, _920 Me, p0 1750 2. 1H ood margaret Jarman, Rural Level of Livi Indexes 33: Counties _o_f_ 1;_h_e_ United States, I935, ureau 0 Agricultural Economics, United States Department of Agriculture, Washington, D.C., 1943. The Rural Composite Index for Michigan was 114, for Kentucky 83, and for Tennessee 77. The higher the index the higher the level of living. 3. Burke found a statistically significant relationship between the diet of the mother during pregnancy and the condition of her baby at birth. In 216 cases studied all still-born premature and “functionally immature" infants were born to mothers whose diets were inadequate. Burke, Bertha.S.,'Nutrition and Its Relationship to the Complications of Pregnancy and the Survival of the Infant,‘ §¥§§ican Journal 2; Public Health, Vol. 35, No. 4, Apr 945, pp. 334-339. II ’I I. 93 dents about 30 percent more than urban people. There are a number of factors contributing to this situation; the young and the old, both considered greater accident risks than the remaining population, are concentrated more heavily in rural areas,1 speed laws are higher in the country and often not so well enforced, and occupational accidents in agriculture are numerous. Suicide occurs more often in the country than in the city, the rates being 16.0 and 12.8 for 1940. This is con- trary to the findings of Dublin (as reported by Gist) in his study of suicide in 180 cities in the United States for the years 1930-31.2' He found that suicide rates were high- er in urban than in rural areas and that they increased as the size of the city increased. Gist points out3 that "one should exercise care in assuming that suicide is always associated with urbanism, or at least that urban rates necessarily exceed the rates for rural areas. Gavan found a greater difference between urban rates in different re- gions than exists between urban and rural rates for the country at 1arge.'4 l. Beegle, J..Allan, Mlchi Po ulation, Special Bulletin 342, Michigan AgricuIturEI 0 age, Agricultural Experi- ment Station, East Lansing, Michigan, 1947, pp. 36-40. 2. Dublin, Louis I., _Tg _Bg _Qr_ Not 29 _B_g. 3. Gist, Noel P. and Halbert, L.L.,‘Urban Societ , Thomas Y. Crowell Company, New York, 1945, p. 383. 4. Gavan, Ruth Shonle, Suicide, University of Chicago Press, Chicago, Illinois, 1933, pp. 46-54. 94 In 1940 the suicide rate in Michigan for rural males exceeded that for urban males by 33 percent, being 25.7 and 19.2 respectively. On the other hand, the rate for urban females followed the national pattern and exceeded that for rural females by 18 percent. The rates are 6.2 and 5.1 respectively. It is the author's Opinion that several factors are cperating to produce these high rural rates for suicide in Michigan. The rural population is preponderant- 1y composed of older males who are frequently the victims of certain dread diseases, such as cancer, coronary throme basis, or senile dementia. It has been found that ill health is a leading motive in 40 percent (Whether success- ful or not) of all attempts at suicide by males; and that domestic difficulties account for another 30 percent.1 Moreover, lumbering drew a large number of unattached men into Northern Michigan. The rapid decline of this industry and the State-wide depression of the thirties contributed further to the personal disorganization of males. (See Chapter VII for counties with high suicide rates.) And finally, it is probable that religious mores among males are weaker and hence do not operate as deterrents of sui- cide to the extent they do among females. Only one negro in the rural areas of the State comp mitted suicide and only 16 in the urban areas took their 1. 800 Metropolitan Life Insurance Company, “Why Do People Kill Themselves?“ Statistical Bulletin, 26; No. 2, February 1945, pp. 9- 0. I. 95 lives. It would appear that suicide in Michigan is essen- tially a rural-white, male, behavior-pattern. Perhaps racial discrimination has built within the colored popula- tion a group consciousness which militates against sui- cidal tendencies among its members. Homicide is largely an urban phenomenon in Michigan. The urban rate exceeds the rural rate by 39 percent. The rates are 3.8 per 100,000 population and 1.5 respectively. Mereover, it is essentially a non-white characteristic. Probably this is often an expression of repressed antagonv ism. In 1940 the rate per 100,000 population was 1.9 for thites and 28.6 for non-whites.1 Undoubtedly, legislation controlling the use of firearms has done much to reduce this death rate. Age-Adjusted 21113;, M; _f_o_p_ Selected Causes When the age distributions of rural and urban pepu- lations are taken into account, the death rates for the highest of the twenty-five principal causes discussed in this section present a someihat different picture. Eight of the leading causes have been adjusted for age and are Presented in Table 4.12.8 This table shows the ratio of letual to expected deaths for each cause, "expected", being A g 1. See Chapter V, p.127 for a detailed discussion. 2. Expected deaths are obtained in the same manner as in the indirect method of adjusting death rates. See Appendix I, pp. 321-325. 96 Table 4.12 AGE-ADJUSTED DEATH RATES FREM SPECIFIC CAUSES, MICHIGAN 1940, BY RESIDENCE Cbserved Death Rates Ratio of Actual to Cause of per 100,000 Expected Deathsz Death (1) (z) (3) (4) Rural Urban 2x100 Rural Urban 4x100 4__ JL 3 Heart 33809 27409 8101 9702 10401 10701 Cancer 12706 11501 9002 9309 10603 11302 Intracranial lesions of vascular origin 11006 7804 7009 10201 9807 I 9607 Accidents (All) 88.3 67.7 76.7 114.9 97.3 84.7 Nephritis 6303 5300 8307 9408 10304 10901 Pneumonia 6002 4900 8107 10006 9909 9903 TUDOIQUlOBi. 3000 3603 12100 9008 12302 13507 Diabetes mellitus 26.4 26.7 101.1 84.4 109.8 130.1 All causes 1097.0 938.4 85.5 96.8' 102.6 106.0 IUrban includes all incorporated places with 2,500 inhab- itants or more in April 1940. Rural embraces all the area outside of cities of 2,500 inhabitants or more in 1940 3 The expected deaths are computed by applying the age-spe- cific death rates in the total population of Michigan from each cause of death, to the corresponding age group of the urban and rural populations respectively, and summing for See Dublin, Louis I. and Lotka, Alfred J., Leggth,gf,Life, p. 95, for an example of this method of adjusting cause-specific death rates for age in rural and all ages. urban areas. Age-specific rates for these eight causes are frcm Linder, Forrest E. and Grove, Robert D., Vital Statisti s Rates.ig the United States, 1900-1940,Ta5Ic 25: PP. £5 6-4570 97 the number of deaths that would have occurred in each area if its death rate for each'cause had been the same at each age as corresponding rates for the State of Michigan. 0f the eight causes adjusted in this table, five have rates that are higher in.urban.than.rural places. The crude heart dis- 0ase rate in urban areas, for example, is 19 percent lower than the rural rate, but when the rates are adjusted for age, the urban rate is seven percent higher than the rural rate. The three diseases which.have higher rates in rural than in urban areas are intracranial lesions of vascular origin, accidents, and pneumonia, the latter two rates being approx- imately equal. Trent -five Principal Causes _o_f_ 29512;; in Michigan by M _a_n_q esidence The specific causes of death in rural and urban areas by race are presented in Table 4.13. Because there are re- latively few colored people in.Michigan,1 an examination of the crude rates for white peeple reveal that race2 has only a slight influence on the total rural and urban death rates in this state. The total rural rate for all causes is 11.0 per 1,000 pepulation and the rural white rate is 10.8, a 1. There are 216,463 colored peeple in.uuchigan of whom 91.4 percent are located in urban areas. 2. Comparisons of death rates of the colored and white population by cause should be made with reservation and caution when given.for rural and urban areas of Michigan, since there are few colored in rural areas. Only seven of the twenty-five causes listed had fre- quencies of more than 20 deaths for rural colored people. 98 difference of only .2, while there is a difference of only .1 between the rates of total urban and urban whites. Thus, it appears that death rates for all causes in Michigan are not influenced very much by deaths occurring to the rela- tively few colored peeple in the population. However, the total urban death rates for a few causes do show an in- fluence of the exceptionally high rates for colored people. These causes are listed below: Urban total death rate Urban bz causes White Difference Pneumonia and influenza 49.0 46.8 2.8 Tuberculosis 36.3 28.2 8.1 Syphilis 9.3 7.6 1.7 Homicide 3.8 2.1 1.7 It is interesting to observe that rates of the colored pop- ulation for these specific causes are particularly high, and it would be unusual if they did not exert an upward influence on.the total rates in urban areas. However, the total urban death rate for cancer, 115.1, was influenced downward by the colored population. It is 2.7 less than the urban segment. This is probably best explained in terms of age. Colored people, among whom cancer takes a high toll, have relatively fewer of their number distributed among the older age groups than do White people.1 . As might be expected, the rates of specific causes for the total rural population are influenced.less by the colored group than the corresponding urban rates. The 1. See also Beagle, J..Allan,‘gp..g;t., p. 41. .‘l I‘ l a ‘. C .u. 5" 0.qu 0'. 0.1! 1‘ .11! .. i \I- III a\.l Ilia 1...! 99 Table 4.13 MDRTALITY RATES FROM COMMON CAUSES OF D H IN'MECHIGAN, 1940, BY RACE AND RESIDENCE Rates per 100,000 population Cause WHITE OTHER Degth Rural Urban Ratio Rural Urban Ratio Rural Rural Urban UrEan Diseases of the heart 337.9 274.0 1.23 438.3 388.7 1.13 OEDOCr 12600 11708 1007 27709 7003 3095 Intracranial lesions of vascular origin 111.0 79.3 1.40 176.4 63.7 2.73 Raphritil 6303 5104 1025 5808 7904 074 Pneumonia and influenza 59.0 46.8 1.26 176.4 85.0 2.07 Accidents (nonrvehicle) 49.1 37.9 1.30 112.2 29.3 3.83 Accidents (motor-vehicle)37.9 31.0 1.22 96.2 17.7 5.43 Tuberculosis 2704 2802 097 27206 17004 1060 Diabetes mellitus 26.1 27.0 .97 53.5 19.2 2.79 Premature births 20.0 25 .0 .80 '55?! 41.5 .77 Suicide 1603 1301 1024 503 706 070 Congenital malformation 14.4 13.9 1.04 '573 . .80 Syphilis 12.7 7.6 1.67 50177 5 . 8.35 Hernia 908 808 1011 503 1206 042 Appendicitis 7.1 10.4 .68 E 17.7 .30 Cirrhosis of the liver 7.1 10.5 .68 --- 71:1 --- Ulcer of the stomach 5.7 6.2 .92 16.0 6.6 2.42 Puerperal causes 4.2 6.0 .70 15.7 '776 1.40 Diarrhea enteritis, etc.33.7 3.0 1.23 _55 571' 1.04 nxophthsimis goiter 2.9 4.1 .71 --- 3:0 ---- Whooping cough 1.5 .8 1.98 3,53 33; 2.65 HOEiOidO 104 201 067 '12:! 3005 035 Scarlet fever 1.0 .7 1.43 --- --- --- Diphtheria 1 0 4 1 e 00 "’ 0 5 “u"- Cerebrospinal meningitis '__._"I ‘73: .25 ~-- :3 --- {‘11 Causes 108208 92600 1017 851706 114303 2020 IBased on pepulation enumerated as of April, 1940. 3 Includes only deaths under 2 years of age. note: Source: Rates based on less than 20 deaths are underscored. of the United States, 1940, Part Vital 8 ti ti a ta 5 c Fla-15 , s enee, a . pp. 223-2, 0 100 causes for which rural death rates of White peeple are lower than the total death rate for corresponding causes in rural areas are listed below: Rural total death rate Rural by cause White Difference Cancer 127.6 126.0 .6 Tuberculosis 30.0 27.4 2.6 Syphilis 1508 1207 301 Thus, it appears that only certain cause-specific crude death rates for all races change materially when they are made specific for the White race. The urban rates are influenced slightly'mmre than the rural rates, because the colored population is concentrated in the urban industrial areas of Michigan and have excessive death rates for these causes. Tuberculosis, syphilis, pneumonia, and homicide among the nonswhites account for a portion of the high rate for these diseases among the rural and urban pepulation. The remaining causes among the twenty-five listed have ap- proximately the same rates for all races in the two resi- dent areas as for the white race.1 Age-adjusted Death Rates for Selected Causes 2y Race and eg ence Table 4.14 enables one to compare age-adjusted rates from the leading causes by race and residence. Observation of these rates along with those of Table 4.12 reveals clear- ly that finite pe0ple in rural and urban places dominate the total death rates in these areas for the causes listed. l._ These crude cause-specific death rates are discussed more fully in Chapter V, Mortality Differentials by Race. 101 Table 4.14 AGE-ADJUSTED DEATH RATES FROM SPECIFIC CAUSES, MICHIGAN, 1940, BY RACE AND RESIDENCE Ratio of Actual to Expected Deaths1 0:?” White 1 Colored math Mina %x 100 233:1 URL. 5:15: 100 Accidents (111) 108.2 95.5 88.1 243.8 75.0 50.8 Intracranial lesions of vascular origin 101.2 97.4 96.2 173.6 135.5 78.0 Pnemnonia 98.5 94.5 95.7 500.0 207.4. 59.1 Heart diseases 96.7 101.5 105.0 136.7 170.4 124.7 Nephritie 94.8 98.0 105.4 31,1 £219.12 255.5 Cancer 92.7 104.7 112.9 215.7 89.1 41.1 Diabetes mellitus 83.5 109.5 150.9 £92.29. M 55.9 Tuberculosis 83.0 83.0 100.0 850.0 503.0 59.2 111 Causes 95.5 99.5 104.5 257.9 157.4 70.4 I The expected deaths are computed by applying the age- specific death rates in the total population of Michigan from.each cause of death, to the corresponding age group of the urban and rural pepulations respectively, and summing for all ages. See Dublin, Louis I, and Lotka, .Alfred J., Le th‘g;,Li£e, p. 95 for an example of this method of adjusting cause-specific death rates for age in ruraland urban areas. It is essentially the principles employed in the Indirect Method of adjusting death rates for age, which are described and illustrated in Appendix I, p321ff.The age-specific rates employed as a standard for these eight causes are from Lindsr, Forrest 3., and Grove, Robert D.,‘gp..git., Table 23, pp. 446-477. 3 Urban includes all incorporated places‘with 2,500 in- habitants or more in.April 1940. 3 Rural embraces all the area outside cities of 2,500 in- habitants or more in 1940. 102 Tuberculosis is the only disease that seems to have been greatly influenced by race. The tuberculosis rates for all races in rural and urban areas are respectively 90.. and 123.2, the urban rate being 35.7 percent in excess of the rural (See Table 4.12). When.the colored group is eliminated, the rates computed for'Whites only are 83.0 in both rural and urban areas. (Table 4.14) The rural-urban mortality differentials for all other causes for whites listed remain approximately the same as the ratios for total rural-urban areas given in Table 4.12. The rates for colored people are in excess of those of the white p0p- ulation for all specific causes presented in both rural and.urban areas with the exception of accident (all) in urban areas.1 Infant Death Rgtes‘ig,Rural and Urban Areas According to data for 1940 presented in Table 4.15, the rural people of Michigan living in places of less than 2,500 population have 42.1 infant deaths per 1,000 live births as compared with 39.3 for those living in urban places of 10,000 population or more. Large cities (100,000 or more population) lose fewer babies per 1,000 live births than any other populationesize group. The rate for these cities is only 37.5. Small cities (2,500-10,000) have the highest infant death rate of all segments considered, name- 1. This generalization eliminates nephritis and diabetes mellitus because these two causes are based on less than 20 deaths for non-whites in both areas. 103 1y, 45.0 deaths. Mediumrsized cities (25,000—100,000) hate the next highest infant death rate among both urban and rural groups. This rate is 43.2. It is evident from this data that infants living in urban places of 10,000 or:more have slightly better chances of survival than rural infants or those living in small cities. However, this slight advantage almost disappears when the rate is aver- aged for the four years 1938-19413 the rate becomes 41.2 -in urban areas and 41.8 in rural areas. Table 4 e15 INFANT MORTALITY RATES MICHIGAN, BY POPULATION SIZE GROUPS, 1940 {By place of residence) Infant deaths under 1 year of Population 812° Group age per 1,000 live births Total P0pulation 40.7 Tatal: 10,000 population or more 39.3 Cities of 100,000 pOpulation or more 37.5 Cities of 25,000 to 100,000 population 43.2 Cities of 10,000 to 25,000 pepulation 39.2 Total: Cities under 10,000 pepulation 42.7 and rural Cities of 2,500 to 10,000 pepulation 45.0 Rural p0pulation 42.1 Source: Linder, Forrest E. and Grove, Robert D.,‘gp.‘gi§., Table 28, p. 589. ' Infant 9.9.2.12 3322; by Selected Causes According to data presented in Table 4.16, rural infant death rates are higher than urban rates for 11 out of 20 classified causes of infant deaths. Urban infants succumb from three of the six most common causes of infant .JI‘ Iris I1" Ills U I‘I' 1|..l“.‘ i‘rlfl,‘ ‘IIIEIII‘..CII‘II\.-IOI‘III.I ‘Illl 1‘ 104 Table 4e16 INFANT MORTALITY RATES, MICHIGAN 1938-1941 (By cause and residence‘ Sglgctod Causes Deaths Per 100,000 live births Total Urban Rural .‘11 CEQBOI 4,13703 4,118e8 4,177e0 Premature births 1,352.1 1,442.5 1,158.2 Congenital malformations 570.0 565.3 579.9 Pneumonia (All forms) 557.9 507.1 666.9 Injury at birth 401.8 425.7 350.5 Other diseases peculiar to first 259.7 265.7 246.8 year Diarrhea, enteritis, etc. 229.5 216.3 257.9 Mechanical suffocation and other accidents 98.1 77.4 142.4 Diseases of thymus gland 96.0 86.6 116.3 Congenital debility 72s? 58e6 102e8 Whooping cough 56.8 46.4 79.1 Other diseases of respiratory systmm55.l 51.2 63.3 Influenza 48.3 33.5 79.9 Diseases of the nervous system. 39.7 44.2 30.1 Intestinal obstruction 35.4 36.5 33.2 Diseases of ear and mastoid process 22.9 25.4 17.4 Syphilis 19e4 20.5 17e4 Diseases of circulatory system. 18.6 14.7 26.9 Ill-defined and unknown causes 18.1 12.2 30.9 Dysentery 17.3 16.2 19.8 Diseases of genito-urinary system. 14.3 15.1 12.7 Measles 12.8 13.3 11.9 140.8 144.6 132.7 All other causes Vital Statistics of the United States, 1938, Part 1, Source: __ “T‘Tab e 24, 939, Part—1'1 940 Part 11, , Table I3; I Table 14; 1941 Part 11, Table. 19. Computed by Commission on.Hos ital Care, in State Hos ital Study Manual, Book IV, Mimeographed), Chicago , nois, Table 13. death at higher rates than do rural, while rural infants have higher rates for the other three causes. These con- trasts are emphasized below: r\ 105 Infant diseases predominant Infant diseases predominant in.g£ban.environment in rural environment Rates Premature births 14.4 11.6 Injury at birth 4.3 3.5 Other diseases peculiar to first I.” 207 2e5 5.7 5.8 Congenital malforma- tion 5.1 6.7 Pneumonia (all forms) 2.2 2.6 Diarrhea, enteritis, etc. These six causes comprised 83 percent of all infant deaths in 1940, and one of them, premature births, is among the ten principal causes of death when ranked according to incidence in the total pepulation. When.these six causes are combined, the rates for rural and urban areas in 1940 favor the rural places slightly. They are 32.8 and 34.2 respectively. Howb ever, the infant death rates from.all causes in 1940 favors the urban over the rural population, the rates being 40.0 and 42.7 respectively. Infant death rates for 1940 adjusted for underregistration of births1 present essentially the same relative ranks as the unadjusted infant rates. These rates for rural and urban places are 41.2 and 39.1 respec- tively. Thus it appears that on the basis of these data, urban infants have a slightly more favorable health level A _— 1. Corrections were made by applying the United States Bureau of Census estimate of 2.2 percent incompleteness of birth registration in.Michigan to the total registered births. The unregistered births Obtained were distri- buted to residence groups by allocating unregistered births at the same ratio as registered births to the State total in the two areas. 106 then rural infant”:L However, it should be pointed out that earlier in the Chapter age-specific death rates for the age group under one year in these two areas definitely favored the rural areas. Maternal Mortglitz Rategz 33 Rural and Urban Areas Death occurs to mothers during child-birth at a higher rate in urban than in rural areas. Mothers living in places of less than 2,500 pepulation die in child-birth at the rate of 2.3 per 1,000 live births as compared with a rate of 3.2 for mothers living in places with 2,500 or more population. When corrected for underregistrati on of births, these two maternal death rates are modified very little, the 1. Thomson points out that reductions in infant death rates may be attributed to four major factors: '(a) the better care that mildren are receiving at home, that is, primarily, the improvement in the methods and sanitation of infant feeding; (b) the decline in the number of children born to a large portion of the mothers, thus embling them to give their children better care both before and after birth, (6) the more expert medical care of children; and (d) the generally more comfortable circumstances in which a large part of the people in the more advanced nations now live.“ In the light of these factors one might point out that in Michigan because of the nature of health organization, the rural population has less opportunity for education in child and health care, less facilities available, and larger size fami- lies than the urban population. (Thompson, Warren 8., 0p. Gite, pp. 221-222) 2. Maternal deaths comprise all deaths of mothers that can be ascribed to childbirth (International List Numbers 140-160). When the number of maternal deaths is express- ed in terms of per 1,000 live births it is referred to as "maternal mortality rate," but when the number of such deaths is related to the total population it is referred to as the death rate from "puerperal causes." 107 urban rats being 3.13 and the rural rate 2.26. Considering all cities of less than 10,000 and the Open country pOpula- tion as 'rural" and, on the other hand, all cities of 10,000 and over as “urban“, the rural areas have lower maternal death rates than urban areas. These rates are respectively 2.5 and 3.2 deaths of mothers for each 1,000 live births. (See Table 4.17) Table 4.17 MATERNAL MORTALITY’RATES, MICHIng, 1940, BY POPULATION SIZE GROUP (By place of residence) Deaths of Mothers Per Population Size Groups 19000 Live Births Total Total Population 2.9 Total Urban: 2,500 pOpulation or more 3.2 Total: 10,000 population or more 3.2 Cities: 100,000 population or more 3.3 Cities: 25,000 to 100,000 population 3.3 Cities:of 10,000 to 25,000 pepulation 2.5 Total: Cities under 10,000 population and rura1 2.5 Cities: 2,500 to 10,000 population 3.4 Rural pepulation 2.3 I‘Deaths are allocated to place of residence. Rates for frequencies of death of less than 20 are underscored. Total urban computed from.Vital Statistics‘ggbthe United Source: Vital Statistics Rates in the United States 1200-1940, Bureau of the Eensus, Table 37, p. 332. The rural pOpulation has lower maternal death rates than any other populationrsize group presented.l Small cities 1. This generalization ignores the maternal death rate of 2.5 for cities of 10,000 to 25,000 because it is based on less than 20 deaths. 108 of 2,500 to 10,000 have the highest rate, 3.4, and the next highest rate, 3.3, occurs in mediumssize cities of 25,000 to 100,000 pepulation. Dorn concluded in his study of maternal mortality rates that there is practically no dif- ference between.maternal.mortality rates in urban and rural_ areas of the Nation, but that there is considerable varia- tion among the States.1 1. Born, Harold F., ”maternal Mortality in Rural and Urban Areas,” Publi Health Re orts, Vol. 54, No. 17, April 28, 1939, pp. 8"596e CHAPTER‘V MORTALITY DIFFERENTIALS BY RACE Of the 52,183 deaths which occurred in Michigan in 1940, five percent, or 2,732, were members of the colored races,l although this group constituted only 4.1 percent. of the total population. Among the 216,463 persons classi- fied as non-whites, 96.3 percent were Negroes, 2.9 percent were Indians, and .9 percent were Chinese and Japanese. Thus, with a population of 5,256,106 in 1940, the State had a relatively small number of colored people.2 As Table 5.1 indicates, the non-whites have been increasing rapidly since 1910, and the Negro segment will probably continue to in- crease in the future. For this reason, among others, it is appropriate to include race mortality differentials among Table 5.1 POPULATION OF MICHIGAN, BY RACE, PER CENT CHANGE OVER PRECEDING DECADE, 1900-1940 POpulation (1) Percent Change Year White Colored White Colored 1900 2,400,789 22,416 1910 2,806,465 25,818 16.9 15.2 1930 4,654,599 179,607 27.5 149.5 1940 5,039,643 216,463 8.3 20.5 1. Estimated as of July 1 for all years except 1940, ‘which is enumerated as of April 1, 1940. Source: Linder, Forrest E. and Grove, Robert D., Vital Statistics Rates in.thg‘United States, T‘O'Teo - 9'4"O"","'"""Tabie ‘i—"", p. 550 ff ."“""" """"'“" l. "Colored" includes Negroes, Indians, Chinese, and Japanese. mericans are included with the "White" group. 2. For a consideration of the problem or death rates based on a small number of deaths, see pp.42-43. 109 110 the important items of analysis in our description of the present mortality-health situation. Race-Specificl'Qgggg.gng,figs-Adjusted.Mortality-Rate; Both the crude resident and the age-adjusted death rates are higher for the colored than for the white popula- tion.2 In 1940 the crude death rate per 1,000 was 9.8 for the white segment, and 12.6 for the colored. The resulting differential: between the two race-Specific mortality rates assumes additional importance as a.measure of health in Michigan when it is considered in relation to the age dis- tribution of the colored races. Actually the age distri- bution of the non-whites favors a low mortality rate in this State, since proportionately there are more white than color- ed persons in the age frequencies above 45 years. It is from those age groups that the chronic diseases take their largest toll. Age-Specific rates will be discussed in the following section. l. Computation of race-Specific death rates are explained on p. 311ff. 2. Thompson states, “It has long been known that the Negroes have a.higher death rate than the whites among whom they live." Thompson, Warren S.,.gp.lgi£., p. 137. Thompson bases his statement on: Dublin, Louis I, "Recent Changes in Negro mortality," MetrOpolitan.Life Insurance Company, New York, 1924; Cover, Mary, and Sydenstricker, Edgar, "Mortality among Negroes in the United States," Public Health Bulletin 174, Government Printing Office, Weanington D. 0., June I927; Sydenstricker, Edgar, Health and Environment, McGraw- Hill Book Company, Inc., New York, 933. 111 On the basis of race-Specific death rates, there are approximately three deaths per 1,000 from the colored pOpu- lation in excess of the death rate for the white segment. This difference is greatly accentuated when the rates are adjusted for age differences. Using the age distribution of the pepulation of the United States in 1940 as a standard, the resulting age-adjusted death rate for the whites is 10.0 per 1,000 pOpulation, whereas the rate for the non-whites is 15.8. This difference of 5.8 is nmre than twice that between the crude resident death rates, namely, 2.8. Age-SpecifJLq 9.29.3.9. £315.93 In Table 5.2, the age-specific death rates for the white and non-white segments are presented. Table 5.2 AGE-SPECIFIC DEATH RATES FROM ALL CAUSES PER 1,000 POPULATION, MICHIGAN, 1940, BY RACE (Allocated to place of residence.) W Rates per 1,000 Ratio: Excess: Populgtion Colored Colored Age Group Wh it e C 01 ore d Wh it e W Total 9.8 12.6 1.3 2.8 Under 1 47e3 66e9 104 19e6 1 ‘ 4 gel 3e9 ~le9 1.8 5 - 14 1.0 1.4 1.4 .4 15 - 24 1.6 5.0 3.1 3.4 25 ‘ 3‘ 2.3 6e8 3e0 4e5 35 - 44 4e2 11.2 2e? 7e0 45 - 54 9.1 21.8 2.4 12.7 55 - 64 20.8 35.8 1.7 15.0 65 ' 74 48e0 57sl 1.2 gel 75 ' 84 114e1 121e7 lel 706 85 0V0: 238e8 129e0 "' "“ Source: Vital Statistics Rates‘gg the United States 1900-1940, Table II, p. 202. 112 These figures reveal that death rates are higher for the colored population than for the white in every age group but one--those 85 years and over. The lowest age-specific death rate among the series is 1.0 per 1,000 pOpulation for whites age 5-14 years, While the highest is 238.8, also for whites, but at age 85 and over. Some of the greater rela- tive differences found between the two groups occur in age frequencies under 45 years. For example, at age 15-24 years the rate 5.0 for the colored segment is more than three times as high as that of the white pepulation, namely 1.6. At age 25-34, the death rate for the colored groups is three times that of the white. Moreover, the number of infant deaths under 1 year of age per 1,000 population is 47.3 for whites and 66.9 for nonrwhites, the latter being approxi- mately 40 percent higher than the former. Among the age groups 45 years and above, the highest relative difference is at age frequency 45-54, where the rate for the colored population is about two and a half times that of the white. An understanding of the full significance of age- specific mortality rate differentials by race is more readily grasped if one observes the rates along with the population age distribution of the respective groups. These data are given in Table 5.3. The white pOpulation in Michigan has a much higher proportion of its number over 44 years of age than does the colored pOpulation. It is well to remember that the higher the proportion of the total population of a given class in the age groups above 44 years, in general, the 113 Table 5e3 POPULATIONl OF MICHIGAN, BY AGE AND RACE, 1940 _gopulationl Per 10,000 POpulation Age Groups White Coloréd White Colored Total 5,039,643 216,463 10,000 10,000 Under 1 80,326 3,481 159 161 1 - 4 334,115 13,462 663 622 5 - 14 846,906 35,866 1,681 1,657 15 - 24 900,303 35,068 1,786 1,620 25 - 34 807,070 41,778 1,601 1,930 35 - 44 719,713 43,920 1,428 2,029 55 - 64 398,202 10,695 790 494 65 - 74 224,713 5,129 446 237 75 - 84 85,956 1,372 171 63 85 Over 13,343 341 27 16 1. POpulation enumerated as of April 1, 1940. Source: Linder, Forrest E., and Grove, Robert D., Vital Statistics Rates in the United States, 1900-19407—__ Mpmsl'ése'i‘ ""‘"" ‘— higher is the risk of death, since it is at these ages that the chronic diseases take their toll. Yet, with this handi- cap, the white race has lower death rates in all age fre- quencies above 44 years except one--those persons 85 years and over. On the other end of the life span, ages 1 through 24, the two groups have approximately equal proportions of their pOpulations in each of the age frequencies. Yet in every instance, the mortality rates for the colored segment are considerably higher in those age groups than are those for whites. Ordinarily, a population with a relatively low proportion of its total number distributed among the older age groups, and with a relatively high proportion in younger age groups, has a low crude resident death rate. However, 114 this is not true of the colored segment in Michigan. Fur- thermore, for the age frequency 25-44, in which almost 4,000 out of every 10,000 colored persons fall, the age- specific rates are almost flares times the death rates of the white pOpulation. This condition stems, in part, from the fact that a large prOportion of the colored people in Michigan migrated from Southern States, where standards 1 of living have been low,, and where Opportunities for health education.and medical treatment have been more restricted. Sex-Mortality Differentials by R_a_c_e_ In Table 5.4, the sexsspecific death rates are given for both the colored and the white pOpulations. Table 5.4 SEX-SPECIFIC DEATH RATES IN MICHIGAN BY RACE, 1940 (By place of occurrence) Rates per 1,000 Population Male 7 Female1 Race Both Sexes Crude Adjusted Crude Adjusted White 9.8 10e8 llel 8e7 8.8 Colored 12.6 13.7 17.0 11.6 14.6 1. Adjusted to the age distribution of the United States pepulation, 1940, by the direct method. Source: Crude data from Linder, Forrest E., and Grove, Robert D., Vital Statistics Rates ig,§gg'United States 1300-1940, Table 10, p. 193. Crude rates for both sexes are lower for whites than for non-whites. They are respectively 10.8 and 13.7 for males 1. See Footnote 2, p. 92. 115 and 8.7 and 11.6 for females--a difference of 2.9 between the racial groups for both females and males and a differ- ence of 2.1 between the sexes for both whites and non- Whites. Thus death rates for males are higher than for females in both racial groups in Michigan.1 It has been shown that the crude resident death rate of 12.6 for the colored races is higher than that of the white race, namely 9.8; and that the resulting dif- ferential becomes still greater when these rates are ad- justed for age differences. The adjusted rates are 15.8 and 10.0 respectively. This situation persists in Spite of the fact that both the age and sex pOpulation composi- tion for the colored segment favors a lower crude rate for them.than for the white group. The colored races have pro- portionately fewer males in their population, which tends to favor low rates since males generally have higher death rates than females. The sex-ratios are 105.3 for whites and 103.8 for non-whites. Sex-specific death rates for the two population groups are also higher when they are adjusted for age differences (using the pOpulation of the United States for 1940 as a base). “The rate for male whites increased .3, namely, from 10.8 to 11.1 per 1,000 population. 0n the other hand, the rate for male non-whites increased 3.3, or from 13.7 to 17.0. 1. This generalization is true for each of the 48 States. Nbles have higher crude death rates than females for the United States and all of the 48 States for both white and colored racial groups in 1940. See: Linder, Forrest E., and Grove, Robert D.,‘gp.‘git., Table, pp. 127-149. 116 The rate for female whites changed only slightlyb-from 8.7 to 8.8 but the rate for female nan-whites increased from 11.6 to 14.6, a difference of 3.0. When the crude resident death rates for the two groups were“adjusted for both age and sex, the death rate for the white pepulation increased only slightly~~from.9.8 to 9.9, whereas the death rate for the colored population increased from.12.6 to 15.8. Sex-specific death rates by age group provides more detailed comparisons of race-mortality differentials. These figures are given in Table 5.5. Table 5.5 Ass AND SEZPSPECIFIC DEATH RATES BY RACE, MICHIGAN, 1940 (By occurrence) Rates Per 1,000 POpulation White Colored Age Grgup ‘ meg: Female Male Female All Ages 10.8 8.7 13.7 11.6 Under 1 54.3 39.9 73.1 61.2.,I 1 ‘ 4 2e3 2.0 4el 3e8 5 - 14 1.1 .8 1.4 1.4 15 - 3‘ leg 1.3 4e5 5e6 25 ' 34 2e5 2e0 7e3 6e5 35 ' 44 4e6 3e7 10e8 11e5 45 ’ 54 10e4 7e5 23e9 19e3 55 ' 64 23e8 17e5 36e6 35e2 65 “,7‘ 53el 42e7 66e7 48e8 75 ' 84 122.2 106e7 141e3 105.0 85 07.! 249e5 230e3 157e1 104e5 Source: Linder, Forrest E., and Grove, Robert D., Vital Statistics Rates in the United States 190O=I§ZU, TablO 23. De 476s As might be expected, the death rates of the colored races are in excess of those of the white race at all ages and 117 for both sexes, except for rates of females at ages 75 and over and for males 85 years and over. In.genera1, the dif- ferences pointed out in the section on age-specific rates by race are true for the Sex-age-specific death rates. In addition, the following observations may be indicated: (1) For whites, the sex-age-specific death rates for males are higher than those for females. (2) Among non-whites, females have death rates in three age groups Which are equal to or exceed those of the males. These age groups are 5-14, 15-24, and 35-44. (3) From ages 0-1 and 5-44 with the ex- ception of age group 25-34, there tends to be less difference between death rates for males and females of the colored races than between male and female rates of the white race. Table 5.6 DIFFERENCES IN AGE-SPECIFIC DEATH RATES, MICHIGAN, 1940, BY SEX AND RACE Age Excess of : Excess of Colored Groups male Over Femgle : Over White White : Colored: male : Female All Ages 2.1 2.1 2.9 2.9 Under 1 14e4 lleg 18e8 21e3 1 - 4 e3 e3 1e8 1s8 5 ' 14 e3 e0 e3 e6 15 ‘ 24 e6 "" 2e6 4e3 25 - 34 e5 e8 4e8 .4e5 35 - 44 .9 ---- 6.2 7.6 45 “ 54 2e9 4e6 13e5 11e8 55 - 64 6.3 1.4 12.8 17.7 65 ' 74 10e4 l7e9 13e6 6.1 75 - 84 15.5 36.3 19.1 ---- 85 - up 19.2 52.6 ----' ---- Source: Linder, Forrest E., and Grove, Robert D., Vital Statistics Rates,;n the United States 1900-I940,p. 476. Note: In this table zero indicates no difference and dotted line indicates a minus difference. 118 The differences indicated in Table 5.6 which seem.to invite explanation are those in which females among the colored pepulation have equal or higher death rates than do males. It is likely that between the ages of 5-44 the col- ored female does not live as sheltered a life as does the white female, and therefore she is exposed to nearer the same risk of disease and death as is the colored male. 'The white female on the other hand occupies a more protected role than the colored female, and She is not subjected to as much physical labor and drudgery. Rural-Urban Differentials A comparison of the crude and age-adjusted death rates in Michigan, as indicated in Table 5.7, reveals sev- eral striking differences between the colored and white segments of the population in both rural and urban areas. One of the most outstanding mortality differentials among those presented, results from.the crude death rates of whites and non-whites living in rural areas. Table 5.7 CRUDE AND ADJUSTED DEATH RATES PER 1,000 POPULATION BY RACE AND RESIDENCE, MICHIGAN, 1940 White Colored Difference Regidence Crude Adjusted Crude Adjusted Crude Adjusted Urban 9.2 10e2 lle4 14e7 2e2 4e5 Rural 10.8 9.7 25.2 24.6 14.4 14.9 Difference 1e6 e5 15e8 9.9 Source: Taken from Table 5.9. 119 The colored pepulation in rural areas dies at almost two times the rate of those in urban areas. The rates are respectively 25.2 and 11.4. This is contrary to the find- ings of Kemp and Smith in Louisiana and of Molyneaux in Virginia but is in keeping with the conclusions of Vance.l This high crude rate among colored groups in the rural areas of Michigan is difficult to explain, and it. shOuld be subjected to special study. Ordinarily a rate as high as 25.2 for a pOpulation group in an area as advanced in health care and sanitary education as is Michigan would be due to an aged population. However, this does not appear to be the explanation in this case. Such a conclusion is indicated by the only slightly lower age-adjusted rate of 24.6. It is probable, in the Opinion of the writer, that the high rate is the result of a combination of factors. It is quite possible that there is a tendency for disabled non-whites to settle in the rural areas of Michigan. It has 2 already been pointed out that Freedman found the incidence of disease greater among migrants than among natives. Per- 1. In Louisiana the rural colored crude death rate in 1940 was 11.4 and the urban 17.9, (Kem , Louise, and Smith, T. Lynn, OD. th., Table II, p. 17 .A similar differ- ential characterizes the colored of Virginia (Mblyneaux, Lambert, 23 al., pp. th., p. 534). Vance found the negro's highest death— rate to be in the urban South and his lowest in the rural South. In the North the stand- ardized rural rate is 18. 2 while the urban rate is 17.1 (Vance, Rupert, _p. th., p. 347). Higher mortality rates for the rural colored pOpulation in the North than for the rural colored segments of the South may be partly due to incomplete registration of deaths in the South (Sea Willcox, Walter F.,'gp.'th., p. 213). 2. Freedman, Ronald, gp. cit., p. 541. 120 haps colored people who find it difficult to compete for a living in the industrial areas prefer to seek agricultural pursuits hero rather than to return to the South. Moreover the economic and educational levels of rural Negroes as com- pared with urban may be such as to contribute to high death rates. Vance has this to say} "Excess Negro mortality is made of the elements that cause excess deaths everywhere. It is related to occupational factors found in rough, heavy, work, to poor housing, heating, and sanitation, to inadequate nutrition and poor medical care, and to that ignorance which condemns a people to both, when they might secure better."l Another contributing factor may be a high death rate among the Indians of Michigan living in rural areas. A comparison of the crude death rates of the colored and white population living in rural areas reveals that rural whites have a crude death rate of only 10.8, compared with 25.2 for the rural colored racial groups. The differ- ence between these two rates is 14.4. This indicates that colored persons in rural areas die at a rate more than twice as high as White people in.the same areas. The colored races living in rural areas have an age distribution some- what similar to an immigrant pOpulation in that they have a larger proportion of their number among the middle-aged groups than the white race, and analler proportions among the younger and older age groups. These data are presented 1. Vance, Rupert, 22, cit., p. 349. 121 Table 5.8 POPULATION PER 10,000, MICHIGAN, 1940, BY RACE, AGE, AND RESIDENCE Number per 10,000 Age Rural fl__ Urban Groups White Colored White Colored 0 - l 175 136 151 163 1 - 4 765 619 607 622 5 - 14 1,875 1,655 1,574 1,657 15 - 24 1,717 1,686 1,825 1,614 25 - 34 1,425 1,586 1,698 1,962 35 - 44 1,221 1,565 1,542 2,073 45 - 54 1,140 1,245 1,307 1,164 55 - 64 873 816 745 _ 464 65 - 74 554 471 386 215 75 - up 255 221 165 66 Total 10,000 10,000 10,000 10,000 Actual Population 1,782,530 18,709 3,257,113 197,754 Source: Computed from.Linder, Forrest E., and Grove, Robert D., Vital Statistics Rates in the United States 19063I616,‘TEEI3‘I77 p. 945. in Table 5.8. In spite of this age distribution, Which favors a low death rate, the colored segment in rural Michi- gan has much higher death rates than does the white segment. The age-adjusted death rate of rural whites is only 9.7 as compared with a rate of 24.6 for rural non-whites. ‘ Apparently the industrial centers of Michigan select the healthier colored individuals, because the mortality differential of white and colored people living in urban areas is not nearly so great as that of these two racial groups in rural areas. The urban white population has a crude rate of 9.2 per 1,000, as compared with a rate of 11.4 122 for the urban colored population. Possibly, colored people in urban areas tend to adapt to proper health habits and to make greater use of medical facilities than do those in rural areas. Although the crude death rate of the urban white pOpulation is only slightly lower than.that of the urban colored pOpulation, the death rate of non-whites, when adjusted for age, increases more than does the rate for whites. These age-adjusted rates are 14.7 for the colored and 10.2 for the whites. Since colored people in urban areas have lower proportions of their number in the older age groups and an excess in the middle-age groups, the colored pOpulation in urban areas should have a lower death rate than the White urbanites if the two groups are practic- ing the same health habits, obtaining the same amount of medical care, and have similar levels of living. Thus, we might contend that although urban colored people have an age distribution which favors a low death rate, they have higher death rates than white people in urban areas mainly for rea- sons that could be controlled with prOper preventive measures. In Tabde 5.9 death rates for the two racial segments of Michigan are classified by size of city. Death rates for the white pOpulation vary inversely with the size of the urban center, the large cities having the lowest rate. This relationship is also true among colored peOple for the crude rates but not for the age-adjusted rates. Small cities have the highest rate for the white pOpulation but for non-whites the rate is higher in.the rural areas. 123 Table 5e9 DEATHS PER 1,000 POPULATION BY RESIDENCE AND RACE, MICHIGAN, 1940 (By place of residence) RatesAper 1,000 population Residence White 5 Colored CrudezzAdjustedl:CrudezzAdjustedI Total 9.8 9.97 12.6 15.77 Urban places 9.2 10.17 11.4 14.66 Cities of 100,000 and over 8.2 9.90 11.3 14.75 Cities Of 10,000 to 100,000 10.1 10.42 11.8 14.16 Cities of 2,500 to 10,000 11.7 10.62 12.8 14.78 Rural 1008 9070 2502 24061 1. Adjusted to the age distribution of the population of the United States for 1940 by the direct method. 2. Source: Vital Statistics Rates Ln the United States, 1900-1940, Table 11, p. 202. Specific Causes 9; M ‘31 M Specific causes of death demonstrate consistently the superior health of the White over the colored pOpulation in Nflchigan. The non-whites have higher rates in 15 out of 25 important causes listed in Table 5.10. The two groups ap- pear to be about equal with respect to degenerative dis- eases. The death rate from.heart disease is 297 per 100,000 population for the white segment and 302 for the colored. Rural rates for heart disease vary more between Whites and .non-whites than do urban rates. The rural rate for the colored races is 438 and for the white 338, a difference of 100, Whereas the urban rates are 289 and 274 respectively, a difference of 15. The second most fatal degenerative disease, cancer, 124 .bsomm .mm .oa magma .HH pnwm .oema .mopmpm cepHgD on» yo moflpmwgmem Haew>umoszcm .conoomnmoqs was amused om menu mama no woman mopsm “opoz n.m m.aamm a.mmoa m.H n.neaa o.eme n.H a.meme m.Hme aeeeeo Has i... u-.. Mm «A m... «H «A m)... M Snefinea nun nun H. m.a mu. a. b.a m. n. mfipawsflsoa Hesammoenonoo .II' II. OOH all- “I” be I." ”I” m. “Ohm“ POHpHdom m.n mam m.a e.m o.m m. m.m n.m o.a emcee meaaoeee e.a a.oa e.a e.ea mhbm H.m a.ma mhmw e.H eeaeasom -nu nun. e.m a. bum. H.¢ m. mhm. s.n neeaom easaeeeeeeam e.a n.e a.n a.H H.@ 0.9 m.H H.@ a.» meeaeeepea ae eoaeeaeoae III- III. a .mapflnmqu .ecmnasaa m.m a.pH m.e e.a mum. o.e s.a mhm. n.e menses assessesm m.m ppma a.m o.H H.e m.e m.a e.e 0.0 eeeaeem men go geese m. n.m m.e e.H a.ma m.m m.H o.ma m.e eoaeeeapeeo III III. quapmman em minnow nu- nun H.a a. a.s m.oa a. m.e n.e geese eee we manoeaaao a. n.e H.s e.H e.sa e.oe m.H e.ea m.e napaeaeaeeea o.em sumo» a.ma m.e .Hhmm e.s 9.0 e.ee e.e maaaeaam a. mum TE m. [ohm as: a. mum: Hz: eefleaaehes Hesseeweoo n. mhmul. m.ea e. e.a H.9H m. e.s m.ea eeaeaem e.a .mn 0.0m a.H e.ee o.nm s.H s.oe n.nm unease assesseam o.m .n H.0m a. m.ea o.am m. a.mm e.em assesses ampeneao m.m m.ee e.sn e. e.aa o.an a. e.em m.nn meeeeaeea eaeaee>.nopes e.e a.mam e.am n.e 4.0aa a.mn e.e m.eaa a.mm maeeeeeaensa n.m a.mae H.ee m. n.em e.an e. m.en e.He .eaee> seep geese apnoeaeea o.n HumbH o.en m.e o.em m.ee m.H a.me H.Hm eueeeaueH s easeseeem e. a.me n.ne e.a e.ea e.an e.H e.se e.em easemenez e.” e.esa o.HHH m. s.ne n.ea m. a.mm H.oe snowmen Heaeeneeapea a.» e.esm o.ema e. n.oe m.eaa e. «.mm e.oma secede n.H a.mne e.enn H.a a.mmm o.eam .o.a s.aon e.eem ences on» no nonsense nwmhmpl neepo opaeaaumbmmml segue eases" .mmme segue opens conoaoo «ucuoaoo "wouoaoo "eaeem "neaeem "neapem names so ems-o H331 . queue " Hosea nofipuasmom ooo.ooa nominepsm I " HOZMQHMMM Q24 Hodm .o¢md .ZdeMOHE ZH deMQ ho mmmbdo 202200 HmB 20mm mmedm demQ 0H.n oanae 125 takes a much higher toll from.the white pOpulation. The rate for the white segment is 120.7 and for the non-white segment 88.2. In other words, cancer proves fatal to about one-third more white persons than colored. The same trend is found between the two groups in the urban population, but in the rural population it is reversed, the rural rate for non-whites being 279 and fcr whites 126. This may be partially explained in economic and educational terms. Organized health education is lacking or inadequate, in- comes are low, and facilities for treatment less available to colored people and particularly so in the rural areas.1 Intracranial lesions follow approximately the same pattern as does cancer for the colored and White pOpulation and for both urban and rural places. As might be expected, transmissible diseases take a greater toll from.the colored segment in prOportion to their numbers in the total pOpulation than from the White segment. For two important communicable diseases, tuberculosis and syphilis, this favorable position of the White group per- sists in both rural and urban areas. The death rate from tuberculosis for the colored races is 179 per 100,000 popu- lation, or more than 6 times that of the white race which is 28. The rate for this disease in urban areas is only 38 for white persons whereas colored people are dying almost four 1. Vance also found cancer rates for the total colored popu- lation lower than for the total white. He did not give any brsakdown by residence. (Vance, Rupert, 22, ctt, P0348. 126 times as fast, or at a rate of 170. In the rural areas the rate is almost ten times as high for the colored as for whites, the rates being 273 and 27 respectively. Syphilis is outstanding as a cause of death among negroes and proves fatal to the rural colored population of Michigan 24 times as often as it does to the White race.1 The rates are 12.7 and 304.7 respectively. The combined rate for three of the well-known transmissible diseases (tuberculosis, syphilis, and whooping cough) is 241 per 100,000 for the non-whites and only 39 for the whites. There seems to be some evidence pointing to the fact that both the educational and economic standards of migrants of the colored races are relatively lower than those of native residents, and much lower than those of native White residents. Further, it would not be unreasonable to suppose that these migrants have found come petition for survival in urban areas ruthless, and have ad- justed by moving into the rural areas of the State. Thus, they not only brought with them lower standards of economic and health care but have also tended to settle in areas- where facilities are lacking to provide adequate health in- 1. Kemp and Smith found that rural colored rates for syphilis were about 12 times as high as those for rural whites. Inasmuch as Michigan has done outstanding work in detect- ing and recording cases this is probably a part of the explanation of the differences in rates between the two States. See Kemp, Louise, and Smith, T. Lynn, 0 . th., Table III, p. 21. Vance referring to the excess ve rates for syphilis among the colored people makes this comment. ”Among the causes of death Which show a higher ratio of Negro to white races syphilis is outstanding. Against this must be placed the fact that syphilis is more likely to be recorded as a cause of death among Negroes." Vance, Rupert, 22o 21.-£0, Do 3480 127 centives and health services. White peOple are more likely to die from.accidents (both non-vehicle and automobile) them.the colored people. Because of our discrimination practices, these people are assigned tasks at hand labor out of proportion to their numbers and because of their low incomes do not have the money to purchase certain products and services which con- tribute to accidents in equal ratio with the whites, for example,--airplane rides, waxed hardwood floors, and auto- mobiles. Suicides in Michigan occur about 50 percent more often among whites than non-whites. The rates are 14.2 and 7.4 per 100,000 respectively. It is likely that supersti- tion in some instances and the strength of the religious mores in other cases deters the negro from.suicide. Fur- thermore racial discrimination has probably developed in- group feelings Which are reflected in a kind of family solidarity that offers security to the individual.. 0n the other hand, homicide accounts for 15 times as many deaths among Colored as White groups. It is debat- able perhaps whether the colored people are less inhibited or whether repressed frustrations ultimately result in overt aggressions against their own people as well as others. Only 1.9 per 100,000 white people died by homicide in 1940 While 28.6 per 100,000 among the colored population died from this cause. 128 *_m_ The infant death rates of White and colored people living in various pOpulation-size groups are presented in Table 5.11. Table 5.11 INFANT MORTALITY RATES BY RACE AND POPULATION-SIZE GROUPS, MICHIGAN, 1940 (By place of residence) Infant Deaths Per 1,000 Live Births ’Excess of POpulationeSize Colored Over Groups Total White Colored White Rate TOtal 4007 4000 5607 1607 Cities of 10,000 and over 39.3 38.2 54.6 16.4 100,000 or more 3705 3600 5204: 1604 25,000 to 100,000 43.2 42.7 63.7 21.0 10,000 to 25,000 39.2 38.2 72.1 33.9 Cities Under 10,000 and 42.7 42.4 76.9 34.5 Rural 2,500 to 10,000 45.0 45.3 25.0 ---- Rural 42.1 41.7 100.5 58.3 Note: Rates based on less than 20 infant deaths are under- scored. Source: Linder, Forrest E., and Grove, Robert D., Vital ‘ggatistics Rates ngghg_United States 1900-1945:- Table 28, p. 589. The infant death rates of non-whites are higher than those of whites in all1 pOpulation-size groups. Column four in Table 5.11 shows the excess of death rates for the colored segment over rates for the white segment. It is interesting to note that this excess between the two death rates tends to 1. Cities of 2,500 to 10,000 are disregarded in this state- ment, since their rates are based on only 3 infant deaths among the colored pOpulation. 129 increase as the size of the population group decreases.1 This pattern is set by the death rates for the colored p0p- ulation. The lowest rate among colored infants is 52.4 in cities of 100,000 and over, while the highest rate occurs in rural areas, namely, 100.0 per 1,000 live births. However, among white infants there is no such steady increase of death rates by size of population group as is found among the colored population. While the lowest rate, 36.0, does occur in the largest size cities, the highest rate, 45.3, does not occur in the rural areas but rather in the smallest size cities, namely 2,500 to 10,000. White infants in rural areas have lower death rates than do in-, tents in cities of 2,500 to 10,000, or in cities of 25,000 to 100,000 population. Another exception is that infants in cities of 10,000 to 25,000 have lower rates than the larger cities of 25,000 to 100,000 pOpulation. . The association of infant death rates among colored peOple with pOpulation-size gnaups is difficult to explain, since it does not resemble that of the rates of white in- fants in.Michigan. The progressive increase of infant death rates among the colored population.from large cities to rural areas is probably best explained, as suggested earlier, by the ec010gical forces which tend to distribute the colored population who migrate to Michhgan from Southern States ac- cording to competitive abilities. The data above seems to l. Vance observes, "For the Nation as a whole, figures in- dicate that the Larger the city, the lower the rate of infant deaths." (1940) Vance, Rupbert,‘gp. cit., p. 377. 130 substantiate the explanation that the colored people Who succeed in the industrial centers tend to raise their stand- ard of living until they gradually approach the level of living of whites, and in this way provide better medical care for their infants than the colored who settle in rural areas after finding life in the big city too difficult. Further, cities provide more facilities than rural areas for medical care, and may be less discriminatory in opening them.to colored peeple. The number of mothers Who die in childbirth per 1,000 live births in Michigan is presented in Table 5.12. This data is important only because it allows one to see how few mothers now die in childbirth. The rates are so low that they are almost meaningless for comparative purposes. For this reason, perhaps the actual number of deaths are more revealing. There were only 285 women Who died from various puerperal causes1 in the State in 1940. Of these 285 deaths, there were 266 among White women and only 19 among colored women. Two of the 19 deaths of the colored mothers occurred in rural areas and 17 in urban areas. When these deaths are expressed as rates, the rates among colored people are higher than those of White people in all of the various size cities and in rural areas. The reader is referred to Chapter 9 for trends in maternal mortality rates. 1. Deaths from puerperal causes are International List Numbers 140’150 0 131 TleD 5012 MATERNALIMORTALITY RATES BY RACE AND POPULATION-SIZE GROUPS, MICHIGAN, 1940 (By place of residence) 1 t— POpulationPSize Deaths of Mothersgper 1,000 Live Births1 Groups Total White Colored Total 209 208 406 Cities of 10,000 and over 3.2 3.1 4.6 100,000 and over 3.3 3.2 1.2 10,000 to 25,000 2.5 2.3 9.3 Cities under 10,000 a rural 2.5 2.5 5.1 2,500 to 10,000 3.4 3.5 --- Rural 203 202 704 I Deaths from.puerpera1 causes are International List Numbers 140-150. Note: Rates based on less than 20 deaths are underscored. Source: Linder, Forrest E., and Grove, Robert D., Vital Statistics Rates in the United States 1900-1940, Table 37’ p. 63 O CHAPTER VI MORTALITYl DIFFERENTIALS IN MICHIGAN COMPARED WITH THE NATI ON3 A fruitful approach in the further analysis of dif- ferential mortality is to compare data in.the United States with that of Michigana Death does not usually occur to people in a given segment of the pOpulation at random. Some death rates are reflections of social, economic, and psydhological conditions, and others are the results of a people grown old without replacement of children. In any event, the populations of the United States and Michigan have varying characteristics, and a systematic analysis of their similarities and differences may reveal aspects of health problems heretofore overlooked. In 1940, there were 52,183 deaths in Michigan and 1,417,269 deaths in the United States. How are these deaths distributed in relation to sex, age, residence, race, and cause of death? On a strictly numerical basis, the United States is losing relatively more lives than Michigan, the crude rates per 1,000 pOpulation being 10.8 and 9.9 respec- tively, a difference of .9. Thus, approximately 118,502 people died in the United States who would have survived the year had the Nation's death rate been the.same as that of 1. ,111 death rates for Michigan and the Nation in this Chapter are allocated to place of residence, except as indicated 0 2. The "Nation" is defined as all of the States, outlying territory, and possessions of the United States. 132 133 Michigan. Differences in the crude death rate become more meaningful, however, When the deaths from.which these rates are computed are related to such composition differences as age, sex, and race in the population of the two areas. One method for reducing the vital phenomena of two populations to more comparable units is to classify and analyze deaths in terms of segments of the pOpulation, such as comparing death rates of males in the United States with those of males in Michigan. Still another method is to relate data from two populations to a common standard so that the resulting rates are relative and hence comparable figures. Both meth- ods will be used in analyzing the data in this Chapter. Since age is so influential in determining the size of the death rate of any given segment of population, it is well to give some consideration to this problem before pro- ceeding to an analysis of particular classes. If age as a variable is held constant by relating the death rates of Michigan and the United States to a standard population1 the resulting age-adjusted death rates throw considerable light on the influence of age distribution on the death rates of the two areas. The crude and age-adjusted death rates are given below: Crude Adjusted Difference United States 10.8 9.5 1.3 Michigan 9.9 8.9 1.0 Difference .9 .6 1. The standard million used here is the population of the United States for 1930. 134 Both adjusted rates were lower than their respective crude rates, which indicates that both populations in 1940 tended to be older than that used as a standard. Although the crude rate for the United States is reduced more than that for Michigan, the age-adjusted rate for the State is still slightly lower than that for the Nation. Thus, if the two population groups had indentical age distributions and the same age-specific rates as occurred in 1940, Michigan's death rate would remain lower than that of the Nation. For purposes of comparison the actual population and the deaths of known ages in Michigan and the United States are present- ed in Table 6.1. It shows the prOportion of the total popu- lation and the total number of deaths in specific age groups expressed in ten thousands. Mortality Differentials By _A_gg The total age-specific death rates for Michigan and the United States are presented in Table 6.2. On the basis of the two series of death rates the following statements may be made: (1) The age groups at which minimum.and maximum.death rates occur are the same for Michigan and the United States. They are 5-14, and 85 years and over respectively. The min- imum.death rates are identical in both populations, namely 1.0, but the maximum.rate in Michigan is 245.1, an excess of 16.2 over the rate for the United States. (2) Age groups from 55 years upward have death rates which are two or more times the rates of all other specific age 135 Table 601 DEATHS AND POPULATION OF MICHIGAN AND THE UNITED STATES PER 10,000 OF TOTAL DEATHS AND POPULATION, BY AGE GROUPS, 1940 Number per 10,000 Age Michigan United States Deaths Population Deaths Population Under 1 773 159 784 153 1 - 4 146 661 174 '647 5 - 14 165 1,680 164 1,704 15 - 24 313 1,780 346 1,817 25 - 34 404 1,615 461 ' 1,621 35 - 44 672 1,453 674 1,392 45 - 54 1,206 1,245 1,162 1,178 55 - 64 1,664 778 1,663 803 65 - 74 2,125 437 2,162 484 75 - 84 1,913 167 1,803 172 85 Over 619 25 607 29 Total 10,000 10,000 10,000 10,000 Unknown age 36 1,702 Source: Computed from.Vita1 Statistics of the United States, 1940, Volume II, Residence, TabIE 9, p. 7 , TaBIe 9’ P. 190. Table 602 ACE-SPECIFIC DEATH RATES PER 1,000 POPULATION, MICHIGAN AND ' THE UNITED STATES, 1940 Rate per 1,000 Population Age Michigan United States Difference All ages1 9.9 10.8 excess .9 Under 1 4801 5409 910083 608 1 ‘ 4 202 209 BXOOSS e7 5 - 14 1.0 1.0 ~- 15 - 24 1.7 2.0 excess .3 25 - 34 2.5 3.1 excess .6 35 - 44 4.6 5.2 excess .6 45 - 54 9.6 10.6 excess 1.0 55 ' 64 2102 2203 excess 101 65 - 74 48.2 excess 48.0 .2 75 - 84 113.6 excess 112.6 1.0 85 Over 245.1 excess 228.9 16.2 Includes deaths of unknown ages. Source: Tables 4.2 and 6.3. 135 . 5H .n .a 3995 .oosmcamom .HH megao> .oema .mmpapm Sopfiqp on» go moflpmapspm HASH>.aone sopsmaoo “mongom .mows «Songs: 90 unused mousaonH H a.mam a.mnm 0.4mm m.amm a.mem o.¢nm a.mam 0.94m a.mmm nope mm a.moa o.mma n.0aa m.ooa s.naa s.soa m.eoa m.Hma a.maa em 3 as n.44 m.Ho a.mm «.mn n.04 s.me o.ae n.4n o.me 4 «s n no 0.0H m.on m.em a.ma H.Hm a.ma a.ma «.0m a.mm do u an m.o a.4H A.HH m.s H.0a m.m o.m a.ma 0.0a «m u we s.4 «.0 n.m n.e H.n 4.4 m.¢ o.m m.n «e u on s.m e.» a.» m.m a.» a.» s.m a.» a.» 4» u mm m.a m.m o.m m.a e.m H.m m.H n.m o.m em . ma a. m.H H.H a. H.H o.H o.a m.a o.H ea s n a.m‘ o.» m.m m.m m.» m.m . s.m a.» m.m e - a m.me 4.40 n.sm m.nw a.mm s.mm s.se a.ao o.¢n a nouns «.0H . a.ma n.aa s.m m.oH m.a n.a a.ma m.oa Hmomd add causes mass dance 38% odes ages cameos .32 .389 .3 zqmmp 14 sampm gases oeoa .mmadam ameHzp .mozmnHmmm n24 .Hmm .ma4 wm cheaqpmom ooo.a mam wages maemq OHSHommm «.10 0.3.95 ” 137 groups for both Michigan and the United States, except for the group under one year of age. (3) The age groups WhiCh take a toll of less than 10 deaths per 1,000 pOpulation are 1-54 for Michigan.and 1-44 for the United States. (4) The United States has higher age-specific death rates than Michigan for all age groups except 5-14 and 55 and over. Mortality Differentials by §_§_x_ Death rates by sex for the United States are present- ed in Table 6.3 and for Michigan in Table 4.2. Both the crude and age-adjusted death rates of males and females given in Table 6.4 are higher for the United States than for Michigan. Age-adjusted rates are slightly lower, and their Table 604 SEx-SPECIEIC DEATH RATES, PER 1,000 POPULATION, NICH13AN AND THE UNITED STATES, 1940 Area male Female Difference (male and Female) Crude Ad3.1 Crude Adj.l Crude AdJ.l United States 12.0 12.1 9.5 9.4 2.5 2.7 MiGhigan 1100 1100 809 900 201 200 Difference 1.0 1.1 .6 .4 1 Adjusted to the age distribution of the United States pOpulation in 1940 by the direct method. Source: Computed from Vital Statistics of the United States, 1940, Volume II, ReSIdence, Table 9, pp. I78, differences are also slightly less. Differences, however, in both the crude and age-adjusted rates for both sexes never 138 exceed 1.1 in the two areas. This indicates a close similar- ity of age distribution within the sexes in the two popula- tion groups. Differences in crude rates between males and females in M1Chigan and those in the United States are 2.1 and 2.5 respectively. Females in both areas possess a no- ticeable survival advantage1 over males. MOreover, this ad- vantage is slightly greater for females in the United States than for those in Michigan. Sex-mortality differentials by age are given in Tables 6.5 and 6.6. The salient conclusions are as follows: 1. Males have higher death rates than females at all ages for both pOpulations, (Table 6.6, columns one and two). The differences, although small, are Similar for both areas, the only exception being in age group 85 years and over*where the rate of males exceeds that of females by 25.7 in the United States. Little weight can be given to this exception, since declarations of age for advanced age groups are subject to great error.2 1. Thompson, in discussing the lower death rates among females, says, "Whatever the reason, they survive in greater numr bers, and, as a result, a comparison of death rates of dif- ferent populations without a knowledge of their sex con- stitution.may, and often does, result in some misconception of their actual mortality conditions." (Thompson, Warren S. ‘gp.‘git., p. 227) Elsewhere, Thompson suggests that the favorable female death rate is probably due to differences in occupational roles. (Thompson, Warren S., and Whelpton, P.K.,‘gp.‘gi§., p. 181, 184) See, also, Landis, Paul E., ‘gp.‘£;t., p. 214, and Appendix IV'of this dissertation. 2. See Densen, Paul M. "Family Studies in the Eastern Health District: II The Accuracy of Statements of Age on Census Records," Th2 American Journal 2§_Hygiene, XXIII, No. 1 Sec. A, 1940,pp. I-38. I.“ Table 6.5 139 SPECIFIC DEATH RATES PER 1,000 POPULATION, MICHIGAN AND THE UNITED STATES, 1940, BY AGE AND SEX Age Michigan United States Groups Male Female Male Female A11 Agesl 11.0 8.9 12.0 9.5 Under 1 5501 4007 6109 4707 1 ' 4 205 200 301 207 5 ~ 14 101 08 102 100 15 - 24 201 105 205 108 25 ' 34 207 202 303 207 35 ‘ 44 500 401 509 405 45 - 54 11.0 7.9 12.5 8.6 55 - 64 2402 . 1709 2602 1801 65 ‘ 74 5305 4208 5402 4109 75 ‘ 84 12202 10502 12105 10405 85 Over 24900 24107 24308 21801 1 Source: Includes deaths of unknown ages. Tables 402 and 6030 Table 606 DIFFERENCES IN AGE-SPECIFIC DEATH RATES, MICHIGAN AND THE UNITED STATES, BY SEx, 1940 Differences Age Male minus Female: Male : Female __ Groups M1 0 U030 3 Mldhe and U030 _i_M10h0 and U080 ,A11 Ages 2.1 2.5 1.0 U.S. excess 6.7 U.S. excess Under 1 14.4 14.2 6.8 U.S. excess 7.0 U.S. excess 1 - 4 .3 .4 .8 U.S. excess .7 U.S. excess 5 ' 14 03 02 01 U080 excess 02 U080 EXCESS 15 - 24 .6 .5 .2 U.S. excess .3 U.S. excess 25 - 34 .5 .6 .6 U.S. excess .5 U.S. excess 35 - 44 .9 1.4 .9 U.S. excess .4 U.S. excess 45 - 54 3.1 3.9 1.5 U.S. excess .7 U.S. excess 55 - 64 6.3 8.1 2.0 U.S. excess .2 U.S. excess 65 - 74 10.7 12.3 .7 U.S. excess .9 Mich. excess 75 - 84 17.0 17.0 .7 Mich. excess .7 Mich. excess 85 Over 7.3 25.7 5.2 Mich. excess 23.6 Mich. excess Source: Table 6.5. 140 2. The minimum age specific death rate for both sexes oc- curs in the age group 5-14 in both Michigan.and the United States, and the maximwm rates in both areas occur at ages 85 and over. The lowest rate in age group 5-14 is .8 for females in Michigan, and the highest rate is 1.2 for males in the United States. 3. In general the excess of male rates over those for fe- males steadily increases from age group 5-14 in the United States and from.age group 25-34 in Michigan through age group 75-84 in both populations. 4. Sex-specific death rates are higher for the United States than for Michigan in all age groups except 65 and over for females and age 75 and over for males. / Rural-Upban Mortality gigferentials Each year approximately 11 persons per 1,000 people living in rural areas of Midhigan die in comparison with 9.8 in the rural areas of the Nation. (See Table 6.9) These crude death rates indicate that the toll of human lives ex- acted from.MiChigan's rural population is relatively greater than that of the rural population of the United States as a whole. Table 6.7 shows that Michigan's rural population has lower proportions of its total number in the ages from 1-34 (than does the Nation and a marked excess over the rural seg- ment of the United States from.the age of 35 years upward. It is also evident from.Tab1e 6.7 that proportionately more rural deaths occur at age 55 or over in MiChigan.than in x . v I s. I I —0’.;’ It; 1211 £— 141 Table 6 0 7 DEATHS AND POPULATION PER l0,000 MICHIGAN AND THE UNITED STATES, 1940, BY AGE AND RESIDENCE Deaths_per 10,000l Populationgper 10,000 .0 O. O. .0 O. Age United United Groups Michigan States Michigan States Rural Urban Rural Urban: Rural Urban Rural Urban Under 1 720 805 974 659 175 151 181 132 1 - 4 153 141 242 130 763 608 786 541 5 - 14 166 165 205 138 1,872 1,579 2,007 1,470 15 - 24 285 330 404 308 1,716 1,813 1,864 1,780 25 - 34 303 465 454 466 1,426 1,713 1,455 1,749 35 - 44 464 799 578 736 1,224 1,572 1,212 1,531 45 - 54 893 1,396 947 1,302 1,141 1,299 1,046 1,280 55 - 64 1,575 1,718 1,444 1,807 872 729 760 836 65 - 74 2,363 1,981 2,104 2,200 553 377 483 485 75 - 84 2,296 1,679 1,932 1,719 220 139 176 169 85 Over 782 521 716 535 38 20 3O 27 Total 10,000 10,000 10,000 10,000 10,000 10,000 10,000 10,000 1 Includes only deaths of known age. Source: Residence, Table 9. 1940, Population, Fourth Series, Midhigan, Table. l. Linder, Forrest E., and Grove, Robert D., Vital Statistic Rates 2;,thg_United States, 1900-1940, TabIe IV, p. 936. \' II. I .‘ 1 ‘III‘ 1.1.! 142 the United States, while larger proportions of rural deaths in the Nation occur under age 55. The rural age-adjusted death rate is higher in Michigan.than in the Nation. Com! parison of the age-specific rates provides one explanation (Tables 4.2 and 6.3). InasmuCh as rural Michigan.has lower age-Specific rates than the Nation for those under 54 years but higher rates for those over, it would appear that the excessive death rates from ages 55 upward partially account for the high rural rate in the State. The ratios of male to female death rates for the age groups in rural and urban segments of Michigan and the United States are given in Table 6.8. These date Show that sex-mor- tality differentials are greater in rural than in urban Mich- Table 6.8 SEIHMORTALITY RATIOS, MICHIGAN AND THE UNITED STATES, 1940, BY AGE AND RESIDENCE Male Death Rate 'Female DeathREte‘X 100 Age GrOuP' Mighigan United States Rural Urban Rural Urban Total Crude 129 120 125 126 0 - l 131 138 130 129 l - 4 120 110 114 120 5 ~14 138 138 122 133 15-24 140 133 126 122 25-34 142 127 118 126 35-44 134 116 119 136 45-54 145 137 135 153 55-64 134 137 134 154 65-74 120 130 122 138 75-84 113 119 113 121 85-up 104 101 112 111 Source: Based on death rates given in Tables 4.2 and 6.3. 143 igan for a majority of the age groups. In the United States the sex-mortality differential is greater in urban than in rural areas for all ages except those 85 years and over and those l5-24 years of age.1 In contrast to Michigan's urban population, the urban people of the United States have higher crude and age-adjust- ed death rates (Table 6.9). Their crude rates are 11.5 and 9.4, respectively, a difference of 2.1. The adjusted rates Table 6.9 DEATHS PER 1,000 POPULATION, MICHIGAN AND THE UNITED STATES, 1940, BY RESIDENCE -_ -:‘ - I Deaths per 1,000 Population Residence United States Michigan Crude Adjustedl Crude Adjustedl Urban TOtal 1105 1104 904 1004 Cities: 100,000 over 11.3 ll.4 8.5 10.5 Cities: 10,000-100,000 11.4 11.2 10.1 10.5 Cities: 2,500-10,000 12.4 11.8 11.7 10.8 Rural TOtal 908 908 1100 1002 1 Adjusted to the age distribution of the pOpulation of the United States in 1940 by the direct method. Source: Crude rates except total urban are from Linder, Forrest E., and Grove, Robert D., Vital Statistics Rates in.the United States, 1900-1940, Table 11, pp. 198, 202. for the same segments are respectively 11.4 and 10.4, a dif- ference of only 1.0. Adjustments for age lovers the urban rate .1 in the United States but increases it 1.0 in Michigan. The two differences provide a rough measure of the influence of age on.the crude rates, indicating that the urban crude 1. See also Appendix IV, Table 4. 1‘ v‘ i’ .:i e- :0 {W A! 0-6 1y I in in J . I I, . id 19: , :02635 '03:; ‘9 {1 ()'J :13 "V - 11) ‘9 ’1 v [‘3 ‘72. a 1 ’Li 0 .15— ,1 . (;‘ 144 rate in the United States is higher than the corresponding rate in Michigan because larger prOportions of the population are in the older age groups (See Table 6.7). The crude death rates of cities in both Michigan and the United States vary inversely with their size, the highest rates being found in small cities. As was indicated earlier this tendency for small cities to have high rates is becoming a matter of growing concern.1 It is obviously not a pattern of certain isolated areas but is characteristic of the United States as a Whole. Various 5126 cities have age-adjusted death rates slightly in excess of the rural areas in both Midhigan and the United States. Large cities of 100,000 and over in.much- igan have rates only .1 in excess of the rural rate, While in the United States the corresponding rate is 1.6 in excess of the adjusted rural rate. Small cities of 2,500-10,000 popu- lation have the highest adjusted rate in both the State and the Nation. The lowest rate in the United States is found in cities of 10,000-25,000, whereas in Michigan.it is found in the large cities of 100,000 population or more. The mortality data by residence given above may be summarized as follows: (1) The urban age-adjusted death rate 1. Mangus points this out, also, when he says, "It is of considerable interest that the death rate was highest among rural residents and among residents of small towns and cities of less than 10,000 population. (mangus,.A. R., ”Health and Human Resources in Rural Ohio," Mimeo- graph Bulletin No. 176, Department of Rural Economics and Rural Sociology, Ohio State University and Ohio Agricultural Experiment Station, Columbus, Ohio, 1944, p. 4. See, also, Footnote, p. 68, Chapter IV. penni ~‘h “am ”1’“ a .‘t. in I. \ \ H ,r" -L a 3 _a .5 OF fro p—A- 145 is lower for Michigan than for the United States, but the rural rate for the State is higher. (2) Both the crude and the adjusted death rates for various size cities are lower for Michigan than for the United States. (3) In both popu- lations, crude death rates vary inversely with the size of the city. (4) The rural segment of Michigan's population has a.higher crude death rate than the corresponding population group of the Nation, a reflection of the excess of older people among the rural pOpulation of Michigan. Mortality Qigferentials _Igy 3393 According to data presented in Table 6.10, the crude death rates of the white and the colored segments of Michigan compare favorably with the corresponding classes of the Unit- ed States. In Michigan 12.6 colored persons per 1,000 die annually, While the corresponding rate is 13.8 for the Unit- ed States. This differential can be partially accounted for by the varying age distribution and partially by the high standard of living of Michigan's predominantly urban colored population.l When the crude death_rates for non-Whites are adjusted for age, the rates are 16.2 for the United States and 15.8 for la NUchigan's colored population is relatively small, com- prising only 216,465, of Whom.9l.4 percent are classed as urban compared with 47.9 percent so classed for the Nation as a.whole. Less than 4 percent of the colored in.Michigan are other than Negro. Of the State's total urban non-White pOpulation, 75 percent live in the city of Detroit. Thus, any comparison of the total colored in Michigan.with those of the Nation is essentially a comparison of an urban group with one Which is more than half rural. 9' 0-HT 3 tea ' I l ‘9'9ac i '“1' "..' u.- ‘ 9' .. 41:48,; ha HM? ,. :h:‘.‘-. 33:5 the T5233 5'- ‘:ac I *3- 31-‘3 0 Pitch: : 3401‘. E? c “in 9h ”A“. WW9: tre 146 Michigan, the difference being reduced two-thirds, or from 102 to 040 Table 6010 CRUDE AND ACE-ADJUSTED DEATH RATES PER 1,000 POPULATION, MICHIGAN AND THE UNITED STATES, 1940, BY RACE Area White Colored '_Crude Adjustedfl) Crude Adjgsted[1) United States 10.4 10.2 13.8 16.2 Michigan 9.8 10.0 12.6 15.8 Difference e6 02 102 04 (1) Adjusted to the United States population in 1940 by the direct method. Source: Crude rates from.Linder, Forrest E., and Grove, Robert D., Vital Statistics Rates in the United States 1900-1940,?{5‘1‘3 11, pp. 198,202. ——" _- Tables 6.11 and 5.5 show clearly that the colored population of the United States has relatively greater num- bers than that of Michigan in the younger ages from 1-24 years and in the older ages from 55 to the end of the life span. In the latter age groups risks of mortality are great. On the other hand, Michigan's colored population has a larger prOportion of its total number in the middle age groups from 25-55 where the risks of mortality are considerably less. It is to be expected, then, that the relative number of colored deaths in the United States and Michigan tend to be in excess at ages Where the respective populations are concentrated. Hence colored deaths in the United States ex- ceed those in Michigan from ages 1-24, and from age 55 up- Wards I! act'" .I : ! June“. I u D‘.‘ h 'u‘d. . __ Table 6011 DEATHS AND POPULATION OF THE UNITED STATES PER 10,000 OF TOTAL DEATHS AND POPULATION, BY AGE AND RACE, 1940 Number_per 10,000 Age Deaths Population White Colored White Colored Under 1 727 1,163 150 180 1 - 4 158 280 630 795 5 - 14 155 228 1,665 2,043 15 - 24 292 707 1,802 1,945 25 - 34 387 953 1,617 1,656 35 - 44 585 1,261 1,392 1,398 45 - 54 1,095 1,601 1,202 966 55 - 64 1,690 1,482 833 542 65 - 74 2,288 1,325 499 354 75 - 84 1,975 663 181 94 85 - Up 648 337 29 27 Total 10,000 10,000 10,000 10,000 Source: Deaths computed from.Vital Statistics of the United States, 1940, Volume II, Residence, Table 9, Po 17g. I! (‘I (I 1 . I 9:4 M "u no: wit: u!!- o I “w“. ‘1 a .11-" . . - H l 0-: ’1‘ by vn‘od tat 1 y u 1““ ‘1 l W... 5‘ if“ s O I s.’ ‘ 't.‘ has ‘4, ; .9 . eff-g. v ‘N ‘. .w M ‘5 :1: C1 7 “ 148 The white people in Michigan die at a slightly lower rate than those in the Nation. (Tables 6.11 and 6.12) The respective crude death rates are 9.8 and 10.4 per 1,000 population, a difference of .6. The age-adjusted rate for the United States is 10.2, While the rate for Michigan is 10.0, a difference of only .2. Thus, the white segments of the State and the Nation have approximately equal death rates, although the State rate is slightly lower than that of the Nation. An examination of Tables 6.11 and 5.3 reveals that the age distributions of the two white populations are quite similar. White people in Michigan, however, are more numerous in age groups under 15, and from ages 35-54, while white individuals in the United States are found more often in age groups 15-34, and in all groups from 55 years upward. Sex-specific crude death rates for the white and non-White populations are slightly lower in Michigan.than.in the United States (Table 6.13). This is more apparent when the age-sex-specific rates are examined (Compare Tables 5.2 and 6.14). Colored males in Michigan have lower rates, for example, at all ages except 75-84 than the corresponding group in the Nation. Differences between Michigan and the United States for white females classified by age are less pronounced. White females in both populations have identical age-adjusted death rates, namely, 8.8 per 1,000. Among whites, male death rates exceed the female at all ages, and this is generally true for the colored group (Table 6.15). In Michigan, however, the rate for colored l._ :2. I" | :3" c- I - 149 Tab1e 6 e 12 DEATHS PER 10,000 OF TOTAL DEATHS, MICHIGAN, 1940, BY AGE AND RACE White Deaths Colored Deaths Per 10,000 Per 10,000 of of Age Number Known Age Number Known Age Under 1 3,799 769 233 854 1 - 4 708 143 52 191 5 - 14 811 164 49 180 15 - 24 1,454 294 177 649 25 - 34 1,819 368 286 1,048 35 - 44 3,014 610 491 1,800 45 - 54 5,736 1,161 553 2,027 55 - 64 8,294 1,678 383 1,404 65 - 74 10,791 2,184 293 1,074 75 - 84 9,807 1,984 167 612 85 - Over 3,186 645 44 161 Total 49,419 10,000 2,728 10,000 Unknown 32 4 Source: Deaths computed from Vital Statistics of the United States, 1940, Vqume II, ResIdaHEe, Table 9, P0190. {:éivd a” 6‘44 P' a av- 3U"! "kl 150 Table 6 013 CRUDE AND AGE-ADJUSTED DEATH RATES PER 1,000 POPULATION, MICHIGAN AND THE UNITED STATES, 1940, BY RACE AND SEX (By place of occurrence) White : Colored : , z : Area Male : Female : Male : Female :(1) : : (lifii: :(1) : : (1) Crude :Adj.: Crude: Adj. :Crude :Adj.: Crude : Adj. United States 11.6 11.6 9.2 8.8 15.1 16.9 12.6 15.0 Aachigen 10.8 11.1 8.7 8.8 13.7 17.0 11.6 (14.6 Diff. 0.8 0.5 0.5 0.0 1.4 0.1 1.0 0.4 44 (1) .Adjusted to the United States pOpulation in 1940 by the direct method. Source: Crude rates are from.Linder, Forrest E., and Grove, I... 55”.], I. -i.‘ I :II AGE-ESPECIFIC DEATH RATES PER 1 BY RACE AND SM, 1940 Table 6.14 151 000 POPULATION, UNITED STATES, (By place of occurrence) Age White Colored Gnu” Male Female Male Female , A11 A858 1106 9.2 15.1 12.6 Under 1 56.7 43.6 101.2 77.4 1 ' 4 2.8 2.4 5.3 4.4 5 - 14 1.1 .8 1.6 1.4 15 - 24 200 104 5.0 5.0 25 - 34 2.8 2.2 8.5 7.4 35 ' 44 5.1 3.7 13.2 11.7 45 - 54 11.4 7.5 24.5 21.1 55 - 64 25.2 16.8 39.5 35.7 65 - 74 54.0 41.5 56.5 46.3 75 ' 84 122.2 105.6 109.7 84.7 85 - Up 249.3 224.7 193.2 156.2 Source:Tfinder, ForrestE., and Grove, Robert D., Vital Statistics Rates.;g_the United States, 1900-I940, Tibia 23, p. 434. Table 6.15 DIFFERENCES BETWEEN MALE AND FEMALE DEATH.RATES, MICHIGAN AND THE UNITED STATES, 1940, BY AGE AND RACE Excess of Male Over Female Death Rates Age White Colored Groups Michigan United Michigan United States States A11 A868 2.1 2.4 2.1 2.5 Under 1 14.4 13.1 11.9 23.8 1 - 4 .3 .4 .3 .9 5 ’ 14 .3 .3 0(a) .2 15 ' 24 .6 .6 ' 1.1(b) .0(&) 25 ' 34 .5 .6 .8 1.1 35 ' 44 .9 1.4 - .5‘b) 1.5 45 " 54 2.9 3.9 4.6 3.4 55 - 64 6.3 8.4 1.4 5.8 65 “ 74 10.4 12.5 17.9 10.2 75 ' 84 15.5 16.6 3603 2500 85 - Up 19.2 24.6 52.6 37.0 a Zero indicates no difference. (b) Source: Minus sign indicates excess of female rate. Based on death rates in Tables 5.5 and 6.14. 152 females exceeds that for males in two age groups, and in one, the rates are identical. Among non-whites in the United States, the sexes have equal rates, also, at age 15-24. Is the sex-mortality differential greater for the colored or the White pOpulation? How do they compare in Michigan and the United States? If the age-adjusted death rate for each sex is expressed as a ratio of the male to the female rate, it is possible to answer these questions. The ratios are given below: White Colored United States 132 113 iichigan 126 116 Thus, the white sex-mortality differential is greater than the colored. Moreover, it is higher in the United States than in Michigan. However, the sex-mortality differential for non-whites is Slightly in excess in Michigan. Rural mortality differentials by race favor the United States, while urban differentials favor Michigan. The race-specific death rates by residence are given in Table 6.16. The significance of these data.may be summar- ized as follows: 1. The crude and age-adjusted death rates of rural Mich- igan for both colored and White people are higher than cor- responding rates for the Nation.1 2. Urban Michigan has lower crude and age-adjusted death rates than the United States for both White and colored 1. See Chapter V, p. 119, Footnote 1. "“"l " sul v" ' nae-cg - . hi." 1...-.. ”4.... mi, h “a, ‘4 I‘fi h—fi (’ 153 groups. 3. Mortality differentials between the United States and Michigan by race and residence are greater for the colored in both rural and urban areas than for the corresponding white segments. Table 6.16 CRUDE AND ACE-ADJUSTED DEATH RATES PER 1,000 POPULATION, MICHIGAN AND THE UNITED STATES, 1940, BY RACE AND RESIDENCE White Colored Area Rygal Urban ' Rural Urban (1) (I I (1) (17 Crude Adj. Crude Adj. Crude Adj. Crude Adj. United States 9.5 9.2 11.9 10.8 12.2 14.4 15.6. 18.3 Michigan 10.8 9.7 9.2 10.2 25.2 24.6 . 11.4 14.7 Difference 1.3\ 0.5 2.7 0.6 13.0 10.2 4.2 3.6 (1) Adjusted to the 1940 United States population by the direct method. Source: Rural crude rates from Linder, Forrest E., and.Grove, 1900-1940,'TEUISHIIT—E§T_I98,202; Urban crude ra33§__- computed from Vital Statistics of the United States, 1940, Volume TIT—RESidence;—TEbI3“9:—ppT—I78{‘I9UT’ Principal Causes 2:.QQEEQ IMortality differentials tend to favor the people of Midhigan when total death rates from.classified causes are compared with those of the Nation. This comparison is pre- sented in Table 6.17. The Table includes all of the specific causes re-classified into eighteen major groups of related causes. These are the 18 major subdivisions of the 1938 ‘F'Tv‘a: N“. we :'-a “0:1. 154 International List of Causes of Death.1 Part I of the Table shows causes predominant in Michigan and Part II those pre- dominant in the United States. All are ranked from.high to low in terms of the death rate for Michigan. From.the data in this Table, the following general observations may be made: 1. iMichigan has lower death rates than the Nation for fourteen of the eighteen classified causes (Part II, Table 6.17). 2. Michigan has lower rates than the Nation for classi- fied causes which take the lives of infants and children. They include the infectious and parasitic diseases, Michigan having a rate of 59.4 per 100,000 for this group compared with 91.0 for the Nation. It will be recalled that in.the early history of Michigan, illness and death from infectious and parasitic diseases were major problems facing not only children but all residents. Crude death rates from diseases peculiar to the first year of life are only slightly less in Michigan than.in the United States, 38.5 compared with 39.2. 3. Michigan has lower rates than the Nation for the pre- ventable causes of death. These include violent or acciden- tal deaths in addition to the infectious and parasitic dis- eases, and diseases of pregnancy and childbirth. 4. Causes of death which are predominant in Michigan tend to be those Which are most difficult to control. In this 1. See Appendix III for complete International List of titles included in each major subdivision. 155 Table 6.17 MORTALITY RATES PER 100,000 POPULATION FOR 18 CLASSIFIED CAUSES1 OF DEATH IN MIC:-:IC-AN AND THE UNITED STATES, 1940 (By place of occurrence) W Causes Rates per 100,000 Population Part I . Causes Predominant in Michigan United Excess of Michigan. (Ranked by States Michigan Mishiaan Rate) Rates Diseases of circulatory system. VII 315.8 314.4 1.4 Rheumatism, diseases of nutrition III 36.5 36.1 .4 Congenital malformations XIV 13.9 10.0 3.9 Diseases of blood and blood forming organs IV 8.5 7.4 1.1 Part II Causes Predominant in United Michigan United Excess of States. (Ranked by Michigan States Michigan .gates) 1_ Rates Cancer and tumors II 124.8 125.2 .4 Diseases of nervous system ‘VI 101.3 103.6 2.3 Violent or accidental deaths XVII 91.8 94.3 2.5 Diseases of genito-urinary system X 70.1 95.4 25.3 Infectious and parasitic dis- eases, including tuberculosis I 59.4 91.0 31.6 Diseases of respiratory system VIII . 58.8 66.1 7.3 Diseases of digestive system IX 53.3 59.2 5.9 Diseases peculiar to first year of life XV 38.5 39.2 .7 Senility XVI 6.4 7.7 1.3 Diseases of pregnancy, etc., XI 5.5 6.7 1.2 Ill-defined and unknown causes XVIII 4.1 16.0 11.9 Chronic poisoning and intoxication V 1.3 2.1 .8 Diseases of bones and organs of movement XIII .7 .9 .2 Diseases of skin and cellular tissue XII .5 1.1 .6 1 The source cited gives the specific causes and the Inter- nation List of rubrics included in each of the 18 classes of causes. See Appendix III for complete International List titles. NOtRCman numerals refer to rubrics in the International List Of 1958. Source: Linder, Forrest E., and Grbve, Robert D., Vital Sta- h United States 1900-1940 (U.S. Bureau tistics Rates Igatne “n- A 9_AR - , , 156 group, diseases of the circulatory system, which includes heart aiLments, ranks first having a crude rate of 315.8 as compared with 314.4 for the Nation. 5. The three leading causes of death for Michigan and the United States are identical, namely, diseases of the circulatory system, cancer and tumors, and diseases of the nervous system. Michigan has lower rates than the Nation for the latter two of these. Table 6.18 presents twenty-five specific causes of death by race and residence for Michigan and the United States. An inspection of this Table reveals that the urban segment of Michigan compares favorably with the Nation, while the rural segment compares unfavorably with the Nation. This is true, in general, for both white and colored pSOple. Both racial segments of Michigan's urban population have lower death rates in twenty out of the twenty-five principal causes of death (Table 6.19). The exceptions for the two population groups are not identical, however. Urban death rates for whites exceed the Nation for motor vehicle acci- dents, congenital malformation, puerperal causes, exOphthal- mic goiter, and scarlet fever. Among the urban colored, the diseases which take a greater toll in Michigan than in The United States are tuberculosis,l suicide, appendicitis, ex- ophthalmic goiter, and cerebrospinal meningitis. (Rates for 1. See Cha ter V, Footnote 1 p. 90, for a discussion of the hi rate for tubercuiosis among the urban colored of Michigan. - fldlznl‘. .“fi\I c lllc A toucllvlfllfieli .. .I 11!! ntlu- I F. 4+lul I HufFl‘lrh rl HIV duh Jul .MF HI ‘4'; 4. uu‘dl 1.)- cenh 1...,- IVs..- I--.A ’II! It”. ‘u IRI- _ In; Gui ”I’ll! .I'. I ‘:‘,fl'. ‘.‘§ - Px 1‘ i.‘."vd‘b‘hr~ - I‘Vnnllw 157 .omm.oam .mm .OH magma .HH chem .oemH .mmpwpm emcee: .o¢ma .noapeasmom mmwwnmadqm no dommm map mo moflpmapapm prfi> “condom .hdmo whack 039 nocsb m .nnpwoc om comp mama no women one money conoomnouqb a o.aoma Mhmeaa e.mOHH :buomm s.mama m.samm c.0eo «.mmoa momsso Had. m.H mu. s. sun. n.m ... m.a .Nv manoepngfio d. m. m. e. p. ... o. HDI. mapawnHQQE Hosammonpohoo o. nun. e. s. a. nun. e. o.H empoe poaasom H.e o.m H.H m. m.s n.m n.m m.H eweoo meagooea e.ee mnbm a.» H.m a.mm s.oa n.» e.H oefioaaom o.m .m m.n H.e m.H nun. e.m m.m nephew oeaasepngowm «.ma Hum. e.m o.» e.ma m.n e.m e.» «.0pm .mspfinmpso .soennsfin m.ea mhm. n.m 0.0 a.ma b.pH 0.0 m.e mousse Hesmmuosm o.o H.o o.m m.o o.e puma m.m e.n nosaopm we» go sweep o.eH mnmu o.m m.m m.m n.m o.e m.a seesaw m.m H.e e.HH n.0a o.» .vnw m.m H.s wopaa me» go mfimonnnflo m.na 5.5H m.oa e.oa m.» n.m m.m H.s quHOflesomma e.no Hhmm e.oa o.s a.me sumo» m.m s.ma ahasemsm m.o mum. m.oa m.na n.n n. m.oa ¢.eH eofipssnoeaas HapHsomnoo e.o o.b H.5H H.na o.n mhmr. n.na m.oa oefiossm e.mm m.ae m.se o.nm a.mn H.mn m.nm 0.0m mnpnfip magnesmnm a.mm «.ma m.mn 0.5m H.HH mnmm. m.om H.om mspfiaaoe uopmpsdn n.0m e.eH 0.0m o.an a.ma «.00 e.om 0.5» Aoaoanop saves” manoefiood n.0aa $.0ba a.mn a.mn >.nm a.mbm a.dn «.sm namoasonopsa e.mn n.0m «.5e o.en o.am «.maa o.oe H.0e .oaoflnopunozv apnoeaooa m.nna o.mm a.mm m.oe o.saa H.0ea o.os a.mn sueoeauefl was manoesoqm o.HeH ¢.os a.ms e.am «.moa a.mn H.es n.no nflpflnnmmz m.sma p.90 o.mm a.me e.ea e.osa a.mm a.daa damage guacamop no msoamoa Hwaqsnownan a.moa n.0e m.nea m.eaa a.mn o.sem b.5m o.oma noonso m.ean s.mmm n.5nn o.eem n.sma a.mne o.n¢m a.snn peace on» go nonsense .m.p .eoaa hm,p shoes umnOHbo opus; nfimoa no cause Hensmu nowadasmom ooowooa Mom mapdm .mmedam QfiBHZD HEB 924 ZfiUHmUHS .mH4MQ mo mfimpdo 202300 fiomm Hmdem HBHAdBmoE “ooqocfimon no oomam 0p copmooaae mnpscnv mozmonmm axe scam Mm .oeaa md.o Danna 158 the last four causes listed are computed on less than 20 deaths.) Two of these diseases, goiter and cerebrospinal meningitis, might almost be termed indigenous to Michigan. The control of these diseases, as well as appendicitis, depend on public enlightenment, reasonably adequate living standards, and available facilities. Under our present social structure, these control measures are not as avail- able to the colored as to the white pOpulation. Perhaps the higher suicide rate among the urban colored in Michigan as cmmpared with the corresponding segment of the Nation can be explained in terms of their rapid urbanization which has weakened the restraining influence of the religious mores while at the same time it has increased personal ten- sion, conflict, and disorganization. An examination of Table 6.19 indicates that rural- white death rates of fifteen of the twenty-five principal causes are higher in Michigan.than in the United States, and that rates of twelve of these causes are higher for the rural colored. Both segments of the rural pOpulation are more likely to die from.degenerative diseases in Michigan than in the United States. More rural peOple in Michigan, whether White or colored, die from accidents, both vehicle and non-vehicle, than in the United States. Suicide rates are higher in Michigan for both whites and non-whites. It is interesting to note, however, that rates for transmissible and infectious diseases, ailments for which modern science has developed means of control, are lower in rural Michigan 159 .ma.o oHQwE "condom .uoapmasmom was» ma omsmo mam» Eon“ unease on one; omega H shamapeaso .noz Hwnammonnohoo «anonpnman .qoz quammonnmhoo no>om poanwom stunts: Hobmu poansom nwsoo wsfimooma IIIHII. nwdoo weanooma ocaoaaom odflofiaom Inuit: Hopflom .nmoxm slats: powwow .nnon napanpno .monenmaa mapanpso .eonanwfin momsmo Honomnmsm autumn menses Honmmnomm nUQEOpm no hooab seasons 90 Hooab «Hanom cashew no>HH no mamonnefio nmpaa ho mamonnnao mutant mfipfioauaommd mapaoficnoggd mfiaaemem maafinssm eanouaoa Hepfisownoc Innis: .anouasa Hepanomnoo nuns-u omHoHsm oufloasm mmpnap endpoaonm mnpnan onspmaonm aspaaaoa mopopwan mdpaaaoa noponman aunts: mHmoHSonoQSB namoasononsa Anopofiv mpaowaood IIIIII AHOpQSV upqocaood A.nm> nozv mpqmcfiood A.no> nozv mpnocfieod sfiaoasonm sfinoasonm napannmoz napfinamoz uqofimoa Huansnosnan asOHmoH Hoanwnoouan 93an 13280 9.355 3.3280 endow ansom nopsem empmqp_nH sewage amen. mopspm meshes mH sameness mw pawMHaowonm nomum pqonaaouonm nopmm conoaoo sweep apnea aspen Anuwanofiz Hanan ma scapsasmom meant on» go copay hp demons mousse“ muam em .oeoa .mmaqam nmaHZD mmH 924.240HmUH2 24mmD_ZH demn ho mwmpdo 202200 fimB ho MoZdZHSOQMmm mHoo oapda I ‘ ‘:l ‘ III- III Ill-I‘Vri III‘Q.I~ .‘ a. ..l I:l!lI.II.IIII|»‘£ 160 0®Ho© mHQwB “OOHSOW .moapmasmom mfimp_qfi mmsmo mam» Son“ mmpmmc on one; whose H smsoepemfia .nmm meflomonnmnmo Hobo.“ poaawom nmdoo wqamooag oufiofiaom noaaow .ngoxm meHHpcm .monhnmfim momsmo Hanomnmsm «Haemm Hm>HH go mfimonnnflo mHeHoaeeoaeq .ahomawa Ampamomqoo H H IIHIII. .ILhIII somEOPm mo mnmoHD IIHIII mHHmmmhm moaofism agenda onspmaonm mspaaaoa mmpmpsflm shnoepegsm nmdoo wqamoona moaoflaom mapflnpnm .sonanwan momsmo Hwnmmnosm mspfloaeaogg4 mnpnfin onskwaoam .amE Hwnfimmonnonoo nobow poHHmom Hmpflow .nmomm nomaovm we mamoHD wfinhmm no>aa no mamonnnfio mHHHemsm .ahohawa Hepanowaoo ouaoasm mapsaaos mopopsan unannn mamoasonmpda mamoasonmpse sinus: A.no> nozv manocfloow Intuit A.no> nozv apnooaood nunnnn AH0poEv upsouaood Intuit Anopozv upoocfiood annasn canoasonn sfisoasosm mflpfinnmoz emphysema nutnns msoamoa Hmfinenoshqu sununn encamma asfissnosnpsH annals noonwo taunt: Hoodoo llllll. Phwmm nus-Illn- PHNQHH nopspm eopfiqp nH namaeosz cw. ampepm eo»HMbqu sewage aflnH pamsfiaowonm,uopmm. passaaooonm nopwm cohoaoo Assam opama Hunsm Admwanodz Hanan ea nofipsasmom spams on» mo nopsn he commas nomnsov moam_wm .oema .mmaaam omeHzp awe 024 zameOHs Admbm ZH madam mo mmmodo 202200 awe mo MoZdZHzcnmmm on.o canoe 161 than in the Nation. Syphilis is the only important trans- missible disease which makes a less favorable showing in this respect but in 1940 treatment methods were not as advanced as they are today. Furthermore, it is reasonable to assume that reporting was probably more accurate in Michigan since this State has been a leader in veneral dis- ease control for a number of years. In the urban areas of Michigan where treatment facilities are more readily avail- able and public health programs are operating with specific regard to this problem, syphilis rates are lower than for the Nation. They are 7.6 and 10.7 respectively for urban whites and 36.4 and 63.7 respectively for urban non-whites. The rural death rates from syphilis in Michigan are 12.7 for the white pOpulation and 504.7 for the colored, While for the United States they are 8.8 and 45.6 respectively. Among the rural colored peOple, syphilis is the second principal cause of death. Rates from tuberculosis, pneumonia and influenza, and premature births are lower in rural Michigan than.in the rural United States. These causes of death are frequently considered among the better indices of the healthful living conditions of a population. If this is true, Michigan's rural White peOple compare favorably in the healthfulness Cf their living conditions with the Nation. In the two areas the death rates from these causes for rural whites are as follows: 162 Rural White Population Area , Pneumonia Tuberculosis Premature births United States 70.0 34.0 23.2 Michigan 59.0 27.4 20.0 Difference 11.0 6.6 3.2 In Table 6.20 the age-adjusted death rates for eight leading causes of death are presented. Rural Michigan has rates lower than the Nation for three causes, namely, pneu- monia, tuberculosis, and nephritis. Michigan's death rate for pneumonia is 26.6 percent less than that for the United States, and the rate for tuberculosis is 30 percent less. Two observations may be made from the age-adjusted rates in this Table: (1) In general, urban Hichigan.has lower death rates from specific causes than the United States. This is true for all causes in addition to specific causes, except for intracranial lesions of vascular origin. The urban rate from this cause in Michigan is in excess of that of the Nation 1.4 percent. (2) In most instances, rural Michigan.has higher age-adjusted death rates from specific causes than the rural United States. This is true for five of the eight causes. Two communicable diseases, tubercu- losis and pneumonia, are among the three causes for which death rates are lower in rural Michigan. Infant mortality Differentials 0n the basis of data presented in Table 6.21, Mich- igan has lower infant mortality rates than the United States. This is true for the total, rural and urban areas, all sizes 163 Table 6.20 AGE-ADJUSTED DEATH RATES FROM SPECIFIE CAUSES, MICHIGAN AND THE UNITED STATES, 1940, BY RESIDENCE (Causes ranked by ratios for rural Michigan) Ratio of actual to expected2 deaths Cause Rural 2 Urban of : : : : : Death : . : : : Mich. 3 U.S. :MiChoxlOO: Hugha: U.S. :MiChoxloo 3 :UoSo 3 3 :UoSo Accident 114.9 104.7 108.7 97.3 97.5 99.8 (all) Intracranial lesions of vascular origin 102.1 96.0 116.8 98.7 97.3 101.4 Pneumonia 100.6 137.0 73.4 99.9 126.6 78.9 Heart diseases97.2 78.6 123.7 104.1 108.0 96.4 Nephritis 9408 15500 7002 10304 14001 7308 Cancer 9309 7706 12100 10603 11201 9408 Tuberculosis 90.8 129.8 70.0 123.2 141.5 87.1 Diabetes Mellitus 8404 7008 119.2 10908 11303 9609 All Causes 9608 9608 100.0 10206 11507 8807 1 Urban includes all incorporated places with 2500 inhabi- tants or more in April 1940. Rural embraces all the area outside of cities of 2500 inhabitants or more in 1940. 2 The expected deaths are computed by applying the age- Specific death rates in the total pOpulation of Michigan from.sach cause of death, to the corresponding age group of the urban and rural pOpulations of Michigan and the United States, and summing for all ages. See Dublin, Louis I, and Lotka, Alfred J., Length of Life,.gp. cit., p. 95 for an example of this method of adjusting cause- specific death rates for age in rural and urban areas. It is essentially the principles employed in the Indirect Method of adjusting rates for age, which are described and illustrated in Appendix I. The age-specific rates used as a standard for these eight causes are from.Linder, Forrest E., and Grove, Robert D., gp.'git., Table 25, pp. 476“4770 164 Table 6.21 INFANT MORTALITY RATES, MICHIGAN AND THE UNITED STATES, 1940, BY RESIDENCE Infant deaths under 1 year of age_per 1,000 live births Residence United Excess States Michigan U.S. Total Population 47.0 40.7 6.3 Urban Total 44.4 40.0 4.4 Cities: 100,000-over 39.3 37.5 1.8 Cities: 25,000-100,000 45.1 43.2 1.9 Cities: 10,000-25,000 48.4 39.2 9.2 Cities: 2,500-10,000 50.7 45.0 5.7 Rural T0tal 5001 4207 704 Source: Linder, Forrest E., and Grove, Robert D., Vital Spatistics Rates lg the United States 1900-I940, Table 28, pp. 578 ff. Urban total computed from Vital Statistics‘gf the United States 1940, Part II, TabIe 10, pp. 210,227. of cities, and for the leading causes of death. The Nation's infant death rate exceeds that of Michigan by 6.3, the re- spective rates being 47.0 and 40.7 per 1,000 live births. The differential between the rates in rural segments of Michigan and the United States is 7.4, the rates for the respective areas being 42.7 and 50.1. In the urban areas of the two pOpulations the rates are much closer together. Urban Michigan's infant death rate is 40.0, compared with 44.4 for the Nation. Infant death rates for the United States decrease as the size of the city increases. The rate in the very large cities is 59.5, as compared with 50.7 in the very small (I I. I. 165 Table 6.22 INFANT MORTALITY RATES FOR SELECTED CAUSES, UNITED STATES, 1940 MICHIGAN AND THE Number of deaths under 1 year of agegper 100,000 live births Michigan United U.S. x 100 States Mich. All Causes 407109 470109 11505 Premature births 1275.5 1370.4 107.4 Pneumonia (all) 600.9 630.4 104.9 Congenital malformation 594.8 467.5 78.6 Injury at birth 419.1 487.5 116.3 Diseases peculiar to 274.5 250.6 91.3 first year of life Diarrhea, enteritis, etc. 153.5 348.3 226.9 Accidental suffocation 100.0 V 115.2 115.2 and other accidents Diseases of thymus gland 99.0 44.5 44.9 Congenital debility 70.7 . 119.8 169.4 Diseases of respiratory system 69.7 53.4 76.6 Influenza 40.4 114.2 282.7 Whooping cough 33.3 85.4 256.5 Intestinal obstruction 29.3 31.5 107.5 111 defined and unknown causes 22.2 240.8 1084.7 Diseases of genito-urinary system. 20.2 14.0 69.3 Syphilis 1602 53007 32702 Diseases of circulatory system 14.1 17.5 124.1 Diseases of ear and mastoid ‘IItI 21.0 189.2 Diseases of the nervous system 9.1 10.2 112.1 Dysentery 'ETI 44.0 543.2 Measles 523; 7 .7 154.0 Note: Death rates based on less than 20 deaths are under- scored. Computed from.Vital Statistics of the United States, 1940, Part II, Residence, Table 2, pp. 69, 77. Table 14, pp. 525, 3290 Source: 166 cities. This generalization is true, also, in Michigan with the exception of cities of 10,000-25,000. They have next to the lowest infant death rate in the State and next to the highest in the Nation. Infant mortality rates by selected causes for Michigan and the United States are given in Table 6.22. Michigan has lower infant mortality rates for ten of the leading 15 causes on this list.1 However, when these are presented by residence (Table 6.23) only the ten leading causes can be compared, because the remaining ones are based on a very small number of deaths. or the ten leading causes, four have higher infant death rates in the United States in both rural and urban areas than in Michigan (Table 6.24). They include premature births, injury at birth, diarrhea, and congenital debility. 0n the other hand, four of the ten leading causes have higher gates in both rural and urban Michigan than in corresponding segments of the United States. They include congenital malformation, other diseases peculiar to the first year of life, diseases of the thymus gland, and other dis- eases of the respiratory system. The other two diseases are pneumonia and accidental. suffocation. The rate for pneumonia in the urban segment of the United States is higher than for Michigan, but the 1. Beginning with syphilis in Table 6.22, the rates for Michigan are based on less than 20 deaths, and are in- dicated by an underscore. .-II". ‘ . I finel- . .u I- ‘. ‘5 dais. ‘IU—a I MN- l.‘~ ‘1‘ ‘ ... .‘ \s I he. “4.21 167 Table 6.23 INFANT MORTALITY RATES FOR SELECTED CAUSES, MICHIGAN AND THE UNITED STATES, BY RESIDENCE, 1940 Number of deaths under 1 year of age gper 100,000 live births Cause Rural Urban United United Michigan States Michigan States All Causes 4,214.8 5,012.0 3,998.0 4,435.3 Premature births 1,075.9 1,272.2 1,378.6 1,454.8 Pneumonia (A11) 761.8 711.9 517.7 560.4 Congenital malformation 592.8 425.6 595.8 503.6 Injury at birth 36705 37900 44508 58007 Other diseases peculiar to first year 249.0 225.4 288.0 272.1 Diarrhea, enteritis, etc. 163.0 402.9 148.6 301.5 Accidents, suffocation 148.2 135.8 75.1 97.5 Diseases thymus gland 133.4 38.5 81.2 49.7 -Congenita1 debility 133.4 175.5 38.3 71.9 Other diseases of the respiratory system 83.0 59.9 62.8 47.8 I11 defined and unknown 35.6 430.8 15.3 77.4 Note: For complete titles of causes of death see source cited below. Death rates based on less than 20 deaths are underscored. Source: Computed from.Vital Statistics of the United States, 1940, Part II, Table 2, pp. 69 and 77, —Tab1eI14, pp. 525 and 3290 I. = 4 9 t“ ‘p'&‘ 168 Table 6.24 INFANT DEATH RATES FROM SEIECTED CAUSES, MICHIGAN AND TEE UNITED STATES, 1940, BY RESIDENCE Number of deaths under 1 year of age per 100,000 live births Residence Causes Predominant in the United States Premature Injury at Diarrhea, Congenital Births Birth Enteritis Debility Total United States 1,370.4 487.5 348.3 119.8 Michigan 1,275.5 419.1 153.5 70.7 . U.S. RatioomXIOO 115.5 116.3 226.9 169.4 Rural Uhited States 1,272.2 379.0 402.9 175.5 Michigan 1,075.9 367.5 163.0 133.4 , U.S. Ratio.——-Mich.1100 119.2 103.1 247.2 131.6 Urban United States 1,454.8 580.7 501.5 71.9 Nflchigan 1,578.6 445.8 148.6 38.3 , U.S. REtiOtfiizfitxloo 105.5 130.3 202.9 187.7 Causes Predominant in Michigan Congenital Other Dis- Diseases of Diseases Malformation eases Pe- Thymus of Respi- culiar to Gland ratory first year System Total NUchigan 594.8 274.5 99.0 69.7 United States 467.5 250.6 44.5 53.4 :_.U__1M1°h- 00 127. 109.5 222.5 130.5 Ratio .8. 1 2 Rural Michigan 592.8 249.0 133.4 83.0 United States 425.6 225.4 38.5 59.9 ‘Mich. . . . 138.6 Ratioeqfirgtxloo 139 3 110 5 346 5 Urban Michigan 595.8 288.0 81.2 62.8 United States 503.6 272.1 49.7 47.8 ,Midho . . 163.4 131.4 Ratio.—U——.S.x100 118 3 105 8 Smurce: Computed from Vital Statistics 2: the United States, 1940, Part II, Residence, Table 2, pp. 69,77, TaEIe 14, PP. 525,529. ‘- (‘I‘ I 159 rate for the rural segment of the Nation is lower than for the corresponding group in the State. Similarly, the rate for accidental suffocation and other accidental deaths in the United States is 15 percent higher than that of Michigan, but the State's rural rate is in excess of that for the rural segment of the Nation. Michigan's urban rate, 75.1, is lower for this cause of death than the corresponding rate, 97.5, for the United States, but rural Michigan's rate, 148.2, is higher than that of the Nation, 135.8. Thus, urban Michigan has lower rates than the urban part of the Nation for six of the ten leading causes of infant deaths, While rural Michigan.has higher rates than the United States for six of the ten leading causes. Table 6.25 NATERNAI. MORTALITY RATES, MICHIGAN AND TEE UNITED STATES, 1940, BY RESIDENCE Deaths of M0thers Per 1,000 Live Births Residence United _ Excess States Michigan U.S. Total Population 3.8 2.9 .9 Urban Total 3.6 3.2 .4 Cities: 100,000 and over 3.1 3.3 -- Cities: 25,000 ‘ 100,000 5.7 3.3 .4 Cities: 10,000 " 25,000 4.0 2.5 1.5 Cities: 2,500 - 10,000 4.3 3.4 .9 Rural Total 4.0 2.3 1.7 —¥ Note: Underscored rates are based on less than 20 deaths. Source: Linder, Forrest E., and Grove, Robert D., Vital Statistics Rates in the United States 1900-1945, Table 57, pp. 6227—632. 170 Maternal Mgrtalitz Differential}. The mothers of Michigan die from puerperal causes at lower rates than those of the Nation in both rural and urban areas, and for all pOpulation-size groups except large cities of 100,000 pOpulation and over (see Table 6.25 on the preceding page). The maternal rates of the Nation.and.thhigan are 3.8 and 2.9 deaths per 1,000 live births, the excess of the Nation being 31 percent. Rural Mishigan has the lowest maternal rate of any pOpulation-size group. Michigan's rural rate, furthermore, is exceeded by the corresponding rate of the Nation by 74 percent, or 1.7, this being the largest differential of the group. In contrast, the urban people of the United States and Michigan have maternal death rates of 3.6 and 3.2 respectively, the urban part of the Nation's rate being only 13 percent in excess of the rate for the State. The maternal rates for cities of the United States decrease as the size of the cities increase, but this is not true for the cities of Michigan. Small cities in the United States, those with 2,500 to 10,000 population, have a rate of 4.3, compared with 3.1 for the Nation's large cities of 100,000 population or more, a difference of 1.2. The Internal rates for cities of corresponding size in Michigan, however, are 3.4 and 3.8, respectively, a difference of only .8. PART III INTERNAL VARIATION CHAPTER VII MORTALITY DATA BY COUNTIES The crude residentl death rates have value in locating the areas in Which the incidence of death and fatal diseases are likely to occur most frequently. The importance of such data is well evaluated by the following statement from a pub- lication of the United States Bureau of the Census written by Linder and Grove:2 "Public health agencies are organized to deal with health problems by areas. The first question for them.is how many pSOple are dying and where are they dying. Important as it is, the question of why they are dying comes second." Death is usually preceded by illness, and thus the need for health facilities is usually great where death occurs at high rates. It is recognized that the number of deaths which occur in a community is a very useful index of the need for hospital beds in that area.3 The practical value of this data, analyzed for localities, will increase in importance as health pregrems begin to devote greater attention than at the present time to those diseases Which are prevalent among the older aged population. Because of these uses of mortality incidence data and also the emerging interest of the general public, 1. Deaths are allocated to place of residence. 2. ‘Linder, Forrest E. and Grove, Robert D.,‘gp. cit., p. 13. 3 See Hospital Resources and Needs, Report of the Michigan . Hospital Survey, W.K. Kellogg Foundation, Battle Creek, IMichigan, 1946, Chapter VII, for a discussion of the bed- death ratio and its use in estimating the number of hos- pital beds needed in communities. 171 1'72 government, and health leaders in adequate care of the sick through hospital planning and extention of health facilities, it is considered desirable to present a detailed analysis of ten-year average crude resident death rates for 111s counties of Michigan. To assist in the analysis of the distribution of deaths by county, several maps have been prepared. These maps show the number of deaths by county for either per 1,000 01' per 10,000 population. In this Chapter, consideration is Sivan to the distribution of deaths by county for all causes and. for the principal causes, followed by an analysis of residential variation by county in Chapter VIII. Crude and Age-Adjusted Death Rates: All Causes Death rates from all causes are presented in Table 7.1 and Figure 7.1. Inspection of Figure 7.1 shows that the toll of deaths from all causes is highest along the southern bor- der of the State and in the western half of the Lower Penin- an«18.. The Upper Peninsula has only three counties where the °rude resident death rate is more than 11.2 per 1,000 popula- tion. With the exception of three counties in the Thumb area (St . Clair, Sanilac, and Tuscola), and Ogemaw and Alcona, the Baa‘I'iern counties of the Lower Peninsula have lower death rates than those in the rest of the State. The twelve lowest and the thlve highest crude resi- dent death rates by county are listed below; the twelve low- 331: have death rates under 10.1 and the twelve highest are IE. .I .... Q0 arm . Q . nu .3‘fl A 4 0.0.. C 8 ...... l V .... . 90 O. L 0 . . x, L III e C . A = “."N". O O .D o .. M ...W.*.. C O C v 0 Q0 . E ......Q... v M % “O“"O“”"“ . O 9 . . R | H w T. AH E O nu 0.».fir1v Wm ...:. e . dpwmzvxzestKW&&£vv uiwmpnu _Azupezv . . Q 00.95.}. I, .0 O C C O 0 ....e . O 2 R ... J in . q. o. p: ..o.. I. .I e. .3 . . '0’.’.D. 9 . I'. II. II. v 0 O O . R m I. m B 0 E m . . . . n_v P D O my. 0 m 2 N 2. 0. 2. . 3. mm ”u e. I. nu m“ .1 TI MM Table 7.1) Figure 7.1 (Source: III-3|- I'lltnll If‘ 1A 51.1.- .11.- ..d ...]- 5:]- ‘1': Iv]! I ‘ .‘Aiflu ll‘wu‘HII 1.". ‘ 111‘ ‘ll‘ ‘ 1 ll." .11 1 iii Ny.‘ y..- N‘Vi \ Ax...‘ 1'73 Table 7.1 AGE-ADJUSTEDl AND CRUDE 3331mm DEATH RATES FROM ALL CAUSES PER 1,000 POPULATION, COUNTIES OF MICHIGAN, AVERAGE 1935-1944 (Based on the 1940 population; by place of residence) State and : Death Rates : State and : Death Rates County :AdJusted: Crude : County : Adjusted: Crude Alcona 9.7 11.6 Lake 10.3 13.3 Alger 1201 1102 Lapeer 1000 1101 Allegan 9.4 12.3 Leelanau 9.7 11.2 Alpena 9.8 10.4 Lenawee 10.1 13.2 Antrm 1005 1208 LiflngSton 909 1209 Arenao 9.6 10.9 Luce 8.7 10.3 Baraga. 11.5 11.1 Mackinac 10.9 11.0 Barry 9 o4: 130 2 Macomb 10 o 5 8 o 8 Bay 10.3 11.0 Manistee 10.3 13.4 3911219 11 . 0 13 . 8 Marquet to 10 . 9 11 . O Benian 1001 1106 Mason 1000 1202 Branch 10.5 14.6 Mecosta 10.6 13.5 Calhoun 10.7 12.2 Eenominee 10.1 11.0 Cass 10.3 14.8 Midland 9.0 8.3 Charlevoix 10.6 13.3 Missaukee 9.9 10.6 Cheboygan 10.4 12.2 Monroe 10.0 10.1 Chippewa 11.8 11.2 Montcalm 10.4. 14.0 Clare 11.3 15.0 Montmorency 10.0 10.6 Clinton 9.4 11.6 Mushegon 10.7 10.2 Crawford 11 . 8 12 .0 Newaygo 9 . 6 11.9 Delta 10.7 11.1 Oakland 9.7 8.2 Dickinson 10.1 9.5 Oceana 10.3 13.7 8L17cm 9.7 12.9 Ogemaw 10.2 11.7 t 11.7 13.2 Ontonagon 11.0 11.3 Ganesee 10.1 8.5 Osceola 10.2 13.4 Gladwln 9.4 10.7 Oscoda 9.0 8.9 Gogebic 11.7 10.8 Otsego 10.0 11.1 I'IBLntS. Traverse 8.6 11.5 Ottawa 8.7 9.7 Gratlot 10.4 13.2 Presque Isle 10.1 9.3 Hilladale 9.6 13.7 Boscommon 9.2 10.2 Houghton 11.1 12.5 Saginaw 10.1 10.4 Huron 9.6 10.8 St. Clair 10.7 12.0 Ingham 9.8 9.7 St. Joseph 10.0 13.5 Ionia 9.6 12.5 Sanilac 9.6 11.9 I08 00 9 .3 11.0 Schoolcrart 10.3 11.1 it on 9.8 9.1 Shiawassee 10.0 12.2 Jaabella 9 .9 10.5 Tuscola 10.0 12.1 Kaekson 10.2 11.3 Van Buren 10.0 14.6 Kalamazoo 9.7 10.8 Washtenaw 9.2 10.8 Kalkaska 9.8 11.3 Wayne 11.1 8.9 19:11 10.4. 10.4 Wexford 10.3 12.6 Wax-law 11.4 12.2 State 10.2 10.1 301. 1 Based on the United States population, 1940, indirect method. 1‘03: Reports of the Michigan Department of Health, 1955- 1944, Table 16. 1’74 13.8 or over per 1,000 pepulati on. Twelve Lowest Crude ‘i'velve Riga st Crude Re sident De ath Rates Resident De ath Rates Oakland 8. 8 Gas s 14.8 Midland 8.3 Van Buren 14.6 Gene see 8. 5 Branch 14.6 lacomh 8 .8 Mont calm 14.0 Oscoda 8.9 Benzie 13. 8 Weyne 8.9 Oceans 13. '7 Iron 9 .1 Hilledale 13. 'I Preeque Isle 9 . 3 8t . .T Oseph 13. 5 Dickinson 9 .3 llacosta 13.3 0t tawa 9 .7 Mannie tee 13.4 Inghen 9 .‘7 Osceola 13.4 M011”. 10e1 Lake 13.3 Oakland county with the lowest death rate (8.8) and Gus county Iith the highest death rate (14.8) are both lo- cated in the southern part or the State." However, Oakland 1: an industrial county whereas Case is predominantly rural. live or the high death rate counties are in the ex- treme southern part or the State, grouped fairly close to- sethar. These are Cass, Van Buren, Branch, Hillsdale, and St. Joseph counties. or the remaining twelve. Denna. Mani ates, and Oceans are on the western coastline of the up“ 1’” part or the Lower Peninsula: and Lake, Osceola. and “None are in the heart of the Lower Peninsula. No Upper P9111 nsula county is included in this group. Hence, we may “M on the basis at the ten-year averages that the 1:011 °r anm 13 greater among those living in the western half °‘ the state. 81: or the counties with 10! death rates are located ‘“ 1=11. southeastern part or the State. They are t1? 1114“" “1&1 or ”mi-rinduetrial counties of Oakland, Genesee. “‘e 59“ Va 1 e‘“: M“ uh. c! e. ‘:‘ 1'75 Macomb, Wayne, Monroe, and Ingham. One county, Ottawa, lies in the western part of the State; three counties, Midland, Oscoda, and Presque Isle are in the northeastern part of the Lower Peninsula; and two, Dickinson and Iron, are in the Upper Peninsula. It is interesting to note that counties of the Upper Peninsula are not found usually in either the lower or the higher death rate brackets. It is likely that the pOpulation is too young to place it in the higher brackets, while its communicable disease rate is too high to place it in the lower brackets. Age-adjusted death rates are also given in Table '7.]. and are shown graphically in Figure 7.2. They are more comparable than crude rates, for they are not in- fluenced unduely by differing age distributions among the counties. Rather, variations between such death rates are usually considered to be a result of conditions which may be improved, such as poor sanitation, lack of health faci lities, lack of income with which to purchase medical care , or poor health habits and practices of the people. For the most part, the high age-adjusted death rates fall in the Upper Peninsula, while low rate counties are found in the Lower Peninsula. The counties with the "'3 :Lve highest and the twelve lowest age- adjusted death rates are listed below: AGE ADJUSTED DEATH RATES IO.| AV. I935' 44 e e e’eae‘ O e ce ’e’ o. . :e ,e ’2 . O 9 . . . 90'... . A... ..H\ e... . C O N. "OMOOOW . I 0 shoe... .e ... e vecseec... I _ e e9 4. 4.. O ...eaneeee r .. e. _ 1:. e e. ...e.e.l.eeco . A7 . / .cc .... ccc . w . c .. e i . . r/ .. ......m... C ‘ C. v... coco...H . O 9 9 9 9 R . o. 4. e 4. e s ...... u.“ 9 9 9 O O nvo v0 ewe”... fl ...I "oneeeecu R R _ _ _ 0.x.» E M 0 O 0 0 0 O O O 0. mass P N o. s o 5. o I 1 U 9 9 O O I m. .. - - T ,,,, I z . A ,9 E V1, 1 D 1“ (Source: Table 7.1) ”sure 7.2 1'76 Twelve Lowest Age-- Twelve Highest Age-~ adjusted Death Rates adjusted Death Rates Grand Traverse 8.6 Alger 12.1 Luce 8.7 Chippewa 11.8 Ottawa 8.7 Crawford 11.8 Midland 9.0 Emmet 11.7 Oscoda 9.0 Gogebic 11.7 Roscommon 9.2 Baraga 11.5 Wadhtenaw 9.2 Keweenaw 11.4 Iosco 9.3 Clare 11.3 Allegan 9.4 Houghton 11.1 Barl'y 9 e4 Wayne 1101 Clinton 9.4 Ontonagon 11.0 Gladwin 9.4 Benzie 11.0 .nger county in the Upper Peninsula has the high- est rate and Grand Traverse in northern Michigan has the ILcnwest. Of the twelve counties with low age-adjusted death rates, one, Luce, is in the Upper Peninsula and the rest are below the Straits. They are for the most part Inlrnl. In contrast to this distribution, seven of the counties with high death rates are in the Upper Peninsula and they include both rural and urban counties. On the basis of these data, the people in the fifteen counties of the Upper Peninsula occupy a less fa“Torable health position than those in most of the °°unties of the State, and one naturally seeks ”an expla- hat; ion. Dorn found that in Ohio the death rates in poor eC’Q’Iiomic areas were greater than corresponding rates in good economic areas.1 If this conclusion is tested for “13 twelve lowest and twelve highest age-adjusted death \ 1' Dorn, Harold F., "Mortality Rates and Economic Status Io. in Rural Areas " Public Health Re orts, Vol. 55, l 1, January 5, 1940, pp. 3-12, (Reprint No. 2126, p. 3). 1'77 rate counties of Michigan by employing standard of living measuresl as criteria of economic status, there is indirect evidence to support the generalization. For the twelve low mortality counties the mean death rate is 9.1 and the mean standard of living index 101, whereas for the twelve high mortality counties the mean death rate is 11.5 and the In351n.standard of living index is 96.2 Death rates in the high mortality counties are about 26 percent greater than in the low mortality counties. Moreover the standard of Jniwring measure is about 4 percent greater in the twelve areas with lowest death rates. This information also sug- gests an inverse relationship between mortality and standards of living which is in keeping with the findings of Dorn and _ l. The standard of living measures employed here are those given in Hospital Resources and Needs, Michigan Hospital Survey, 92. cit., Table 13, p. 66. The standard of liv- ing measure s an average of ten characteristics after expressing each item in standard deviation units. The items included in the average follow: percentage of homes with radio, mechanical refrigeration, central heating, electric lighting, running water, occupied by owner, needing major repairs, less than 1.01 persons per room, average value of homes, and percentage of persons 25 and over having completed more than 6 years of school. 2’ The standard of living measures for the counties with the twelve lowest age-adjusted death rates are as fol- lows: Grand Traverse 104, Luce 85, Ottawa 114, Midland 103, Oscoda 87, Roscommon 94, WaShtenaw 115, Iosco 97, Allegan 106, Barry 107, Clinton 108, and Gladwin 91. Those for the counties with the twelve highest age- adJusted death rates are: Alger 91, Chippewa 96, Crawford 92, Emmet 99, Gogebic 98, Baraga 88, Keweenaw 89, Clare 94, Houghton 98, Wayne 112, Gntonagon 90, and Benzie 102. Arithmetical means given in this Chapter for the 12 high and 12 low death rates or their corre- sponding standard of living measures prevent counties ‘iith large populations from.determining the size of the figure. O O C v I l I . g D . O O . O p t . _ . t t . O I n _ ' \ b. | n . ~ ‘ e ' e .7 . F . ' \ . e I . . t t. e 178 others.1 Wayne county, with approximately one-half of the State '8 pOpulation, ranks eleventh among the counties with the twelve highest death rates, while Berrien county, with around 80,000 pOpulation, ranks sixth highest. These two counties are pOpulated predominantly by people neither very young nor very old, md many conditions resulting in high death rates are largely preventable. The Principal Causes _g_i_'_ Death: Cause-Spa cific Death Rates The predominant causes of death to be discussed by county are given below with their relative rank for the Statee Rate per 1,000 population Rank (Tenryear averages) Heart Diseases - 2.85 Cancer 1.16 Apoplexy .88 .Accidents (all) .76 Pneumonia .57 Nephritis .55 Tuberculosis .36 Premature births .27 Diabetes mellitus .26 Arteriosclerosis .19 Suicide . .12 Te11~year average crude resident death rates were computed 1’ See: Coombs, I..C., "Economic Differentials in Causes of Death,” Medical Care, Vol. I, July 1941, pp. 246-255; Mott, Frederick D. (M.D.), and Roemer, Milton I. (M.D.), Rural Health _a_i.1_d_ Medical Care, McGraw—Hill Book Company, New York, 1948, p. 71-73; Registrar General, The Regis- trar General's Decennial Supplement; England Eng Wales, H.M. Stationery Office, London, England, 936, VOL 11, p. 211, and pp. 191-210; Thompson, Warren S., _op. 31:31., pp. 234-239. 1'79 because they provide a more stable measure of the occurrence of deaths. This is particularly important in shell counties there a variation of a few deaths would affect the size of the rate materially.1 These causes comprise approximately 80 percent of all deaths occurring to residents of the State during the ten year period 1935-1944. It is the object of this section to show how the incidence of death from these causes are distributed among the counties of the State. To aid in the presentation of the data, maps have been prepared showing the average number of deaths per 1,000 pepulation in each county for each of the senses listed except arterio- sclerosis and suicide. 1. m Discus! (All Forms). Heart diseases take £greater toll than any other cause of death in all of the counties of the State (See Appendix II, Table I for rank). - 1. Death rates should be interpreted with caution vhen based on a small number of deaths. According to Linder and Grove ( . it., p. 93), a death rate based on 20 deaths "has as an ard error of roughly 28 percent of the magnitude of the rate." Many of the death rates by causes for the counties are based on fewer than 20 deaths per year and this is indicated by underscoring the rates. It might be pointed out that the number of deaths that occur in a county is not a sample but a complete enumera- tion. Since Michigan's registration system is among the more efficient ones of the Nation, inconsple te registra- tions being less than 2 per cent, inaccuracies from this source have been reduced. To counterbalance county var- iation in the number of deaths due to seasonal and unusual occurrences, the deaths in each county for ten years (1935-1944) have been averaged and related to the popu- lation of 1940. It is believed that these death rates present a fairly accurate estimate of mortality among counties. (See Linder, Forrest E. and Grove, Robert D., . £13.09 pp. 93.94 for a discussion of the problem of eath rates based on small numbers of deaths; See also this thesis, pp.42-43.) 180 The ten-year average death rate from heart diseases for the State is 2.9 per 1,000 pOpulation (Table 7.2). Observation of the map, Figure 7.3, reveals, as might be expected, that the high death rate counties follow Closely the pattern of death rates for all causes. With the exception Of Muskegon and Ottawa counties, rates are at or above the State average for the entire western half of the Lower Peninsula. There is also a group of high rate counties running along the Eastern Shore line from St. Clair as far north as Alpena county. Counties of the Upper Peninsula have low rates from.these causes. Only four of the fifteen counties, namely, Chippewa, Luce, Sohoolcraft, and Houghton have rates as high as 3.2 per 1,000 pOpulation or higher. The belt of low death rates from.heart diseases in counties running up the eastern half of Michigan from.Wayne to Presque Isle is a mortality condition caused primarily by younger aged popu- lations in these counties.1 Twelve Lowest Death Rates from.Heart Disease Twelve Highest Death Rates from.Heart Disease iidland 2.02 Van Buren 4.63 Oakland 2e03 M8111 stee 4e46 . Genesee 2.16 Branch 4.45 Macomb 2.22 Emmet 4.35 ' Presque Isle 2.30 Livingston 4.15 Baraga 2.32 Charlevoix 4.04 Iron 2.36 Eaton 4.02 Wayne 2.41 Montcalm. 3.99 Oscoda 2.44 Mecosta 3.97 Alger 2.57 Benzie 3.96 Gogebic 2.58 Wexford 3.93 .Ingham. 2.62 Cass 3.90 1. see B66819, J. Allan, .92. Cite, pp. 49-51. HEART AV. I935 ‘44 3 2.852 0 DE ATH RATES '- \ O s . O O . C C e ..eeeee cue 9e. 0000. A . O. O c. e co DEATHS PER 00.. I000 POP. O . 5.9..“ .O I I II. e0...” ’0 o. . "I. 0.0. 0.. O O ...e“‘. Os. . ee. e D 9 9 Av. .Il/ln.» .. . WA 2.50 - 2.84 @ ass - 3. l9 a 3.20- 3.54 .0. . 0". 0 D .O . eeee. ween. . e - 3.90- ovss ratio 7.2) (Source: 1'16“" 7.3 181 Table 7.2 CRUDE DEATH RATES FROM HEART DISEASES AND CAflCER PER 1,000 POPULATION, BY COUNTY, MICHIGAN, AVERAGE 1935-1944 (Based on the 1940 population; by place of residence) Heart : Heart (hmunty Diseases Cancer : County Diseases Cancer (so-95) (45-55)1: _L90-95) 445-55)l Alcona 3051 1.12 Lake 3e79 1.38 Alger 2.57 .98 Lapeer 3.10 1.19 Allegan 3.85 1.37 Leelanau 3.77 1.23 Alpena 3.34 1.01 Lenawee 3.62 1735 Axitrim. 3.73 1.41 Livingston 4.15 1.44 Arenac 3.30 1.05 Luce 3.85 .69 Baraga 2 . 32 .233}. Mackinac 2 .96 1753 Barry 3.78 1.39 Macomb 2.22 . Bay 3e04 leg]. Manistee 4e46 le65 Benzie 3.96 1.27 Marquette 2.97 1.27 Berrien 3.43 1.30 Mason 3.49 1.32 Branch 4.45 1.47 Mecosta 3.97 1.25 <3a1houn 3.66 1.27 Menominee 3.08 1.36 09.88 3.90 leVl Midland 2e02 e88 Charlevoix 4.04 1.57 Missaukee 3.07 .85 Cheboygan 3.54 1.44 Monroe 2.95 1731' (Shippewa 3.36 .99 MCntcalm. 3.99 1.41 <31are 3.87 1.19 Montmorency 2.71 1.20 (Slinton 3.06 .17 Muskegon 2.72 1.06 (Drawford 3.08 1.51 Newaygo 3.76 1.23 Delta 2e93 le30 Oakland 2e03 e92 Dickinson 2.71 1.10 Oceana 3.64 1.47 IEaton 4.02 1.54 Ogemaw 3.19 .96 IEmmet 4.35 1.41 Ontonagon 2.73 1711 (Genesee 2.16 .90 Osceola 3.89 1. 9 (Gladwin 2.95 1.16 Oscoda 2.44 1.10 (Gogebic 12.58 1.14 Otsego 3.55 1.15 (Grand Traverse 3.47 1a16 Ottawa 2.93 1.34 (Gratiot 3.57 1.34 Presque Isle 2.30 1.04 Hillsdale 3.85 1.55 Roscommon 3.05 1.09 Boughton 3.46 1.59 Saginaw 2.95 1.14 IHuron 2.81 1.26 St. Clair 3.35 1.38 Ingham 2.62 1.14 St. Joseph 3.78 1.51 101113 3e70 1e41 Sanilac 3e35 1.21 Iosco 3.36 1.44 Schoolcraft 3.22 1.31 Iron 2.36 1.18 Shiawassee 3.20 .49 Isabella 2.96 1.13 Tuscola 3.75 1.37 Jackson 3.37 1.29 Van Buren 4.63 1.78 IKalamazoo 3.04 1.22 Washtenaw 3.14 1.32 KalKESKQ 3e4l 1.10 Wayne 2e41 1.05 Kent 3e41 le34 WBXfOI‘d 3.93 lel? Heweenaw 2.85 1.22 State 2.85 1.16 1. International List number, 1939. Nets: Source: 1944, Numbers 64-74, Table 16. Rates based on less than 200 deaths during ten years are underscored. Annual Report, Michigan Department of Health, 1935- 182 The counties with the twelve lowest and the twelve highest death rates from heart diseases are given on the .preceding page. In general, the counties with the twelve lowest death rwates from heart diseases have heavy concentrations of urban people, while the counties with the twelve highest rates from these causes are esSentially rural. The highest rate (14.63) is in Van Buren county and the lowest rate is in Mid- land,--2.02 per 1,000 pOpulation. 2. Cancer. The death rates for cancer are given in 'Table 7.2 and are presented graphically in Figure 7.4. The rate for the State is 1.2 per 1,000 pOpulation. It is the second highest of the principal causes of death in Michigan, and is second highest for a majority of the counties. In 53 of the 83 counties (64 percent) cancer is the second highest cause of death while in 26 counties (31 percent) it ranks third. The incidence of death from.cancer follows closely that of heart diseases. The higher rates occur in the tier of counties along the southern border, in the central, and in the western areas of the Lower Peninsula. In general, the Upper Peninsula counties have lower death rates than those of lower Michigan. The counties with the twelve lowest and those with the twelve highest death rates from cancer are listed below: ,(I DEATH RATES: CANCER AV. l935- 44 = l.|56 I t. '3 ‘3 . "1' ‘3» fi’K‘ deg/Q, ”A ... r ,,.J .. N O O 19 O. O . $000 9 'DO // ”a, 1 of m5- ' '/ ' saw/714g . ° ‘69.: (a 49' J . A"; V V v f v V..v vv‘ ‘ ' .90... ..e.'e.o,:¢:o:q.e:e.o‘f J, .4, . /" H ' 4' 0’ 9’ o o’o’o’o‘cv’O’o‘ t W x 0' 0’ //;:/ UNDER 95 x -~‘ w .c . n+4... swab-,6 { .0 ' v. oiV?“'€'¢%W% 0k , / ..o o ’flflfif e 04 .3 .0 0.0-0.4 5... v ' fififlw.°. d’V5 . z . ‘ . . . .0 O ' i s / . 0 5.0.0.1; O . % ... .’ . / /, .95 -l.09 ° sun: .0. “4.3.30 \. O 09‘ ' o w ~ 0 \ 71.15. V \ o O ’ . - l.24 V’fi'.’ .0 : C O L4. oft». . o o. X [/ R: DEATHS PER IOOO POE, I . 0. . D HO . 0 L25 - L39 3.; x 9'... c'o‘s‘ o ' o 0 90' o c o .....O‘ o c ‘0 o e ' o c l 40 - ' 5 4 " "9.. “" ... ‘3. o . o ‘.?&?‘9.' . . - . . ' - vov.v.v .' .V .V .V .7 ' 30040 c c c w - ." u""°3".°‘."."o°’~ ' ’ ' "0‘ 1r. \ I.55 OVER ,1..O‘; 93.4164. 6"" g If! 5) p 0.] | ) . O {J . . a g. a J . Figure 7.4 (Source: Table 7.2) 183 Twelve Lowest Death Twelve Highest Death Rates from Cancer Rates from Cancer Luce .69 van buren 1.78 Missaukee .85 Cass 1.71 Midland .88 Manistee 1.65 Genesee .90 Houghton 1.59 Baraga .91 Charlevoix 1.57 Oakland .92 Hillsdale 1.55 Cgemaw .96 Eaton 1.54 Alger .98 Crawford 1.51 Chippewa .99 St. Joseph 1.51 Macomb 1.00 Osceola 1.49 Alpena ' 1.01 Shiawassee 1.49 Mackinac 1.03 Branch 1.47 Five of the twelve counties with low rates for cancer are located in the Upper Peninsula and most of them are pre- dominantly rural; three are in the northern part of the Low- er Peninsula; and four are located in southern Michigan. Two of the last four counties mentioned are predominantly urban and two have large urban pOpulations. Seven of the 12 counties with high death rates are essentially rural and are characterized by large numbers among the older ages. They are located in southern Michigan. Van Buren county in southern.Hflchigan has the highest death rate, namely 1.78, and Luce in the Upper Peninsula has the lowest, .69. 3. Apoglegy.l Apoplexy is another chronic disease which takes its toll from among the older aged persons. Almost one person per 1,000 (.88) dies from.this cause in Michigan (See Table 7.3 and Figure 7.5). Although apoplexy is the third principal cause of death for the State, it occupies this position in slightly less than 50 percent (48.2) 1. Includes Cerebral hemorrhage (83a) and Cerebral embolism and thrombosis (83b). - 11.1-flunk A .m: nu. \IJ \J Nahum“ a E I \ nnnnnnnnnnnn OOOOOOOOOOOOO DEATH RATES '- APOPLEXY l935-44 =.880 AV. IOOO POP. DEATHS PER UNDER . II] II VIA .se - .99 m LOO -1.u / // // A, o 1 9.9... a .... . . .. . cc 0. 00 9009. co. 9 c. 99 O o e . O 6 he.» ...... . .... ...... . | 90000. .0000. o o e oocee. .3 "one". . n l.24 -l.35 Table 7.3) (Source : “an" 705 184 Table 7.3 CRUDE DEATH RATES FROM APOPLEXY AND ACCIDENTS PER 1,000 POPULATION, BY COUNTY, MICHIGAN, AVERAGE 1935-1944 (Based on the 1940 popula- tion; by place of residence) County Apoplexy Accidents: County Apoplexy Accidents (83ab) (169-195): (ssab) (169-195) .Alcona 1028 1008 Lake 1052 1021 .Alger .64 1036 Lapeer 103i 095 Allegan 1.33 .02 Leelanau 1.09 .79 .Alpena 1.05 4§1_ Lenawee 1.42 1707 Antrim. 1047 076 LiVingSton 1011 1013 .Arenac fggg ITIE Luce .62 1.06 Baraga ‘ggg 1.35 Mackinac 1765' 1721 Barry 1.69 1.04 Macomb .86 . 0 Bay 092 086 Manistee 1059 086 Benzie 1.18 1.24 Marquette .93 .83 Berrien 1.05 .98 Mason 1.18 ';39 Branch 1.68 1.10 Mecosta 1.47 .96 Calhoun 1.32 .90 Menominee 1.05 jzzg Cass 1.64 1.29 Midland .Lél 'ng Charlevoix 1.34 1.03 Missaukee 1.27 .72 Cheboygan 023 100 Monroe .86 1004 Chippewa 089 097 Montcalm 1053 1018 Clare 1.15 1.41 Montmorency .60 1.15 Clinton 1.14 .85 muskegon .85 .70 Crawford .séé 1.14 Newaygo 1.31 'ng Delta 1015 1003 Oakland .67 073 DiCkinSOn .80 083 Oceana 1059 1011 Eaton 1.30 1.04 Ogemaw 1.72 1.09 Emmet 1.41 ‘;98 Ontonagon 1.33 '1708 Genesee .65 .70 Osceola 1.59 1.09 Gladwin 1.10 [:98 Oscoda ;Z_. ,;12 Gogebio 1.04 1.06 Otsego l;zg 1.00 Grand Traversel.41 .86 Ottawa .99 .63 Gratiot 1.08 .91 Presque Isle ;§Z_ .71 Hillsdale 1.62 1.04 Roscommon .76 ‘758 Houghton 1.28 .85 Saginaw “755 731 Huron 1.02 .95 St. Clair 1.17 1.05 Ingham. 1004 079 St- Joseph 1052 096 Ionia 1051 085 Sanilao 1049 091 Iosco 1.18 ‘;21 Schoolcraft ';gz 1.01 Iron .;_2 ,;g§ Shiawassee 1.17 .96 Isabella .99 .91 Tuscola 1.23 .96 Jackson 1.08 .87 Van Buren 1.64 1.17 Kalamazoo 1009 088 Washtenaw 1008 085 KalkaSka 1042 ‘;Z§ Wayne .56 058 Kent 1.01 .73 Wexford 1.16 :83 Keweenaw 1.57 .485 State .88 .76 1. International List number, 1939. Note: Rates based on less than 200 deaths during ten years are underscored. Source: Annual Report, Michigan Department of Health, 1935- 1944, Numbers 64-74, Table 16. 185 of the counties. The distribution of counties by rank are as follows: Rank Number of Counties Percent 2 20 24 3 4O 48 4 l6 l9 5 5 6 6 1 1 In one county, Emmet, apoplexy and cancer have the same death rate, namely 1.41. As might be expected, the in- cidence of death is high in the southern tier of counties and in the western part of the State. The counties with the twelve lowest and the twelve highest rates are given below: Twelve Lowest Death Twelve Highest Death Rates from ApOplexy Rates from Apoplexy Midland .51 Ogemaw 1.72 Wayne .56 Barry 1.69 Presque Isle .57 Branch 1.68 Montmorency .60 Cass 1.64 Luce .62 Van.Buren 1.64 Alger 064 Hillsdale 1062 Genesee .65 Oceana 1.59 Crawford .66 Osceola 1.59 Oakland .67 Manistee 1.59 Iron .69 Keweenaw 1.57 Oscoda .71 Montcalm 1.53 Otsego .72 St. Joseph 1.52 Four of the highest twelve rates are for counties in the southern tier, and two are in the adjoining tiers. These counties are Barry, Branch, Cass, Van Buren, Hills- dale, and St. Joseph. Only one of the high rate counties, Keweenaw, is located in the Upper Peninsula. The low rate counties are distributed throughout the State; three are in the Upper Peninsula, six including Midland are in the northern part of the Lower Peninsula, and three are the 186 industrial counties, Wayne, Oakland, and Genesee, in south- eastern Michigan. The highest rate for apOplexy is 1.72 in Ogemaw and the lowest is .51 in Midland. 4. Agcidental Deaths. The distribution of deaths from accidents of all kinds is shown in Table 7.3 and Figure 7.6. It is the fourth principal cause of death in the State, the rate being .76. In 55 counties, or 66 percent, it is likewise the fourth principal cause of death, but in 22 counties, or 27 percent, it occupies second or third place. The counties with the twelve lowest and the twelve highest death rates follow: Twelve Lowest Death Twelve Highest Death Rates from.Accidents Rates from.Accidents Wayne .58 Clare 1.41 Ottawa .63 .Alger 1.36 Muskegon .70 Baraga 1.35 Genesee .70 Cass 1.29 Presque Isle .71 Benzie 1.24 Missaukee .72 Lake 1.21 Oakland .73 Mackinac 1.21 Midland .73 Montcalm 1.18. Kent .73 Van Buren 1.17 Antrim. .76 iontmorency 1.15 Menominee .78 Crawford 1.14 Kalkaska .78 Livingston 1.13 Among the low mortality counties Wayne, Genesee, Oakland, Kent, and Muskegon are predominantly urban, while 1midland, Ottawa, and Menominee have large urban populations. In contrast, the high mortality counties are all predominant- 1y rural, three being in the Upper Peninsula, five in north- ezpn Michigan, and fcur in the southern part of the State. There are probably several factors contributing to this ..v .. \. Ru 3 (1 I) H w .. \I-O N-..» 311w a MN! . DEATH RATES: ACCIDENTS AV. I935 ‘44=.759 EATHS PER lOOO POP 9 UNSER.76 2223 .715 -,35 W ~88 ~.95 m ~96~Los o .0‘ .o’ .«- .. lai’ RV‘QVJNWNE o . LOG-HS .__-,§.:.::,:§:,:$®' c. ’ 0’... 0.0. e . -'-‘6-OVER ‘ - ..°O Figure 7.6 (Source: Table 7.3) 187 distribution. Supervision of motor vehicle traffic is more detailed and speed laws are lower in the city than in the country. It is also probable that the insurance liability of employers has resulted in improved safety devices and campaigns Which have materially reduced accident hazards in industrial plants. Similar incentives are lacking in rural areas. Mining and lumbering, the two chief industries of the Upper Peninsula, are de-centralized occupations which are considered hazardous to life. 5. Pneumonia. The death rates from pneumonia are given in Table 7.4 and are presented graphically in Figure 7.7. Although it is the fifth principal cause of death for the State, it ranks sixth in 42 of the counties and fifth in only 31 of’them. A few highly populated counties with excessive death rates determine the high rate for the State, giving it fifth rather than sixth place as a cause of death. For example, in Wayne County, which has 38.3 percent of the State's pOpulation, the death rate frOm pneumonia is exceeded only by heart diseases and cancer (See Appendix II Table 1). The death rate in Wayne county is .61 as compared with .57 for Michigan. Moreover, death rates from.pneumonia constitute the fourth principal cause of death in Mhskegon, Kalkaska, and Alger counties, their respective rates being markedly higher than.that of Michigan. They are .71, .83, and .76 respectively. The twelve counties with the highest and those with the lowest death rates are listed below: DEATH RATESI PNEUMONIA I935-44= .57I AV. ‘ .r o ...... 171. v 000. 00 o inky ‘ . 4.0. O O 0 .0 00 a!“ . O O O ..0 00000A qddflw .x.... cm 0 0 ‘ . 000...: 0.. 0.. C 010.09.". O. O. r .‘ .... 0.... // / J$v II/ A %% .u a. g V “on.“ .0 .o 7 1 3.“. 3.". . V b. u 0 1"...”7 ”77..uu. . 00. 0. .. v . 00 '00. “”000. D l D o - coo . 0 _ e0 v.0... 0.0 o e mvunmv IOOO POR .uuoen .48 was - .52 w.”- .57 m - n .63-.57 DEATHS PER Table 7.4) ”8“” 707 (souPOC: 188 Table 7.4 CRUDE DEATH RATES FROM PNEUMONIA.AND NEPHRITIS PER 1,000 POPULATION, BY COUNTY, MICHIGAN, AVERAGE 1935-1944 (Based on the 1940 population; by place of residence) , County Pneumonia Nephritii: County PneumoniaNephritis1 (107-109) (loo-152) : (107-109) (130-132) Alcona .432 144 Lake :29. :29. Alger ';Z§ :EZ. Lapeer :52 .62 “108811 047 064 Leelanau fig 070 Alpena .45 .53 Lenawee .66 . Antrim :7: 781 Livingston _._'_7_4_ .229. Arenac .64 756 Luce .54 .47 Baraga E . 6 Mackinac :72 E Barry _0_5_6. 092 Macomb e55 053 Bay .53 .79 Manistee .56 ‘;41 Benzie 4% _._8_5 Marquette '75? .53 Berrien 057 076 18.8011 .44 fl Branch _.;7_§ .88 Mecosta _._"5'_'§ _._9_9_ Calhoun .65 .60 Menominee ‘ng .49 Cass _.__6_g 1.51 Midland .44 1755 Charlevoix _._5_5, _._;7__5_ Missaukee ‘2‘5‘I 4.411. Cheboygan _._§§ :41 Monroe 755 .42 Chippewa .56 .;§_ Mbntcalm. .54 .81 Clare :9: .:.§.§ Montmorency 783' :_8_9_ Clinton :12 1.29 Muskegon 771' .52 Crawford [:53 ,ggp Newaygo .64 “:68 Delta .68 fig Oakland 745 .40 Dickinson $1.3. .3 Oceana .50 5.14 Eaton _._§_:_5_ T81 Ogemaw E .49 Emmet _._5_9_ :24 Ontonagon :2? '._‘_5_5 Genesee .62 .46 Osceola ‘;42 .74 Gladwin .63 .74 Oscoda 4gp 2‘83 Gogebic 76-0 :53 Otsego £7. :86- Grand Traverse-£61 :__9_ Ottawa _.__2_9_ . 6 Gratiot ';59 .75 Presque Isle .;53 ,;42 Hillsdale .66 .75 Roscommon .60 .;65 Houghton 2'51 .63 Saginaw 763 .59 Huron .63 .60 St. Clair .58 .84 Ingham e42 059 St. Joseph 073 1063 Ionia ‘;§§ .62 Sanilac “:58 .65 Iosco _._4_2_ :91 Schoolcraft _._6_1 3'66 Iron .52 ';61 Shiawassee .56 l. Isabella TEST _._4_§_ Tuscola .58 .70 Jackson 'TZS .66 Van Buren .56 .90 Kalamazoo e53 072 Washtenaw 053 047 Kalkaska ,;§§ ‘;4§ Wayne .61 .44 Kent .41 ..52 Wexford .511 1‘15 Keweenaw ‘:42 .ng State .57 .55 1. International List number, 1939. Note: Rates based on less than 200 deaths during ten years are underscored. Annual Report, Michigan Department of Health, 1935- 1944, Numbers 64-74, Table 16. Source: 189 Twelve Lowest Death Twelve Highest Death Rates from Pneumonia Rates from Pneumonia Ottawa .29 Clare .91 Kent .41 Lake ~ .83 Keweenaw .42 Kalkaska .83 Iosco .42 Montmorency .83 Ingham .42 iackinac .77 Mason .44 Alger .76 Midland 04:4 Livingston e74 .Alpena .45 St. Joseph .73 Allegan .47 Brandh .72 Jackson .48 muskegon .71 Dickinson .48 Wexford .71 .Alcona .48 Antrim .71 Coastal counties predominate among the low death rate group. They include Iosco, Alpena, and Alcona on the east coast;.A11egan, Ottawa, and Mason on the west coast; and Keweenaw and Dickinson in the Upper Peninsula. Six of the twelve counties with high rates are in northern Michigan, two are in the Upper Peninsula, and four are in southern.thhigan. The highest death rate from.pneumonia is .91 in Clare and the lowest is .29 in Ottawa county. 6. Nephritis. The sixth principal cause of death in.Michigan is nephritis, the rate being .55 per 1,000 pepulation. This disease is a common cause of death in later life, and it tends to have high rates in counties where other chronic diseases take a large toll. (See Table 7.4 and Figure 7.8) Wayne county's low death rate of .44 forces the rate for the State downward. In Wayne and five other counties, nephritis ranks seventh as the principal cause of death, but in 44 of the counties it ranks fifth or above. Because of Wayne county's low rate nephritis ranks sixth rather than fifth in the State. As might be expected -44?.546 DEATH RATES: NEPHRITIS AV. I935 ac! .. . ' shxx. ... .dciwedkaaw O. ... . 4% ...... .....u.. ...u.».. IOOO POP DE ATHS PER UNDER .45 '///. .45—.55 Table 7.4) “6“" 708 ($UN.’ 190 from their population composition, counties of the Upper Peninsula have low rates from.hephritis. This is indicated by the light cross hatching in Figure 7.7. The twelve highest death rate counties and the twelve lowest are listed below: Twelve Lowest Death Twelve Highest Death Rates from Nephritis Rates from.Nephritis Dickinson .32 Cass 1.31 Baraga .36 Clinton 1.29 Oakland 0 4:0 111 dland l e 09 Monroe .42 St. Joseph 1.02 Presque Isle .42 Mecosta‘ .99 Missaukee .44 Barry .92 Wayne .44 Van Buren .90 Isabella .45 Montmorency .89 Genesee .46 Branch .88 Cheboygan .47 Otsego .86 Washtenaw .47 Crawford .85 Luce e47 B9n216 085 The counties with low death rates are distributed throughout the State in localities in which the younger age groups are fairly well represented. Five of these counties are located in the southern part of the State and are predominantly urban or have large urban populations; four are in the northern part of Michigan below the Straits; and three are in the Upper Peninsula. The twelve counties with high rates for nephritis are all predominantly rural. Six of them.are in the southern part of Michigan and six are in the northern part. No county of the Upper Peninsula is in this group. Midland appears to be exceptional. In.this county nephritis ranks second as the principal cause of death, although for other chronic diseases it has low death rates. 191 7. Tuberculosis. Death rates from.tuberculosis are given in Table 7.5 and in Figure 7.9. It is the seventh principal cause of death in Michigan, the rate being .36 per 1,000 pOpulation. Without Wayne county, it would rank as the ninth principal cause of death in the State, the rate, less Wayne county, being 25.3. Since it is the sixth prin- cipal cause of death in Wayne county, this influences the State rate unfavorably. When counties are ranked from.high to low according to death rates from this cause, Wayne occupies eighth position. The twelve highest and the twelve lowest death rates by county are presented below: Twelve Lowest Death Twelve Highest Death Rates from.Tuberculosis Rates from.Tuberculosis Midland .12 Gogebic .73 Barry. - .12 Houghton .69 Missaukee .14 Ontonagon .59 Gladwin .14 Keweenaw .57 Ingham. .15 Dickinson .57 Hillsdale .15 Baraga .56 Eaton .15 Alger .55 Oscoda .16 Mayne .53 Clinton .16 Menominee .52 Ionia .16 Delta .47 Ottawa .16 Marquette .45 Alcona .16 Iron .44 The low rate counties are concentrated in southern Michigan. In contrast, all of the high rate counties are in the Upper Peninsula except Wayne. The highest rate is .73 in Gogebic and the lowest is .12 in Midland. Since tuberculosis is considered a preventable disease, these excessive rates in a particular area must reflect inadequate TUBERCULOSIS .359 DEATH RATES 1 AV. I935 — 44: l/ .A r//.., , .0000” 0 I]. as. V .4 W0» 10.1.” / / I I I .00000. Q ‘ . v . V ...0.0.. I. v a. O 00 DEATHS PER IOOO POP / x/ 00000.. f.’ I / / v0.0... // / ,1, 0. VI. . 0 0.0. ... . o A .. . . A ‘0‘. .... . v0.00m... v ‘ 00. .23 '//A O . '090’9. ////./o, ”my/2 was m .24-.29 I /// a / / /// /// % .r/JWA n .30- .35 ."O/ O 4. . 0 0 . 10000. N0?» 00000. .00000. v0.00. 00.00. .00.000. .. A ..0.0m0. [I 000 n 0 n... O ’ . Figure 7.9 (Source:- Tabl. 705) .n oi“; . ,I I. I 1 - . u u a q. d I. I q I: I: If- - I III . - I. . u I . \ . RI . I. by firm Pub :1 v We L in . . 0 III‘ “:4 and \ 9‘4 A a I \ PU h. a ‘4 . 0 U “I“ IL n lib IL IL .... «.... a. u and .- n .d halv Rant 9?...“ wring hypo». MIC .HV n..J . ... MM Flu 1m \ a NJ M .u A y. . ...: a: ~u~ 192 Table 7.5 CRUDE DEATH RATES FROM TUBERCULOSIS AND PREMATURE BIRTHS PER 1,000 POPULATION, BY COUNTY, MICHIGAN, AVERAGE 1935-1944 (Based on the 1940 population; by place of residence) County Tuber- Premature: County Tuber- Premature culosis Births : culosis Births (13-22) (159) : (13-22) (159) Alcona :1§_ .:§§ Lake .29 .10 Alger .55 :41 Lapeer E 71‘6- Allegan .71-9' :2_2_ Leelanau :34: 758 Alpena .26 :31. Lenawee .25 '751 Antrim :2_O :_2__ Livingston :19 :30 Arenac :33 :22 Luce :31 _.____Q_ Baraga .56 :29. Mackinac .:§§ .:§g Barry TIE _.__2__ Macomb . 25 . 25 Bay 721 . 55 Mani stee £3 :11; Benzie :3 £3: Marquette .45 .29 Berrien .29 .32 Mason .40 .19 Branch _._1_9_ :29 Me costa E E Calhoun .27 .35 Menominee .52 .35 Cass :23 . 25 Midland 2T2 1‘2‘8 Charlevois _._2_§_ :_——3___8'_ Missaukee E E Cheboygan .:24 .:gg Monroe .25 .30 'Chippewa :28, .37 Eontcalm .20 ':gg Clare :_2_2_ :51 -.Iontmorency :_2_3 :_3_6_ Clinton ‘:lg ‘:26 Muskegon .23 .41 Crawford ‘:21 ‘:§Z Newaygo :;§, .28 Delta _.:4_7_ _._5_l Oakland . 20 TEI Dickinson .57 .:§z Oceana, .18. ‘:14 Eaton 2:5. :22 Dawn '38.. .44.. Emmet “:35 “:32 Ontonagon . 9 ':32 Genesee .26 .34 Osceola '7I9 ,:§_ Gladwin ':14 l:§§ Oscoda . 6 ':24 Gogebic _._'7_5 :gz Otsego TI"? . 2 Grand Traverse:§1 .32 Ottawa . 6 .:§_ Gratiot :18 I:§Q Presque Isle ':31 .27 Hillsdale .:_§ .19 Roscommon .:§Z 'TIZ Houghton . 69 ‘.—2'I Sagi naw . 29 7'55 Huron 022 052 St. Clair 025 027 Ingham 015 030 St. Joseph 022 020 Ionia :16 :_2_O_ Sanilac :33 :22 Iosco .:23 .:gg Schoolcraft ':3§ “:gg Iron :44. ‘:23 Shiawassee .17 .31 Isabella ,:21 .:39 Tuscola .22 .24 Jackson .20 .26 Van Buren .23 .23 JKalamazoo .19 .23 HaShtenaw .26 .25 Kalkaska L§_5_ fi]: Wayne 0 5 3 0 25 Kent 0 21 0 2]. 'I'IeXfOI'd fig fl IKeweenaw .57 ':22 State .36 .27 :1. International List number, 1939. riots: Rates based on less than.200 deaths during ten years are underscored. ESource: Annual Report, Michigan Department of Health, 1935- 1944, Numbers 64-74, Table 16. 195 social planning in the field of health. Franklin Roosevelt in discussing public health conditions in New York while Governor of that State said, "...there is no reason for tuberculosis to be twice as prevalent in some counties as in others;..."1 Vance associates high rates for tuberculosis with areas in which standards of living are low. He says, "Preventable diseases of youth and middle age, like tuber- culosis, influenza, and diseases of infancy are highest in regions of low living standards."2 If a mean standard of living measure3 is computed for the twelve counties with the lowest death rates and for those with the highest, those counties with the highest rates have the lowest standard of living index. The mean index was 102 for counties with low rates and 96 for counties with high rates.4 Thus, these data tend to substantiate the conclusion that there is an inverse relationship between size of death rate from tuber- culosis and the standard of living of the area. 1. Roosevelt, Franklin D., The Governor's Foreword, from Public Health :E New York State, Department of Health, Albany, New York, 1932, p. 7. 2. Vance, Rupert B.,'gp. cit., p. 341. 3. The standard of living measures employed here are those used in.Hospita1 Resources and Needs, Michigan Hospital Survey,.2p. cit., Table 13, p. 66. ‘4. The standard of living measures for the counties with the twelve lowest rates for tuberculosis are: Midland 103, Barry 107, Missaukee 91, Gladwin 91, Ingham 116, Hillsdale 104, Eaton 108, Oscoda 87, Clinton 108, Ionia 107, Ottawa 114, and Alcona 90. Those for the counties with the twelve highest rates are: Gogebic 98, Houghton 98, Ontonagon 90, {eweenaw 89, Dickinson 101, Baraga 88, Alger 91, Wayne 112, Menominee 97, Delta 98, Marquette 98, and Iron 94. 194 8. Premature Births. Table 7.5 gives the ten-year average annual number of deaths from premature births per 1,000 pOpulation for each county and in Figure 7.10 they are presented graphically. Premature birth is the eighth principal cause of death for the State, and for 25 of the 83 counties. The death rate for the State is .27. In 23 counties, this is the seventh principal cause of death while in one county, Missaukee (which has the highest death rate in Michigan from premature births),it ranks sixth. It is interesting to observe that in the belt of counties running from Genesee to Presque Isle in which death rates from chronic diseases are ordinarily low, rates from premature births are higher than the State average. This is shown by the cross hatching in Figure 7.10. The counties that have the twelve lowest and those that have the twelve highest rates are as follows: Twelve Lowest Death Rates Twelve Highest Death Rates from.Premature Births from.Premature Births Lake .10 Missaukee .45 Roscommon .14 Ogemaw _ .44 Oceana .14 Muskegon . .41 Lapeer . .16 Alger .41 Mason .19 Luce .40 Hillsdale .19 Mackinac , .39 Ionia .20 Cheboygan .39 Ottawa .20 Charlevoix .38 St. Joseph .20 Chippewa _ .37 Houghton .21 Crawford .37 Kent .21 iMontmorency .36 Sanilac .22 Wexford , .36 In general, the high.rates from.premature births are found in the northern part of lower Michigan and in the eastern part of the Upper Peninsula. Muskegon is the only DEATH RATES: PREMATURE AV. I935—44=.27O '0 . O‘. 0000. 0203030202. ’0'0'0 9.0.0 4 3 ?0?0 V DEATHS PE R UNDER .20 "/1. .20 -.23 I000 POP v v v 0 0 01 $0201 " .0‘ '20:: L9... .0 . L... ”I 0 0'0’0’0 0’ ///b 0 5‘0‘9‘.’ 0 r 0' A ”a A ”gun 7.10 (Source: Table 7.5) BIRTHS 195 county in the southern half of the State listed in this group. On the other hand, seven of the counties with the twelve lowest death rates are located in.the southern part of the State and only one, Houghton, is found in the Upper Peninsula. Some authoritiesl consider the infant death rate an excellent measure of the general level of living of a com- munity and since the death rate from premature births is the largest single item.in.this group, one might assume an in- verse relationship between it and level of living indices. Eniploying the standard of living measuresz mentioned pre- viously, there is some evidence in support of this generali- zation. The mean death rate for the counties with the twelve lowest rates is .18 and the mean of the standard of living measures for the same counties is 102. In contrast, the mean death rate for the counties with.the twelve highest rates is .39 and the Dean of the standard of living measures 18 9303 1. Smith, T. Lynn, 2p. cit., p. 168; and Tate, Leland B., _020 Gite, P0 100 2. The-standard of living measures employed here are those given in Hospital figsources and Needs, Midhigan Hospital Survey, pp: g:§., Table 13, p. 66. For a list of the items included in the measure, see Footnote 1, p.177of this Chapter. 3. The standard of living measures for the counties with the twelve lowest death rates from premature births are: Lake 86, Roscommon 94, Oceana 95, Lapeer 102, Mason 103, Hills- dale 104, Ionia 107, Ottawa 114, St. Joseph 108, Houghton 98, Kent 114, and Sanilac 99. Those for counties with the twelve highest rates from premature births are: Missaukee 91, Ogemaw 86, Muskegon 108, Alger 91, Luce 85, Madkinac 87, Cheboygan 94, Charlevoix 99, Chippewa 96, Crawford 92, Montmorency 83, and Wexford 101. 196 9. Diabetes mellitus. The distribution of death rates from diabetes mellitus is shown in Table 7.6 and in Figure 7.11. The death rate for the State from this cause is .26. There are 20 counties in which diabetes ranks seventh rather-than.ninth as a principal cause of death. Some of these counties suCh as Branch, Cass, and Allegan, have large prOportions of their numbers above 65 years of age, but others, for example fiiose in the Upper Peninsula, do not have old aged pOpulations. Diabetes, however, is not considered primarily a disease of old age, but may be expected in any age group. Dr. Bortz states, "Diabetes is most likely to appear between the ages of fifty and sixty. Following sixty, statistics Show that there is a decrease in the chances for develOping diabetes." But he also makes the following observation, "In the decade past, the mortal- ity rate from diabetes in the group under forty-five years of age has gradually decreased, while it has steadily in- creased in the age group of sixty-five years or more."1 Twelve Lowest Death Rates Twelve Highest Death Rates from Diabetes Mellitus from.Diabetes Mellitus Otsego .lO Keweenaw .45 Kalkaska .12 Mecosta .44 Luce .13 Manistee .40 Baraga 015 Houghton . 040 Oscoda .16 Crawford .40 IOSCO 018 Ste Joseph 038 Genesee .18 St. Clair .38 Midland 018 Lake ‘ 038 Oakland .19 Oceana .37 Ontonagon .20 ienominee .37 Presque Isle .20 Shiawassee .36 Gladwin .21 Wexford .36 l. Bortz, Edward L., "Diabetes Mellitus," from Stieglitz, Edward J".22' Cit0, pp. 226'2270 v. . ..s 0.0. 0 0 AV. I935‘44 = .26I DEATH RATES: DIABETES P. .. . 00. ~05... // UNDER .2I DEATHS PER ' Iooo PO 1 ‘ .u 7 0MOMOM0“ 0 0.0.0.. .0 .0 .00.. .. O 0 0.. .3.” ’0 O ... «0M0. .. 4 4 00000 .0 0 lav. . . / I Q 9 . M .2! - .25 w .25 -.3o a .3: - .35 Figure 7.11 (Scum: $010 7.6) 197 Table 7.6 CRUDE DEATH RATES FROM DIABETES MELLITUS AND ARTERIOSCLEROSIS PER 1,000 POPULATION, BY COUNTY, MICHIGAN, AVERAGE 1935-1944 (Based on the 194g2population; by place of residence) County Diabetes Arterio- : County Diabetes Arterio- scleroiis : scleroiis (61) (97) : (61) (97) Alcona :22 :12 Lake :22 :22 Alger .:2§ .:2§ Lapeer .34 :22 Allegan .34 :22 Leelanau .33 .:22 Alpena ':2_ :22_ Ienawee .36 .41 Antrim .23 :22 Livingston .24 :24 Arenac .LEZ :44. Luce .13 .07 Baraga :12 :14 Mackinac :24 :2 Barry :_3_1_ :22 Macomb . 25 .09 Bay 03.]. £322 1::53181339 040 04:1 Benzie :52 :52 Liarquette .55 F2 Berrien .32 .20 Mason ,:22 .;§Z Branch :22 _._2_2 liecosta :_4__ :51 Calhoun 0 27 0 25 Menominee 0 37 0 lg Cass . 55 :21 Midland :1}; . O Charlevoix _._2_8_ :22 1.115 saukee :22 TOT Cheboygan ‘:22 .:21 Monroe .23 '7I7 Chippewa :24, [:22 Montcalm. :22 .47 Clare “:22 .:22 Montmorency .23 'TTE Clinton ':22 .:42 Muskegon .28 . 6 Crawford :42 :22. Newaygo .30 :22 Delta :32 .16 Oakland TIC .16 Dickinson .:22 ‘:22 Oceana .:22 .57 Eaton :22 .:_2 Ogemaw .26 ‘TIE Emmet .28 ,:22 Ontonagon 'TEO '7I1 Genesee .18 .22 Osceola 'TEI .32 Gladwin :2; .15 Oscoda TIE TIE Gogebic _._2_5_ 14 Otsego :_ E Grand Traverse.32 ,:22 Ottawa :22 .24 Gratiot ‘:22 .:22 Presque Isle :22 'TIO Hillsdale :32 :62 Ros common . 22 :_6_ Houghton .40 .15 Saginaw 725 .24 Huron 024 050 St. Clair 038 024 Ingham. .25 .16 St. Joseph .38 .52 Ionia ,:§2 ':42 Sanilac .27 .:;2 Iosco .18 .:22 Schoolcraft ‘:22 ':22 Iron 3 :22 Shiawas see .36 .42 Isabella ,:22 ,:42 Tuscola .29 .19 Jackson .31 .33 Van Buren .29 .30 Kalamazoo 025 025 I’I'ashtenaw 024: 041 Kalkaska .:;2, ':22 Wayne .23 .09 Kent .29 .28 Wexford :2§_ .15 Keweenaw _gg; .229 State .26 7T9 1. International List number, 1939. Note: are underscored. Source: 1944, Numbers 64-74, Table 16. Rates based on less than 200 deaths during ten years Annual Report, MiChigan Department of Health, 1935- 198 The counties with the twelve lowest and the twelve highest death rates are given on the preceding page. Three Upper Peninsula counties are among those with low rates, but three other counties in the same area are also listed among those with high rates. Six of the high mortal- ity counties, namely Oceana, Keweenaw, manistee, Houghton, St. Clair, and Menominee are on the coast. The highest rate is .45 in Keweenaw and the lowest is .10 in Otsego county. 10. Arteriosclerosis. Death rates from.this cause are given in Table 7.6. The rate for the State is only .19 per 1,000 pepulation. Since some of the county rates are small, they are given per 10,000 pOpulation. The counties with the twelve lowest rates and those with the twelve highest are as follows: Twelve Lowest Death Rates Twelve Highest Death Rates from.Arteriosclerosis from Arteriosclerosis per 10,000 pepulation per 10,000 population Iron .64 Hillsdale 5.98 Luce .67 mecosta 5.74 Schoolcraft .74 Oceana 5.67 Missaukee .75 Livingston 5.42 Lake . 83 Ben zie 5 . 26 Wayne .86 St. Joseph 5.20 Macomb .91 Ionia 4.87 Presque Isle .98 Montcalm. 4.65 Midland 1.00 Clinton 4.24 Arenas 1.08 Shiawassee 4.22 Ontonagon 1.14 Isabella 4.16 Leelanau 1.19 manistee 4.12 Seven of the counties with high rates are in southern Michigan, and the remainder are in the northern half of lower Michigan. No Upper Peninsula county is among the high rate group. In contrast, four or the low mortality counties are in the Upper Peninsula, namely, Iron, Luce, Schoolcraft, and 199 Ontonagon. Six of the remaining counties are in northern Michigan, and two are the industrial or semi-industrial counties of Wayne and Macomb. The highest rate is 5.98 in Hillsdale county and the lowest is .64 in Iron. 11. Suicide. The incidence of death from suicide is given per 10,000 pOpulation in Table 7.7. The average annual rate for the State is 1.2. In five counties, Baraga, Lake, Otsego, Hillsdale, and Oceana, suicide death rates exceed those of some of the ten principal causes, although for the State suicide ranks much lower. Counties in Which suicide ranks ninth or tenth also have high accident death rates as indicated by the following: (See Appendix II, Table 1) Rank 3: Accident Baraga 2 Hillsdale 4 lake 4 Luce 2 Alger 2 Arenac 2 The twelve highest and the twelve lowest death rates from suicide, by county, are given below: Twelve Lowest Death Rates Twelve Highest Death Rates from.Suicide per 10,000 from Suicide per 10,000 Kalkaska .19 Baraga 2.67 Manistee .87 Keweenaw 2.50 Osceola .90 Case 2.24 Presque Isle .90 Ontonagon 2.20 Antrim .91 Alcona 2.20 Luce .94 Sanilac 2.11 Emmet .95 Lenawee 2.11 Ottawa .96 Lake 2.08 Kent 097 Eaton 2008 Muskegon .98 Van Buren 2.08 Midland 1.00 Hillsdale 2.06 Missaukee 1.00 Calhoun 1.93 200 Table 7.7 CRUDE DEATH RATES FROM SUICIDE PER 1,000 POPU- LATION, BY COUNTY, MICHIGAN, AVERAGE 1935-1944 (Based on the 1940 pOpulation; by place of residence) County Suicide 1 : County Suicide (163-164) : (163-164)1 A10 one 0 22 Lake 0 21 Alger ‘71: Lapeer ‘TTI Allegan E Leelanau . 12 Alpena .12 Lenawee ‘721 Ant rim _._"_6_"9_ Livingst on 719 Arenac :12 .Lqu :59 Baraga .27 Mackinac TIE Barry 71's Ma comb TE Bay _T_I_I Menistee TOO Benzie :12 Marquette TF7— Berrien :12 Mason :15- Branch .19 mecosta 'TIS Calhoun :2 Menominee TE Cass ‘:gg hfldland TIC Charlevois :11 Missaukee ‘TIO Cheboygan :11 Monr oe 71'4- Chippewa :12 Montcahm ‘TIO Clare :15: Mont morency 713 Clinton :19_ MMskegon ‘710 Crawford .19 Newaygo '715 Delta '.'_‘1‘_5_ Oakland TI'I Dickinson .:1§ Oceana .18 Eat on :21 Ogemaw E Emmet ‘:Qg Ontonagon .22 Genesee .ll Osceola .09 Gladwin ‘:14 Oscoda '71? Gogebic ‘:1§ Otsego ‘71? Grand Traverse :l_§ Ottawa 2E Gratiot :14 Presque Isle .09 Hillsdale .21 Roscommon 'TIE Houghton .717 Saginaw . 'TIO Huron E St . Clair 717' Ingham .12 St. Joseph~ :12 Ionia :5 Sanil so 721 Iosco .12 Schoolcraft (712 Iron :12 Shiawassee .713 Isabella .:1Q Tuscola .13 Jackson .15 Van Buren .21 Kalamazoo '714 Washtenaw .16 Kalkaska '702 Wayne .11 Kent 'TIO Wexford .14 Keweenaw .Laé State ° 1. International List number, 1939 . Note: Rates based on less than 200 deaths during ten years are underscored. Source: Annual Report, Michigan Department of Health, 1935- 1944, Numbers 64-74, Table 16. 201 Of the twelve low rate counties, eight, Kalkaska, ianistee, Osceola, Presque Isle, Antrim, Emmet, Midland, and Missaukee are in northern Michigan; one is in the Upper Peninsula, namely, Luce; and the remaining three are Ottawa, muskegon, and Kent in southwestern.thhigan. The lowest death rate tram suicide is .19 in Kalkaska county and the highest rate is 2.67 in Baraga. Seven of the twelve counties with high rates from.suicide are rural counties in southern Michigan. These counties are characterized also by high death rates from.chronic diseases. These data support the findings of the MetrOpolitan Life Insurance Company reported by Byrd. They found that "among males ill health ranked as the lead- ing motive, accounting for nearly 40 percent of all attempts at suicide, whether successful or not."1 It will be re- called that suicide in.Michigan is essentially a male White behavior pattern.2 Under these circumstances one would ex- pect a high incidence of suicide in these rural counties. Infant Mortality gates The infant death rates are given in Table 7.8 and Figure 7.12. In this Table, the average (1935-1944) annual number of deaths under one year for each county is related to the annual number of live births. Ten-year average l. Byrd Oliver E. ‘32. cit., p. 197, from.metropolitan Life,Insurance Company, "Why Do People Kill Theme selves?" Statistical Bulletin, 26, No. 2, February 1945, pp. 9'10. 2. See Chapter'V,p31270f this dissertation. INFANT ”RTALI TY RATES (by place of residence) neurone, rs- xm AVE, 1935-1944 : 42.0 -,. . I I f\ .' . O ." “n . ‘7' O O D O ( f‘= . _, ”Axial 1MEM..% J o 300; 6’, '9...» O 'f, I I ".49-“d3 )- J' g m A MM): \ ! '0 O v v i DEATHS PER 1000 0 ' 'o'o'o’ 0 LIVE BIRTHS UNDER 33,0 1")”. .0: .. . we «as 38 - 1.2.9 .0 e 500:0}; 0M “3 ' ”-9 (54 ; - M #8-529 ”’QQ' b“ ' V vvvy v v r H! 53 - 57.9 /‘.o’ ’ ’ 3?;5ffié’ ,’ ' ‘v ‘r v v ‘ - v - 58 - OVER 1,._.. ’ 15.1"“: . 0'” Figure 7.12 (Source: Table 7}8} 20 N Table 7.8 INFANT MORTALITY RATES PER 1,000 LIVE BIRTHS IN COULTIJES OF MICHIGAN, TEIN YEAR AVERAGES, 1935-1944 (Exclusive of stillbirths. By place of residence) State and : : State and : County : Rate : County : Rate Alcona 54.56 Lake 49.57 Alger 57. 4 Lapeer 43745 Allegan 41.33 Leelanau 56.89 Alpena 44.16 Ienawee '40707 Antrim. 48.67 Livingston 42.54 Arenac 48.50 Luce 67. 6 Baraga 56.01 Mackinac '52748 Barry 35.26 mecomb 43.14 Bay 47.07 Manistee 46.71 Benzie 55.86 Marquette 49. Berrien 41.53 Mason 37.91 Branch 43.03 Mecosta 4 .69 Calhoun 47.20 Menominee 40.55 Cass 53.66 Midland 42.06 Charlevoix 52091 IflSSfiUkSG 55091 Cheboygan 53.62 Ionroe 44.66 Chippewa 53.91 Montcalm 53.52 Clare 49.33 Montmorency 60.76 Clinton 39.38 Muskegon 52.86 Crawford 68.24 Newaygo 47.13 Delta 59.60 Oakland 40.50 Dickinson 45.37 Oceana 47.48 Eaton 37033 Ogemaw 5 e65 Emmet 48.97 Ontonagon 59.55 Genesee 43.54 Osceola 49.62 Gladwin 46.90 Oscoda 46.24 Gogebic 45.96 Otsego 49. 2 Grand Traverse 46.23 Ottawa 30.37 Gratiot 51.20 Presque Isle 44.33 Hillsdale 39.90 Roscommon 46. 2 Houghton 51.87 Saginaw 44.56 Huron 51.22 St. Clair 47.74 Ingham. 41.11 St. Joseph 32.61 Ionia 42.66 Sanilac 45.80 Iosco 31.61 Schoolcraft 52.21 Iron 39. Shiawassee 47.99 Isabella 54.44 Tuscola 46.37 Jackson 45.77 Van Buren 44.68 Kalamazoo 34.06 WaShtenaw 39.93 Kalkaska 37.49 Wayne 38.45 Kent 35.02 Wexford 53.98 Keweenaw 56.91 State 42.02 1. Ten year average births and deaths. Source: of Health, 1955-1944, Table 8. Note: are underscored. Computed from.the Reports of the Michigan Department Rates based on less than 200 deaths during ten years 203 infant death rates are not as liable to be influenced undue- ly by unusual conditions in a county as are annual rates. This is important in small counties where infant death rates are based on a small number of deaths. For example in Oscoda county, only 24 deaths occurred during the ten year period. Observation of Figure 7.12 shows that below the Bayhmuskegon line, only six counties have infant death rates in the upper three frequencies, 48 and above, while above the line there are 18 counties with rates in this range. The twelve counties with the highest and the twelve with the lowest infant death rates are given below: Twelve Lowest Death Rates Twelve Highest Death Rates from Infant Deaths from.Infant Deaths Ottawa 30.37 Crawford 68.24 10300 31.61 Luce 67.06 St. Joseph 32.61 Montmorency ' 60.76 Kalamazoo 34.06 Delta 59.60 Kent 35.02 Ontonagon 59.55 Barry 35.26 Alger 57.14 Eaton 37.33 Keweenaw 56.91 Kalkaska 37.49 Leelanau 56.89 Mason 37.91 Baraga 56.01 Wayne 38.45 Missaukee 55.91 Iron 39.18 Benzie 55.86 Clinton 39.38 Alcona 54.56 Among the high rate counties six are in the Upper Peninsula, and the remaining six are in the upper part of the Lower Peninsula. Seven of the twelve counties with low death rates for infants, namely Ottawa, Kent, St. Joseph, Kalamazoo, Barry, Eaton, and Clinton, are concentrated in the southwestern part of Michigan. The remaining counties in this group in- clude three counties in the northern part of the Lower 204 Peninsula, Iron in the Upper Peninsula, and Wayne county. The lowest infant death rate is 30.37 in Ottawa and the highest is 68.24 in Crawford county. As was true for tuberculosis and premature births, variations in the infant death rate tend to be associated with levels of living. In the twelve counties with low rates the mean of the standard of living measures1 is 108 whereas the mean for counties with high rates is 91. 1. The standard of living measures for the counties with low death rates are: Ottawa 114, Iosco 97, St. Joseph 108, Kalamazoo 113, Kent 114, Barry 107, Eaton 108, Kalkaska 82, Mason 103, Wayne 112, Iron 94, Clinton 108. Those for counties with high death rates are: Crawford 92, Luce 85, Montmorency 83, Delta 98, Ontonagon 90, Alger 91, Keweenaw 89, Leelanau 96, Baraga 88, hissaukee 91, Benzie 102, and Alcona 90. See Hospital Resources and Needs, Nichigan,50spital Survey, Table l3,p.66. CHAPTER‘VIII mom-ms! nm 13! occurs: (a NTINUID) useless-run vmuxo - ,EEEQSHBIILSQEIWEIEEB£522”! The ten-year average crude death rates8 in rural and urban areas of the State are respectively 10.48 and 9.88, their difference being .60 per 1,000 population. The difference between the crude death rates in the same areas for the year 1940 is 1.6, the rural and urban rates being 11.0 and 9.4 respectively. Similarly, in the coune ties, corresponding differences between teneyear averages l. The term.'residenee' refers to the rural and urban classes of pepulation as defined by the United States Bureau of the Census. “Rural" includes the pepulation living outside of incorporated places of Less than 8,600, while “urban? designates all population living within incorporated places of 2,500 population or mere. 2. These are crude resident death rates based on.deaths which have‘been allocated to place of residence. Deaths classified by place of occurrence are usually referred to as “recorded“, but these are not employed in this analysis. InlMiohigan, deaths have been allocated to place of residence since 1933. 3. In some of the counties of Michigan, there are towns whose populations increased to 2,500 prior to 1940 which have been granted IIII incorporation charters; these were accordingly classified as urban places by the united States Bureau of the Census in April 1940. The 1940 Annual Report of the MHchigan State Department of Health classified the deaths occurring in these particular counties during the calendar year of 1940 as rural or urban according to the United States Census of 1930, because 1940 census figures were not available when the report was issued. The death rates of counties by residence in this chapter, being ten-year averages 1935- 1944, also define the rural and urban pepulation of each county according to the United States Census of 1930. Counties which have urban places in 1940 which were not so classified in 1930 are indicated. 205 206 are likely to be less than for a single year. An accident taking the lives of five or six people, an epidemic, or an infection in the maternity ward of a single hospital re- sulting in several deaths, may materially influence the total annual death rate or those for specific causes in a given area, particularly, where the number of deaths under normal circumstances is small. The ten-year average death rate tends to smooth out such variations, and it is used throughout this chapter. M11. Mortggty mg. The death rate for the rural segment of each county is presented in Table 8.1 and is shown graphically in Figure 8.1. Over half, or 56 percent of the 83 counties, have death rates above 11.00. These counties, for the most part, lie in the western half of the State or along the southern border. Less than half of the counties in the Upper Peninsula have rates as high as 11.0 and in no case do they exceed 11.9. There is a belt of low rates, 9.9 or below, in the castern’part of the State running from Monroe county to Gladwin, and a similar pocket of low rates in the western part of the State comprising Muskegon, Ottawa, and Kent. These are, for the most part, predominantly urban counties. 'me twelve highest and the twelve lowest death rates are given immediately below: RURAL DEATH RATES : ALL CAUSES IICHIGAN Av., 1935-1944 a 10.5 ‘\ v ‘ ‘ g ,r I r, a ,/ N naco 1,4 *3 C etc aleoa I’lmos. I I l I / I e, - A [I / ’0: s~ w ALCIR ua ama CN-P' /,/ / , / I / I h \\ ‘ - c ' Q“ 7 ,‘ I I ~ 9 ,1 8 s? J / r. \ o e .1 6 , " a . m 00! m «— a. I -- J J I gs: , \ .4/3/ '. - ANYmu 073160 sou-(v A Pen 1 {Luau / /" /¢/ . l/ I \ I/l/ | OCN I YMVIIS ulmn CMV'OI CSCOOA " I ,1 W37" vmoa Is cosmos 0611:». If” “i t | . I 11‘ a a u L“: escrow Ciel»: cu m . {V DaATHS PER 1,000 POP. ) 0,, g 0., I \ /, U l ’ I, ":_V / / //// m, " ocuu utcosu menu "0“” a name TI IJSIICOI ILAG “no? ' I/// I/// ‘ / I/ll l. - ccmsu 1"“ 5' c ‘ ‘ OWN“ nun tON'A ch'on w as I l/ / / ”‘0',“ ; // // [1”1’ /// ’1 ll, ' //////z /| L a Iv A 'Ncfifll Lmugg'os OAKLAND /, / III III [I [I ’I// VAN FALA- IE HAZOO CALH39N 410,50” vuuruaw wavuz ’/ /' SY “In!!! Case J r as / . - __ on u s_~_CH mum “mm" C - — - Figure 8.1 (Souroezsfable 8.1) 207 Table 8.1 RURAL MORTALITY RATES FROM ALL CAUSES OF DEATH PER 1,000 POPULATION, MICHIGAN COUNTIES, 1935- 1944. (All rates are ten-year averages; based on the 1940 population; by place of residence) State and Rural Death : State and Rural Death County Rate : County Rate Alcona 11.64 Lake 13.34 Alger 11.31 Lapeer 10.96 Allegan 11.93 Leelana 11.23 Al ena 9.03 Lenawee 12.98 An rim 12.78 Livingston 12.07 Arenae 18.90 Luool , 10.25 Baragel 11.67 Mackinicl 11.03 Barry 12.64 Macomb 8.76 Ba 9.35 Menistee 12.80 Benzie 13.78 Marquette 9.99 Berrien 11.11 Mason 11.41 Branch 13.37 Mecosta 18.79 Calhoun 10.08 Menominee 9.27 Cass 13.96 Midland 8.37 Charlevoix 13.39 Missaukee 10.37 Cheboygan 10 .7 9 Monroe 9 .70 Chippewa 10.87 Montcalm 13.99 Clare 13.08 Montmorency 10.63 Clinton 11.03 Muskegon 9 . 84 Crawford 12.01 Newaygo 11.87 Delta 10.66 Oakland 8.26 Dickinson 9 .30 Oceana 13.67 Eaton 11.66 Ogemaw 11.63 llnnnet 18.37 Ontonagon 11.34 Genesee 8.08 Osceola 13.33 Gladwin 10.66 Oscoda 8.93 Gogebic 11.96 Otsego 11.09 Grand Tiaverse 18.70 Ottawa 9.41 Gratiot 11.99 Presque Isle 9.46 Hillsdale 13.63 Roscommon 10.22 Hough on 11.90 Saginaw 9.36 Huron 10.82 St. Clair 11.17 Ingham 9.63 St. Joseph _ 14.61 Ionia 12.00 Sanilac 11.98 Iosco 10.97 Schoolcraft 9.68 Iron 8.11 Shiawassee 12.08 Isabella 10.58 Tuscola 11.48 Jackson 9.33 Van Buren 14.39 Kalamazoo 9 . 31 Washtinaw 10 .38 Kalkaska 11.34 Wayne 6.81 Kent 9.27 Wexford 12.87 Keweenaw 12.19 State 10.48 l._ Residence according to the 1930 population. Source: nua afigts, Michigan Department of Health, 1935. 9 ’ T‘ O 208 Twelve Lowest Death Rates Twelve Highest Death Rates Rural: All Causes Rural: All Causes um. 3.2.1 . Ste Joseph 14e61 Gene sec 8.08 Van Buren 14.39 Iron 8.11 Montcalm 13.99 Oakland 8.26 Cass , 13.96 Midland 8.37 Benzie ’13.78 Macomb 8. 76 Oceana 13.67 Oscoda 8.93 Hillsdale 13. 63 Alpena 9 .03 Branch 13.37 Kent 9 .27 Charlevoix 13.39 Menominee 9 .27 Osceola 13 .33 Dickinson 9 . 30 Lake 13. 34 Kalamazoo 9 .31 Clare 13.08 From the above data, it may be concluded that the lowest rural death rates are found chiefly in the highly urbanized counties, while the highest rural rates are, by and large, in the wholly or predominantly rural counties of the State.1 Urban Mortality Rates. given in Table 8.8 and are presented graphically in Figure The urban death rates are 8.2. These rates are generally higher than the rural rates, 73 percent of the 36 counties with urban populations having rates above 11.0. The highest urban rates are concentrated in southern Michigan and include the best agricultural counties, whereas the lowest urban rates are for the most part in the industrial areas. _1___ Rates for the Upper Peninsula 1. Insofar as crude rates are concerned, this corroborates the findings reported by Leonie who says, I'According to the Michigan State College surveys, the more rural the county or area the higher the death rate..." ( Loomis, Charles P., “Measuring Rural Medical Needs," First Annu ichigan Ru a1 Health Conference, Michigan Rate 0 age, 31 Michigan, I947 20. If rates are adjusted for age, however the picture may be different. See Chapter IV, Tab1e4.l p. 68 for State crude and age-adjusted rates by residence. " r. URBAN DEATH RATES ALL CAUSES MICHIGAN Av., 1935-1944 : 9.9 “v11“ n \ \\ ONT NAGON " /~ 0 Seneca H ‘ OUCTT 5c L ‘ LUCC MINSON uacwmac CHIPPEWA r . 03“, \\\ ._ . ) \\ Rx / 3 1" ‘ x _ / net 8 Q f j .2— l I ’/ I , o r 9 f“ /I//, I x . C'(.° "! ISL‘ (‘lé / /’ ,w - ’2’3/ ’ nos?- Auvnm 075100 sou-CV ~ Ltumsu 1‘ r A ) sum must Mus“ cuvroso oscoaa ALCO’M L! / I ‘ r was": v s: ‘0’- 061 w DMTHS PER 1,000 POP. (,1 (ONOII “Zap.“ I / / ' ‘ , t 7'; \ a uasou um: osctou c n: cum-N ., \ )I I I . ”NW-R 9.0 L I v/ ‘ ‘ "\ . . I * —.so~ ’/ I’ll -:- \ the .. ‘ OCU A (t t m gAH / 9.0 - 9.9 . I 0 cars I3 Lu | ‘ utmvco Tu: ‘ a m 10.0 - 10.9 ~ m, I/I’. . ’/ VI \ clu 511 “’n‘ ‘ a 11 e O C 11 .9 ' “(a KIN? yon-A CLINYON wasst'. / 4”” ”00;. . [/l’// I, I m ‘ I’l’ 1’ I”, N 12.0 - 12.9 ) ALLIGAH Daslv [AYON was; kmeas' 0“~“"° / C" ’ l,’ V A. - 13-0 - OVER .. - I HAIOO CALHOUN JACKSON \- In- eA'vnl )1 - ‘ is no URBAN POP. Ev ./ 4 I cue JOSI'N 60.:ch " mum: urn-u . O — - Figure 8.2. (Source: Table 8.2) 209 Table 8 . 2 URBAN MORTALITY RATES FROM ALL CAUSES OF DEATH .2138 1 , 000 POPULATION, MICHIGAN COUNTIE , 19 35- 1944 . (All rates are to n-year averages; based on the 1940 population; by place or residence) St ate and Urban Death : State and Urban Death County Rate a County Rate 110 mm --- 1.31:. mm Alger ll .05 Lapeer 11. 89 Alle gan 13 .80 Laelana --- Alpena 11. 24 Lenawee . 13. 89 Antrim --- Livingston 16 .40 Area ac --- Luc e1 --.. Baraga]- --- Mackin c1 --- Barry 15.00 Macomb 8 e 88 Bay 11.88 Manistee 14.00 Benzie --- Marquette 11 .46 Berrien 12.12 Mason . 15.08 Branch 17.15 Mecosta 15.54 Calhoun 15.58 Menominee 15.58 Case 17 .88 Midland 8.50 Charlevoix 12 .88 Missaukee «- Cheboygan 14.21 Monroe 11.10 ChipPO'l 11 e“ MOntOm 13e93 Clare ’ --- Montmoren ey . -s-- Clinton 14 .52 Muskegon .10 . 59 Craer crd --- Newayg «- Delta 11.42 Oakland 8.09 Dickinson 9 . 56 Oceana --- Eat on 15 . 13 Ogemav --- anet 14.19 Ontonagon --- Genesee 8 .87 Osceola «- Gladwin --- Oscoda --- Gogebic 10.16 Otsego --- Grand Traverse 10 .80 Ottawa 10.01 Gratiot 12.85 Presque Isle . 8.82 Hillsdale 14 .14 Bee comon --- Houghton 14.01 Saginaw 11 . 01 Huron --- St . Clair 12.80 Ingham 9.78 St . Joseph 12.04 Ionia 15. 82 Sani lac --- Iosco --- Schoolcraft 12.17 Iron 11 .01 Shiawassee 12. 26 Isabella 10e57 13.180018 18e66 Jackson 13. 07 Van Buren 15 . 95 Kalamazoo 12.07 Wash tin at 11. 26 Kalkaska --- Wayne 9 .01 Kent 10 . 94 lexrord _12. 20 Ke'eenaw --- State 9 . 88 1. Residence according to the 1930 pepulation. Source: Agnufi Re orts, Michigan Department or Health, 911‘??— able 16 . 210 counties tall in the middle ranges, except for Houghton and Mbnominee where they are high. The distribution or urban death rates at both extremes are given immediately below: Twelve Lowest Death Rates Twelve Highest Death Rates Urban: All Causes Urban: A11 Causes Oakland 8.09 Tuscola 18.66 Midland 8e30 08.88 17e58 Genesee 8.67 Branch ~ 17.13 Presque Isle 8.82 Livingston 16.40 macomb 8.88 van Buren 15.95 Wayne 9.01 Mecosta 15.34 Dickinson 9.56 Eaton 15.13 Ingham. 9.78 Barry 15.00 MUskegon 10.39 Clinton 14.52 Isabella 10.57 Cheboygan 14.21 Grand Traverse 10.80 Emmet 14.19 Kent 10.94 Hillsdale 14.14 or the twelve counties with low rates, nine have large urban centers and three, Isabella, Midland, and Presque Isle, are rural counties with small urban pOpulations. On.the other hand, all of the twelve counties with high rates are predomi- nantly rural, having small rather than large cities. gm _gf_ 29.3.3.1}. _i_._n_ M M: Non-communicable .Although one might expect the mortality rates from.heart diseases, cancer, apoplexy, and nephritis to follow rather uni- formly the age characteristics or the counties, there are in- teresting variations from.this expected pattern. From.the data given in Table 8.3 and presented graphically in Figures 8.3 to 8.7, the rural mortality from the five leading causes1 are sum- marized in the following paragraphs. 1. Pneumonia is not discussed in Chapter VIII, although its rate is slightly higher than.that or nephritis. This ap- pears to be Justified in view or the conclusion in previous chapters, namely, that chronic diseases are the mmJor’health problems or the State. RURAL DEATH RATES : HEART IICHIGAN Av., 1935-1944 : 29.9 (LE: ‘;: :.:.'f.: ‘} SN \ .1 .‘ fih‘ \:N ms... "in... O O./// ’0‘. ‘SCOD‘ ~ ’1'“. e :1; :z‘. STATHS Pas 10,000 per. unnsa 25.0 2Seo ’ 28e9 29.0 - 32.9 36.9 40.9 33.0 37.0 41.0 " om 7 ‘ J SID-n _lAPeCN‘ Figure 8.3 (Source: Table 8.3) DEATHS PER 10,000 POP. UNDER 10.2 11.4 12.6 13.8 15.0 HUHAL DEATH HATES : CANCER NICHICAN Av., 1935 - 1944: 11.4 10.2 11.3 12.5 . 13e7 f . 0 . . I 4m. Lam-n, mr‘.‘ aeg‘ , I / O , 14.9 . ’o' m, '14”: (gm 1 I...‘ . 'l 01‘." ‘ (‘ time 8.4 (Source: Table 8.3) RURAL DEATH RATES : ATCPLExx IICHIGAN Av., 1935-1944 = 10.5 . C - v v W ......S ‘. UNDER 6.6 6.6 - 8.5 8.6 - 10.5 10.6 - 12.5 12.6 - 14.5 14.6 - OVER Figure 8.5 (Source: Table 8.3) ACCIDENTS RURAL DEATH RATES IICHIGAN AV., 1935-1944 I 9.0 ~ y. T I 7 . u. I Is 5 a § . n I u . a A a . . K K N K N \- ‘ A '- fa fl. DEATHS PER 10,000 POP. e. /'-. / , -.\~v.. . waved ' aa_a~P // / I ‘l‘~ is We.‘ figure 8.6 (Source: Table 8.3) RURAL DEATH RATES : NEPHRITIS NICHICAN Av., 1935-1944 = 5,9 ” llf C sc-uo LC'l" U'J A.C. UACII’s‘C r0001! :2 aid 1. ' x r‘) I - r ‘/ / “Q‘s/ . .1, T, 1 a A \ I ‘x\. 5' . L , / . ‘~_4 ens . / '%( ,4“ ' / ‘.~- 9 L, / 'y . hum ours/A. NI Owl 5.! ",3 ' ’//? ., . ‘a .-‘ - / / A ' . . nor- . ' - m Dvstso «u -. _PMA 14 A a ) \ J ' , umsr ”.01.“ (aura-r 933’; A. i ’r‘ I 5v ‘ vS' I "\'s~.‘n I‘nseJ. ’ ,(u V. 0:.“ , /\.\ \. 9 DEATHS PER 10,000 POP. u , , ‘1 . " A I / I /// WDER 3’6 \. u u uttts'a Stein ‘5‘“ DA 5.9 7.1 8.3 OVER 3- BIL!" CASS Josreu - Aw \ , _ 6 .sAv-g ‘ - a ' “ 30 e i t ”M'- . -~ - l ‘ O r 9 .J'. A. , x I S-‘ar ‘1 '“t Yu‘V‘ INY ., (“u'o‘ «ASSN. I l ;' ~(C‘W 3"“ A C1 «on. ~s.>'5.a renew: M. INA. . ‘ / . Q I UAIC‘.“. "M. A. H '4‘! ”a. “'94! ‘ USN-'1 lignr. 8.7 (Source: Table 8;!) 211 Tab]. 8.5 RURAL MORTALITY RATES FROM THE PM PRINCIPAL CAUSES 0T DEATH PER 10,000 POPULATION, MICHIGAN COUNTIES, AVERAGE 1935-1944. (Based on 1940 population; by place or residence) ==================================aa=========================== State Em:a1 Death Rates and _Eeart Cancer A opiexy Accidents Nephri ti s County 1(90-95) (45-55) fSSab) (169-195) (ISO-132) Alcona 35.15 11.17 13.81 10.79 8.48 Alger 25.01 8.51 6.25 14.93 6.07 Allegan 37.35 13.19 13.04 10.15 5.34 .11p8n. 85.70 9.80 9.30 8.98 3.51 Antrim. 37.30 14.14 14.08 7.57 8.11 Arenao 33.03 10.51 9.80 11.15 5.84 Baragaz 83.19 9.09 9.19 13.40 3.08 Barry 37.45 13.48 15.80 9.98 9.34 Bay 84.87 10.58 7.10 9.44 8.39 Benzie 39.08 13.09 11.79 18.43 8.40 Berrien 34.39 12.41 10.19 9.36 8.17 Brandh 41.83 18.84 15.73 9.94 8.37 Calhoun 31.38 8.85 11.30 7.85 4.11 0822 37.45 15.38 16.59 18.30 13.25 Charlevoix 43.80 15.11 12.44 10.50 7.30 Cheboygan 35.50 10.41 10.04 9.88 5.13 0h1pp0'l 30.95 8.01 8.80 9.01 4.34 0183. 58.74 11.90 11.45 14.07 8.18 Clinton 29.44 11.15 10.88 8.35 12.00 Crawford 30.81 15.14 0.04 11.42 8.49 D0193 29.15 11.00 9.98 11.00 4.91 Dickinson 27.85 9.94 9.81 8.70 8.88 Eaton 34.97 14.11 11.15 9.48 0.01 Emmet 48.47 14.03 13.30 9.98 5.52 Gene see 22.03 7 .94 6 .87 7 .17 4. 24 Gladwin 89.58 11.01 10.97 9.80 7.35 0089010 28.45 10.97 10.04 15.43 5.39 Grand TSQVBIBO 39.73 11.43 14.44 10.39 0.37 GIGtiOt 35.34 13.28 10.84 9.15 0.95 H111.d&10 38.30 15.81 10.91 10.31 7.30 Houghton 32.33 14.54 18.08 8.47 0.50 Huronz 88.08 18.55 10.19 9.48 5.01 Ingh‘l 80.07 9.93 10.91 9.01 4.88 IOB1I 30.08 14.35 18.41 8.53 0.50 10800 33.04 14.37 11.80 9.11 0.05 Iron 80.78 10.39 5.14 9.78 5.91 1865811. 38.03 11.07 10.70 9.10 4.58 Jaekson 88.08 10.31 9.25 7.38 5.18 Inlamazoo 30.08 9.90 9.89 8.08 5.91 ‘Kalkaska 34.12 11.05 14.15 7.75 4.84 Kbnt 89.78 11.05 9.38 5.75 4.58 waeenaw 38.47 12.84 15.73 8.49 8.34 212 Table 8.3(Continned) RURAL MORTALITY'BATES rnom man 3173 PRINCIPAL CAUSES OF DEATH PER 10,000 POPULNTION, MIGHIGAN’COUNTIES, AVERAGE 1935-1944. (Based on 1940 pepulation; by place of residence) ¢ age - Deat- 'ates an Heart Gancer A oplery Accidents Nephritis °°“n*y 1(go-95) (ge-ss) £83ab) (169-195) (130-13g1 1:919. '37.93' 13.78 15.21 12.08 6.87 Lapeer 31.62 11.33 13.27 ' 9.42 6.16 Leelsns 37.70 12.33 10.91 7.94 0.99 Lenawee 36.39 13.48 14.43 10.57 6.15 Livingston 40.37 14.08 11.22 9.81 4.84 Lucez 38.53 6.87 6.20 9.1. 4.71 Mackingea 29.5. 10.28 10.50 12.07 4.97 iMaeomb 22.17 9.86 8.87 8.33 5.66 Manistee 42.54 14.76 13.53 9.43 5.02 Marquette 24.14 11.40 17.85 8.38 4.09 M3805 28.57 12.92 13.49 8.80 5.71 Mbcosts 35.00 12.17 15.36 9.31, 8.64 lenaminee 27.71 11.87 8.53 0.82 2.93 Midland 22.25 8.77 5.01 7.03 11.03 Missaukee 30.74 8.46 12.70 7.21 4.35 “OHIO. 29.04 10.80 9.34 10.20 4.28 Montcalm 38 .09 14.57 15.48 12.51 8.10 Montmorency 27 .08 11.98 5.99 11.45 8.85 lbskegog 25.49 10.05 8.32 7.95 5.07 NOVQJBO 37.04 12.29 13.07 7.93 0.84 Oakland 20.52 9.17 7.09 8.03 3.78 Ooetne 38.39 14.05 15.93 11.13 5.40 Ogemss 31.88 9.65 17.20 10.89 4.93 Ontonagon 27.29 11.09 13.29 10.82 5.54 080001. 38.92 14.88 15.85 10.89 7.43 Oscoda 24.38 11.01 7.08 7.86 8.25 098080 35.52 11.50 7.21 9.95 8.58 Ottawa 28.58 13.30 10.17 6.28 5.36 Presque Isl. 23.86 10.46 6.32 6.86 4.17 Roscommon 50.53 10.91 7.63 9.81 6.54 Saginaw 25.99 10.63 8.41 8.62 4.95 St. Clair 30.85 12.87 8.36 11.13 7.95 St. Joseph 41.74 16.27 16.49 10.43 10.55 8.21180 33.54 12.09 14.94 9.09 0.28 Schoolcraft. 25.94 9.21 10.67 10.42 5.82 ahiewsssee 32.47 15.05 15.40 9.29 9.38 Tuscola 35.50 12.78 11.68 9.10 6.93 Van Buren 45.41 17.72 16.30 11.65 8.69 ‘Weehtgna' 27.11 12.32 11.08 9.28 4.58 Wayne 16.73 6.13 3.76 7.46 2.20 Wexrord 39.40 12.07 12.50 7.34 9.11 ‘figggg 29.88 11 42 10.46 9.94? 5.90 1. International List Number, 1939. 2. Residence according to the 1930 population. Source: Annual Reports, Nflchigan Department or Health, 1935- 213 The highest incidence of rural deaths occurs in the western half of the Lower Peninsula and in the southern tier of counties. This generalisation describes the in- cidence of mortality from heart diseases and apoplexy more accurately than from cancer and nephritis. It is not true for accidents. Lew rural death rates occur most frequently in the highly urbanized areas of the State. This is true, for the most part, for all of the five causes of death. The distribution of rural deaths from cancer is more dispersed than for diseases of the heart, apOplexy, and nephritis. There is a heavy concentration of high rates from cancer, however, in the southern tier of counties, some noticeably high rates in and around Ionia county, and others along the western coast running from Oceana tc Benzie county inclusitely. Low death rates are found in the belt of counties running from Homes on the southern border to Midland and Bay county and thence to the Straits. The Up- per Peninsula is generally characterised by low rates, 10 cf the 15 counties having rates under 11.3. These varia- tions among the counties are probably due to the differing age distributions, the mean age of deaths from cancer being somewhat lower than for other dircnic diseases. Dublin points out, for example, that the average age at death for cancer in the United States in 19:50 was 63.35 years, com- pared with 66.21 for heart diseases.1 1. Bee Dublin, Louis, figs 0 . 93.3., p. 115. 214 The Upper Peninsula counties tend to have 10' rural death rates for the chronic diseases. On the other hand, accidental deaths in rural areas occur most often in this area. Although the highest death rate from accident in the state is in Bay county (18.87), the second and third highest rates are in Upper Peninsula counties, namely, Gogebic and Alger. Menominee has a rate of only 6.82, the third lowest among all counties. It is probable that some of the high accident death rates in the Upper Peninsula are associated with the mining industry.1 In the Lower Peninsula high death rates for rural people ‘ occur more frequently in the northern than in the southern counties. Rates for nephritis follow the general pattern for chronic diseases except that high rates tend to be concen- trated in the inland counties. Although for other degen- erative diseases, counties along the western coast from Oceana to Bennie generally have high rates, for nephritis their rates are average, ranging from 4.8 to 6.9. The highest rate, 13.25, occurs in Cass county. As might be expected, the Upper Peninsula counties generally have low rates. Rates are also usually low in the predominantly urban areas. The relative ranks of the five principal causes of l. Darn, Harold 1., o . c t., p. 6, found the accident death rate to be gher in mining than in non-mining areas of Ohio. 215 death in rural segments of the counties are given in Appen- dix II, Table 2. The death rate for heart diseases occupies first position in every county. Cancer is the second most fatal disease. However, in four counties of the Upper Peninsula and in one county of northern Michigan it fell to fourth place. Apoplexy, the third principal cause of death in the State, dropped to fifth position in four coun- ties, all of them in the northern part of the Lower Peninsula. Perhaps the greatest variation among death rates occurred for accidents. Although it is the fourth cause of death in the State, it Jumped to second place in six counties of the Up- per Peninsula, in three counties of the northern part of the Lower Peninsula, and in Wayne. Hephritis, the fifth most fatal disease,ranks second in Clinton and Midland counties, and third in Oscoda. To further aid in the interpretation of the data, the twelve highest and the twelve lowest death rates in rural areas for each of the five principal causes are listed in Table 8.4. Vans and Genesee are among the lowest death rate counties for all five causes; Oakland, Midland, and Presque Isle for four of the five causes: and Baraga and Alger for three of the five. Among high death rate coun- ties, Van Buren is listed for all five causes; St. Joseph and case for four of the five; and Montcalm and Osceola for three. Thus, low death rates tend to be found in the rural segments of counties with large urban populations thereas high death rates are found in the corresponding TIMI. 8e‘ PRINCIPAL omens IN RURAL mus, MICHIGAN, mmor 19:55-19“. 216 TWEBVI.HIGHEST AND LOWEST DEATH RATES FOR FIVE Heart Disgases Highest Death Rates Lowest Death Rates Oakland 30e58 Iron 80 978 Genesee 22.03 Macomb 22. 17 Midland 22 . 25 Baraga 23 e19 Presque Isle 23.86 Marquette 24.14 Oscoda 24.3% B” 24 e near our Wayne 6 e13 L110. 6 e87 Genesee 7.9: NHssaukee 8.46 3:8“. e 5 ippewa . Calhoun ‘8783 Buflg‘ 9 e09 Oakland 5.:3 Schoolcraft 9 Alpem m layne 3.3: Montmorency 5.99 13011 e “00 e Alger a e 55 Presque Isle _.__3_2_ Crawford - 6 e6 Genesee .0; 080 06.3 7 e Oakland 7.59 Bay 1.2.1.2 Cancer 1291122: Van Bureau 45.41 011.131.6701: ‘2 e86 llanistee 42 . 54 Emmet 42.47 Branch 41.83 Ste Jouph 4107‘ Livingston 40.37 Grand Traverse39.73 Benzie 39.62 'OIde 39 .40 Osceola 38.92 Van Buren 17.72 St. Joseph 16.27 Hillsdale 15.81 Cass 15.38 Crawford 5 e ‘ Charlevoix 15711 Shiawassee '15755 Osceola 14.88 Mani stee I4 . 73 Oceana 'ITTEK Hought on 14e5‘ Marquette 17.85 Ogemaw 'I77EU Hillsdale I621! Cass 16.69 St. Joseph 15.49 Van Buren 16 e 30 Oceana 15.93 Osceola 15.85 Branch 15.75 Keweenaw 15.73 Barry 5 e Mont calm 15.48 T‘bl. Be‘ (Continued) 217 mm: HIGHET AND LOWEST DEATH RATES FOR rm PRINCIPAL omens IN RURAL AREAS, MICHIGAN, AVERAGE 1935-1944. Accident! Losest Death Rates Highest Death Rates Ottawa 6.28 Gogebic 15.43 Kent 6.75 Alger It?! Hsnominse 6.82 Clare . ' Presque Isle '3755 Baraga §§.gz Genesee . 77177 Mont calm T2731 Missaukee 7 21 Benzie 12.43 Wexford ‘VfEI Cass . Jackson 7738' Lake 12.08 Wayne 7 .45 Mackinac 13.757 Antrim. 7.57 Delta . Ilidland, . ‘Van.Buren 'IITKK talkaska . Montmorency 11.45 Nephritig wayne 2.20 Cass 13.25 Dickinson 3.82 Clinton 12.00 mummC e9 Ste JOIOPh 10055 Cheboygan 3&3 Mid land m {Alpena . Shiawassee 9.38 Baraga '37?! Barry 9. Oakland '5778 ‘lbxford 5.?E Marquette 4.09 Montmorency 8785 Calhoun 4.11 Alcona m Presque Isle 4.17 Van.Buren '8739 Genesee 1.24 Mbcosta 8.64 Monroe .28 Otsego “8765 Note: Rates based on less than 200 deaths for the ten year 80‘11'0. 3 period 1935-1944 are underscored. Table 8 e 1e 218 segments of rural counties. In the foregoing pages, certain counties have ap— peared consistently within the column of the twelve high- est er the twelve lowest death rates, while other counties appear occasionally, and some not at all. The extent to which counties with extremely high or low death rates are concentrated in designated areas seem.to follow certain patterns of mortality incidence. Some of the generalisa- tions in this Chapter regarding these patterns are further demonstrated in the following analysis. Table 8.5 indicates the number of times counties in a specified area, for ex- ample, the Upper Peninsula, are found anong either the twelve highest or the twelve lowest rural death rate counr ties. Since there are five principal causes of death, a single county has five opportunities to place among those with the twelve highest rates, five opportunities to place among those with the twelve lowest rates, and ten oppor- tunities to place in either extreme. According to Table 8.6 rural segments of counties in the Upper Peninsula placed among the twelve lowest death rate counties 22.7 percent of their opportunities, or about twice as often as among the twelve highest (10.7 percent). Counties in both the northern and southern part of the Lower Peninsula are found about as often in.the twelve highest death rate counties as among the twelve lowest. Counties in the Upper Peninsula occupy positions among the twelve lowest death rates relatively more often 219 Table 8.5 NUMBER AND PERCENT OF TIMES COUNTIES ARE FOUND AMONG THE TWELVE HIGHEST AND ‘IHE TWELVE LOWEST RURAL DEATH RATES FROM THE FIVE PRINCIPAL CAUSES, BY GEWRAPHICAL AREA, MICHIGAN, 1955- 1944 Area Number Place- Time s Per- of ment Placed cent Countie s Ch ances Upper Peninsula 15 75 Lowest 17 22. 7 Highest 8 10 . 7 Northern Michigan8 33 165 Lowest 22 13. 3 Highe st 26 15 . 8 Southern Mich igan” 35 175 ‘ LG" I‘ 21 12 e 1 Highe st 26 14. 9 Urban Counti ea3 11 55 Lowest 18 32 .1 Highs st 0 ---- Rural Count lacs 24 120 LOWOBD 5 2 e 5 Highe st 26 21 . 7 1. Southern Michigan includes all counties below an imag- inary line running from the most southern point of the coastline at Bay county to the northern boundry line of Muskegon county. Bay county is included anong the northern counties and Muskegon among the southern counties. All of the numb area is included among the southern counties. 2. A county is classified as urban if more man'so percent of its population in 1930 was urban. The remaining counties are rural. Source: For crude rates see Table 8.4. 220 than counties in either area of the Lower Peninsula, the percents being 22.7 in the Upper Peninsula, 13.1 in northern Michigan and 12.1 in the southern part of the State. If the counties of southern Michigan are divided into counties predominantly rural and counties predominant- ly urban, the rural segments of urban counties are found among the twelve lowest death rate counties while the rural segments of rural counties are found among the twelve high- est death rate counties. It is interesting to observe that crude death rates from chronic diseases are someahat lower in the areas where rural levels of livingl are low, and that the prosperous farming regions in the southern part of Michigan have the hiaest death rates for these causes.2 For example, the 1. The measures of level of living employed are these con- structed by Margaret Hagood, _o_p. cit., pp. 22-23. It includes percentages for each of 25'; following: farms with gross income of more than $600, with auto of 1936 or later model; dwelling units with fewer than 1.51 persons per room, with radio, with running water, with mechanical refrigeration; median grade of school com- pleted by persons 25 years of age and over. l'or an , account of the development of the indexes, see "De- velopment of a 1940 Rural-Farm Level of Living Index for Counties,‘ Hagood, Margaret Jarman, Rural Sociol- , Vol. 8, Ho. 2, June 1943, pp. 171-185: flBur vol of Living Indexes, 1940', in Notes, Burg Sociol , September 1943. . ' 2. Vance writes, "In the main it can be said that the de- generative diseases of age have higher rates in the regions with higher levels of livin and show an in- crease from 1930 to 1940 (Table 100 . This is true in both the Nation and the region for heart disease, malignant tumors, cerebral hemorrhage, etc.“ Vance, Rupert D., 32. 3gp, p. 341. , (”n - 221 twelve high mortality counties from cancer have a mean death rate almost twice the size (82%) of the correspond- ing rate in the low mortality counties. The mean rural level of living index is about 10 percent greater in the areas of high mortality. It is unlikely that mortality differentials of this size would be reversed if adjusted for differing age distributions. Dcrn found that in Ohio ”the standardized mortality rates from cancer, cerebral hemorrhage, heart disease, and nephritis were slightly higher in to good economic arose.“ In attemptirg to explain, he continues in his sumary as follows: "fie fact that the death rates from the important diseases of late adult life are somewhat lower in the poor economic regions would appear at first sight to support the theory that modern medical and public health practices tend to lessen the effects of natural selection and to preserve a larger proportion of the weak and unfit than would otherwise be true. Accord- ing to this theory, high death rates during infancy and childhood eliminate the least physically fit mem- bers of society so that attempts to decrease mortal- ity at those ages, if successful, would weaken the race. It on not seem necessary to examine the validity this theory at this time, especially in- asmuch as there is practically no direct evidence pro or con. It is unquestionably true that modern health activities do preserve for many years the lives of mapy persons who under conditions existing a century ago would have succumbed at an early age to some dis- ease which is now prevented or cured. Whether or not this affects the physical vigor of the race is a de- batable question.‘ Mott finds itudifficult to accept the findings of x ' Dorn in Ohio, namely that areas of poor economic status tend l. Dcrn, Harold F., _o_p. git" p. 9. 8. Don, 332016. ’0. ibidP’ p. 8. 222 to have lower death rates for degenerative diseases than the good economic areas. He suggests that these mortality differentials are probably due to "poor reporting of deaths as well as faulty diagnosis" in the poor economic areas. However, assuming that reported“, differences exist, he explains the high rates in the good economic areas thus: “Perhaps more obesity in the prosperous sections contri- hates to a higher occurrence. of the degenerative diseases of later life."1 He rejects, along with Dorn, the thesis that the differentials are reflections of the preservation of weaklings and “the .unfit' in early life among the better economic classes.‘ It appears to the writer that the high death rates from degenerative diseases in the better agri- cultural areas of Michigan can best be explained in terms of selection and migration, although without doubt the registration of deaths is more efficient and faulty diag- nosis less prevalent in the southern counties than in other areas of the State. Many rural areas of the Upper Penin- sula, as well as of northern Michigan, offer little entice-’ ment or encouragement to the elderly farmer either to re- main or settle there for his last days. On the other hand, many older farmers in the good agricultural areas of southern Michigan are enjoying all the conveniences of modern life. Accidents (all forms) is the only non-chronic cause of death presented by county and res1dence. As might be 1. Mott, Frederick D., 239,, 93. git., p. 72. I. II 223 expected, it does not follow the patterns of chronic disc cases when related to the rural level of living index Just presented. Death rates from.accident tend to be high in areas there levels of living are low. In the twelve counties with low accident rates, the mean level of living index is 9.5 percent greater than for the high mortality areas. This finding is also in.keeping with the Ohio study.l Dorn explained the condition as resulting in part from mining accidents. This explanation.might likewise apply to the Muchigan data,--especia11y since five of the twelve counties are in the Upper Peninsula. Thompson, in an analysis of occupational mortality differentials, points out that the differences are not due to occupational hazards alone, and that selectivity plays an.important role. “Common unskilled laborers, dockers, and stevedores are frequently men who cannot succeed at other Jobs and in the course of time drift into the poorly paid classes ihere living conditions are very bad... It is perhaps the living conditions enforced by poorly_paid. work rather than the nature of the work itself that in many cases, though not in all, cause the high death rate."2 It is quite possible that in.Michigan, also, processes of selectivity account for some of the excessive death rates in areas where levels of living are particularly low. 1. Dorm, Harold 1., 22. ci§., p. 6. 2. Thompson, Warren 8., pp. cit., pp. 238-239. DEATHS PER 10,000 POP. UNDER 25.0 25.0 - 28.9 29.0 - 32.9 33.0 - 36.9 37.0 - 40.9 41.0 - OVER Ho URBAN POP, URBAN DEATH RATES : HEART IIUHIGAN AV., 1935-1944 : 27.8 .... .0 .‘ ' .... 4 tan: .Oszeu guano. ”w. 8.8 (“W't MI. 8.5) URBAN DEATH RATES : CANCER MICHIGAN Av., 1935-1944 : 11.6 DEATHS PER 10,000 P0P. 10.2 11.4 12.6 13.8 15.0 DIIEEM UNDER 10.2 \ . d .' "Ten: as“ e. 4"“ A ,_ \\‘0“ A n“- "3 U” do enx5an “QI— \. K hares.L \ a n( 0.5-N /2779 . ' ~ \ I } 1’ .\ I ... 11.3 \ 0((... (\.‘S'l ‘3‘.(AL‘/~“‘~n \\ u ~M u u unam‘ 'V‘ '1. 12.5 “150 ‘ muse I I / .~ . I // / . K. ... \H a 13.7 v . ,\, ~ . "asst: ) “a“ / / / f /’ Uh9 ///,,A Lt N A ..v-e a . v‘ean U-K‘ . .1. OVER ,’ I. M - aa a. must ‘ L“ us 0 ~ u 0 U NI\‘ “ NO URBAN POP. ,‘ .' .. .RE A _tosq p. - ‘h. ‘ --— Pigure 8.9 (Source: rstle 8.6) URBAN DEATH RATES : APOPLExx MICHIGAN Av., 1935-1944 : 7.6 DEATHS PER lo,ooo POP. Figure 8.10 (Sourceri'able 8.6) URBAN DEATH RATES : ACCIDENTS HICHIGAN Av., 1935-1944 : 6,3 an I IAJI'JII DEATHS PER 10,000 POP. ,. (522;. UNDER 7,5 ),m H” ANM,LJH.JW- // ’II \6" v 7 0 S " 8 e 9 \ /, ’ o: w ‘ qun.‘ {C 0'. A .1. LL‘ “'1‘ AND \ . . -.t I Al is >> \ rel .uvrr YuSCO' ‘ , ' l 9 . O - 10 . a F mangns gauiL AC. ‘ / / «lav ( Y 'J'uaev / I I /’ l LN 10.5 - 12.9 fivfiwf n/i/ . A I turn «541;! ...'s”./‘ (t s ‘ J. A l/ I g/ I ”'0 ’ 14°“ ‘ // ’x g t ‘ a "gun“ ~..r,5v ’ A A” I. I / I 14.5 - OVER V . H. ,mm. ix; u“ “A," I HIEEEM NO URBAN POP, / ...... c... 0.}... .. - - -- u .___._5. I c Figure 8.11 (Source: Table 8.6) 0'" noon. DEATHS PER 10,000 POP. \ \\‘\ \ HIIEENl 3.6 4.8 6.0 7.2 8.4 UNDER 3.6 NO URBAN POP , URBAN DEATH RATES : NEPHRITIS MICHIGAN Av., 1935-19441:-5.2 ‘ngel Canaan LUCC ‘1 UCT uacsuuac t i ”W CIA A102! nJ / / /’ , // » / , s \ g I i // a 9 K I Y I llii‘ i / ‘\¥/ / if . / c es. ISL r-I-\I , ' 1 ’ I ‘ | r$>‘/ . I -.\ '-’ . ' 4' - ' 4 “new orsioo sou-(v L '46 dunno I K G ’- Dt~lvt 'uvus ULOJII cuwoso oscooa MCON‘) .03- cl" ’ m: we newt! cot-nos can.“ OSCO / AILNAC . ' ,.,/ $477 ./ ' r I/. ) uc~ Lnii CSCtOLI (.Afli CK‘W'N .‘y \\ /" Inf “W" - 4.7 I: , (its! «(to A Aetit 0‘ HC~T At“ stemco U ' .- 5.9 3 auto emu-AC IAYIOT "‘ A 5v MA at '7’.:1 ms t St L 7“ stuv kl CLIN'ON 3 - 8 ,3 , c .sllv (avou n Lmeo ‘*~*"° as oVER . I Ate U AC rain in- ! ¢ It” A C 53- JOit'n OLA!“ “.... — O nw~,—121- (Sour-cor ‘rableca‘fi ) 224 mm 2; m _i_n. Mn £3233: Non-communicable Urban.death rates for the five principal causes are given in Table 8.6 and Figures 8.8 to 8.12. These data may be summarized briefly as follows: (1) Deaths from chronic diseases in.urban areas occur relatively more frequently in the southern and western counties where cities are small and rural pOpulations are predominant: (2) The low urban death rates from.chronic diseases tend to occur in areas of large urban pOpulations: (3)‘Urban places of the‘Upper Peninsula.tend to hate mortality rates uhich are lower for chronic diseases than.thcse of the Lower Penin- sula. In this area cancer has high rates compared with other degenerative diseases. Prom.a.comparison of urban and rural data, certain mortality characteristics of the urban segments of the population appear to be quite similar to those of the rural segments. For example, it may be concluded that urban rates in Michigan counties tend to be high in the areas where rural rates are high: and where urban rates tend to be low rural rates are likewise low. The relative ranks of the five principal causes of death in the urban segment of each county are given in Table 3 of Appendix II. ‘As was true for the rural population of counties, heart dis- eases occupy first place as a Leading cause of death in all instances. Cancer holds second position in urban seg- ments and in no instance falls lower than third place. Por the rural segments, cancer is fourth in 4.8percent of 225 Table 8.6 URBAN MORTALITY RATES FROM.THE FIVE PRINCIPAL ‘ CAUSES OF DEATH PER 10,000 POPULATION, MICHIGAN COUNTIES, AVERAGE 1935-1944. (Based on 1940 population; by place of residence) W State ‘Urban Death Rates Keweenaw pa..- and Re art Cancer A We County 1(90-95) (45-55) fB3ab) (169-195) (l30-l32) .Alcona ----- ----- ----- ----- ----- 1118.! 88054 11057 8088 11079 5081 Allegan 03.50 15.72 14.33 10.53 6.66 Alpena 38010 10008 8010 7068 8040 mnac ----o ----s- ----.- ---00- ..--.-- Barasa‘ 00---- 0--.. ----- ----.0- . -..--- Barry 39005 15087 81028 12017 8050 Bay 33051 1‘051 10054 8017 705‘ Dennis ----- ----- ----- ----- ----- Berrien 34.18 13.68 10.92 10.36 7.02 Brandh ‘51007 18075 19047 13078 9041 Calhoun 90000 15021 14038 9071 7027 0888 48014 82077 15038 16.78 13078 Charlevoix 33003 17058 18053 9029 8036 Cheboygan 38008 19098 15051 11088 8087 Chippewa 33.32 10.85 8.98 9.71 _ 5.42 Clare ----- ----- ----- ~---- ~---- 011nt°n 38041 14070 18008 9004 17041 cra'rord ----I- Oeue-.. ----- Isc-as-n -0000-.. 8.183 89.39 13.89 12.62 9.29 3.35 D18k1n80n 88087 11087 7034 8011 5000 Baton 49.19 17.60 16.08 11.99 10.79 IHNBU 48019 14089 15088 9088 5018 OOHQIOO 31088 9.80 6039 6089 4080 Gladfiln ----~ ----- ----- ----- ----- Gogebic 24.38 11.60 10.53 8.17 5.98 Grand Tiaterse 31.55 11.69 13.84 7.47 4.21 0’83103 37085 1‘008 10089 807‘ 9018 31118081. 39049 14028 13003 11088 801‘ Houghion 40.43 19.58 14.72 8072 5.91 Huron ----- ----u use-.- ----- . -00--- Ingham. 85097 18009 9019 7032 8043 Ionia 37.97 13.66 14.69 9.06 5068 10.00 -..-_- ----- 00-..- one... an... Iron 88091 14031 8036 9000 8051 18800118 81050 10088 8008 9018 4087 Jackson 38.71 13.80 12.08 9.78 6.96 ‘Kalamazoo 33053 10005 13011 9039 8038 Knlkaska. "--' "vrr “"“ “"' ""‘ Khnt 33.96 14.53 10.38 7.50 3.40 226 Table 0.. (Continued) URBAN MORTALITY RATES FROM THE rm * . PRINCIPAL 040323 or DEATH PER 10,000 £0PUL4TI0N, MICHIGAN COUNTIES, AVERAGE 1035-1944. (2.2.0 on 1940 pepulation; by place or residence) State , Urban Death Rates and He afi Cancer A oplexy Acci dent We County 1( 90-95) (45-33) 83ab) (109-193) (130—132) m. ' ----- -----‘ ----- ‘ ----- ----- LEPOOI 27098 14.73 13030 8078 8013 n.13na A ----- ----- ----- --..-- ----- I‘DGIO. 35.77 14.83 13.42 11003 9034 Livingston 46.42 16.01 10.67 17.87 5.60 no.2 ----- ----- ----- ----.- ----- Mackingos ---‘- ----C ...-O. ----- ----- Hloomb 22.17 10.08 8.25 7.65 4.94 th10900 46.93 18.52 18.63 7.70 4.37 Earqnotte 32.20 13.37 9.96 . 8.31 5.86 Mason 4808‘ 13058 9077 10057 7013 “000833 50093 13023 13003 10042 18083 Menominee 35.19 16 .13 13.39 9 .09 . 7 .78 Midland 18.75 8091 5032 8088 1008‘ M12 saukee ----- ----- ----- ----- ----- Monroe 89008 12030 7004 10071_ 4011 JMontcaln. 45.29 11.84 14.28 8.64 7.89 Montmorency ----- ----- ----- ----- ..---- lunakegog 28002 10088 8053 8047 3027 N.'ay‘g° ---- ----- ----- ----- ---.. Odkland 20.22 9.20 6.31 6.68 4.21 0°.“a ----- ----- ----- ----- ----- Ogmnaw -"-' -~--- '---- ----- ---~- “tongs” I'll-CI- ----- ----- ...-.0. ...-0.- 08°. 01‘ ----- ---- ----- ----- ----- Oscoda ----- ----- ---------- ----- OtSOBO "-" """ "“' -"“' “---- Ottawa 30080 13058 9058 6035 0008 Presque 1810 20.51 10.09 3.91 7.81 4.23 Roscommon ----- ----- ----- ----- ----- Saginaw 31.60‘ 11.79 9.88 7.83 6.47 Ste Clair 35097 14083 11097 9093 8091 89. JOBOph 32075 13050 13085 8081 9070 Sanilac -"“. ”"" “"‘ "“' ""‘ Schoolcraft 37004 18.11 8089 9081 8011 Shiawassee 31 .45 14.76 11.79 9 .97 7 .01 Tuscola 58.31 23.13 18.89 14.33 7.49 Van Buren 52.27 18.34 17.07 12.22 10.96 Weahtgnaw 35.32 14.00 9.97 7.72 4.79 Wayne 25020 10077 3.71 3008 4.49 chrord 39.17 11.36 10.76 9.43 6.19 8:59. 27089 11.82 700; 5081 5.2;_ 1. International List Number, 1959. 2. Residence according to the 1930 population. Source: 1944 , Annufi Reggtg llichigan Department or Health, 1935- ab ’ 227 the counties. Rates for apoplexy, the third most fatal disease, dropped to fifth place in the urban areas of two counties, Midland and Presque Isle. Four counties occupy this position for rural areas. Accidents, the fourth principal cause of death, rank second in the urban seg- ments of two counties, Isabella and Livingston, whereas for the rural segments it is second in ten counties. 81:: of these counties are in the Upper Peninsula. In both the rural and urban areas of Midland and Clinton counties, nephritis,the fifth cause of death for the State, is second. This is difficult to explain and is a problem requiring further study. The twelve hiaest and the twelve lowest death rates for each of the five principal causes of death in urban areas are listed in Table 8.7. Oakland, Genesee, Hacomb, and Wayne counties are among those with low death rates for all. five of" the principal causes; Midland, Dick- inson, and Presque Isle for four of the five causes; and lluslcegon, Alger, and Isabella for three of the five. Many of these counties are over 50 percent urban, and the remain- ing ones have large urban populations. Van Buren, Branch, Eaton, and Cass counties are among the high urban death rate counties for all cf the five principal causes of death; Tuscola and Barry for four of the five; and Manistee, Liv- ingston, and Cheboygan for three of the five principal causes. 'l'hcse counties are over 50 percent rural with small cities whose populations without doubt include many Table 8e? PRINCIPAL CAUSES IN URBAN AREAS, MICHIGAN, AWE 19935-1944 228 TIELNI HIGHEST AND LOWEST DEATH RATES FOR FIVE Midland Oakland Presque Isle Genesee Macomb Gogebic ' Wayne Ingham Alger Dickinson Lapeer Muskegon Midland Oakland Genesee Macomh Presque Isle Isabella Alpena Wayne Chippewa Mhskegon Wexford Dickinson Alger Presque Isle Nfldlsnd Wayne Oakland Genesee Alger Monroe Dickinson Isabella Alpena. Macomb Iron Heart Diseases Lowest Death Bates .... ...- 0| Q 3:3”; 3 f3 000 c! 0 Q 33 I? on a» Cancer hassles: Highest Death Rates Tuscola Van Buren Branch Mbcosta Eaton Man istee Livingston Montcahm Emmet Cass 3A1Legan mason Tuscola Cass Cheboygan Houghton Mani stee van Buren Eaton Charlevoix Branch Menominee Schoolcraft Livingston .Allegan Barry Branch Tuscola A Manistee Van Buren Charlevoix Baton . Cheboygan Cass Emmet Houghton Ionia 31:33- ..s \‘l e 0| 0 33333 ,‘3 8’» 3333333 333- Table 8e7 (Continued) 229 mm: HIGHEST AND LOWEST DEATH RATES FOR rm PRINCIPAL CAUSES IN URBAN AREAS, MICHIGAN, AVERAGE 1935-1944 Lowest Death Bates Wayne 6.68 Ottawa 6.35 ‘Muskegen. 37:? Oakland 6.68 Genesee . Ingham. 7e33 Grand Traverse7.47 Kbnt 5.50 Macomb. 7e66 Alpena 7.65 Manistee 3 Dickinson 3.40 Mbnroe ITII Grand Travarse§.8§ Oakland 4.2 Presque Isle 4.23 Menistee. .3' Wayne 4.4 Washtensw 4.79 gengsie 4.80 as e la 4 87 IMacomh 'ZI§Z Accidents Nephritis Highest Death Rates Livingston Cass Tuscola Branch Van Burean Barry Eaton 3Alger Cheboygan Hillsdale Lenawee Menroe Clinton Mbcosta Cass ‘Van Burean Eaton Midland St. Joseph Branch Lenawee Gratiot St. Clair Barry 17.87 .3 N01 e 333333333 33333333333: sNote: period 1935-1944 are underscored. Source: Table 8.2. Rates based on less than.200 deaths for the ten year 230 retired farmers. Thus the urban segments of predominantly urban counties tend to have low death rates, whereas the corresponding segments of predominantly rural counties tend to have high death rates. Althougi the high and the low death rates are in part reflections of the age com- positions of the respective counties, the implications of the magnitude of death rates are factors in the distribu~ tion of health services which cannot be overlooked. Table 8.8 shows the number of times counties in specified areas, for example, the Upper Peninsula, are found among ei ther the twelve highest or the twelve lowest urban death rate counties for the five principal causes. According to these data, the urban segments of counties in the Upper Peninsula and in northern Michigan tend to be found in the low mortality group more often than among the high group whereas in southern Michigan the reverse is true. Counties of southern Michigan utilised 86.6 percent of their opportunities to place in the high mortality group as contrasted with 18.8 percent in the low group. These counties appear among those with high rates more than twice as often as do counties of theUpper Peninsula. Upper Peninsula counties are found twice as often among those with low rates as among those with high rates, the percent of the total placement opportunities being. 80.0 and 10.0. If the counties of southern Michigan are .class- ified as predominantly urban and predominantly rural, the urban segments of urban counties are found more often 231 Table 8.8 NUMBER AND PERCENT OF TIMES COUNTIES ARE POUND AMONG THE TWELNE HIGHEST AND TWELVE LOWEST URBAN DEATH RATES FROM THE FIVE PRINCIPAL CAUSES, BY GEOGRAPHICAL AREA, MICHIGAN, 1955- 1944 Area Number Place- Times Per- of ment Placed cent Counties Chances Upper Peninsula 10 50 tLowest 10 20.0 Highest 5 10.0 Northern Michigan8 13 65 Lowest 80 50.8 'Highest 14 21.5 southern Michiganz as 165 LOWOIt 31 18.8 Highest ‘2 25e5 Urban Counties3 11 55 Lowest 22 40.0 Highest 1 1.8 Rural Counties?’ 22 no LOWOIt 9 8.8 Highest ‘1 37e3 1. In 1950 there were 56 counties ‘with urban population. 2. See footnote to Table 8.5. 3. A county is classified urban if more than 50 percent of its population in 1980 was urban. are rural. Source: Table 8.7- The remaining counties 232 among the low mortality group whereas the corresponding segments of rural counties are found more often among the high mortality group. These data support indirectly the previous gen- eralisaticn that theUpper Peninsula tends to have low rates for chronic diseases. (Four of the five causes under analysis are chronic diseases.) These data also corroborate. the findings that, in general, where, urban rates are high, rural rates are high and vice versa. m Maggitz m p; Besidegg Infant death rates for the rural part of each county are given in Table 8.0 and graphically in l'igIre 8.15. Examination of the map shows clearly that rates of rural people in counties of the northern part of the Lower Peninsula, and in the Upper Peninsula, are higher than the corresponding rates of the southern counties of lower Michigan. The counties with the twelve highest and the. with the twalve lowest rates, given below, emphasize this concentration of high rates in the Upper Peninsula and in northern Michigan. Twelve Lowest Death Rates ITwelve Highest Death. Rates Rural Infant Deaths Rural Infant Deaths Ottawa I 30 .5 Crawford 68.8 Iosco 1.6 Luce . 37 .1 Barry . Gogebic 3375 Mason 3 e Alger W St . Joseph - . 2 . lexford m Menacinee §5 . 5 Montmorency 60 RURAL INFANT MORTALITY RATES (Allocated to residence) MICHIGAN, AVE, , 15235-1944 : 44.5 .V 8‘ - e’e’e‘e ~ ”e e' e‘ .‘e 'e e 'e’e’i‘f Mam. . . .H.'.'.:) @10 /" :.:W .0 dafifié.{i-.—&n$-.u¢t) :mz0z0; .30 H... 0 / mu! PBS 1000 :é.‘ fidfifl LIVE mm / :‘M_""—"’" M \aie’. -° We .‘c‘ .4 0‘~ 0. _0 ' .MMW- 3". r” 0 ..0 0 0. . f 0 0 0 0 0 ’/’/’, Under 58.0 0 . I 58e0'4109 m 42e0'45e9 m 46eO-49e9 n.00.0. ...0 “00.0.. m 50.0-54.9 o’o’o‘: _0 ’ - 55.0-Over Figure 8.13 (Source: Table 8.9) 233 Clinton 35 e ’7 M1 ssaukee 59 e 9 Baton 33 .5 Ont onagon m Hillsdale 5313' Leelanau 3675 Kent m Keweenaw m Kalkaska 37.5 Baraga m Ionia 3779‘ Benzie ' m or the twelve high rates, six are in counties of the Upper Peninsula, and six are in counties a northern Michigan. In contrast, eight or the twelve counties with low mortal- ity are in southern Michigan. The exceptions are Iosco, Mason, and Kalkaska in northern Michigan; and Menominee in the Upper Peninsula. The lowest rural infant rate is 30.6 in Ottawa, and the highest is 68.8 in Crawford county. There is a clear-cut inverse relationship indicat- ed between infant death rates and rural levels or living1 in the data Just presented. In the twelve counties where rural infant mortality is high, the average level or living index is lower than in the twelve counties where infant nortality is low. Death rates in the high mortality counties are 76 percent in excess of rates in the low mortality counties, and the average level or living index is 88.4 percent greater in the counties with low infant death _ rates. This is not‘ surprising since it is in keeping with the expectations as well as the results or several studies hearing on the relationship or mortality and economic I. Source or rural level or living measures: Hagood, ‘ Margaret Jaman, 'Rural Level of Living Indexes for counties or the United States, 1.940," Bureau of Agri- cultural Bcononics, United States Department or Agri- culture, Washington, D. C., 1943, pp. 532-23. 234 Table 8 .9 RURAL INFANT MORTALITY RATES PER 1 , 000 LIVE BIRTHS IN COUNTIES or MICHIGAN, 10 YEAR AYER- AGBS BIRTHS 1ND DEATHS, 1935—1944 (Exclusive of stillbirths; by place of residence) State : Total : : State : Total 1 and :Number’: Rate a and :Number 3 Rate County :Deaths : : County : Deaths : silcona 58 54.6 Lake 40 49.6 Alger 67 60.9 Lapeer 816 39.3 Allegan 855 41.8 Leelanau 90 56.9 Alpena 71 48.4 Lenawee 337 46.8 Antrim. 118 48.7 Livingston 110 39.9 Lrenacl 81 48 . 5 Luce 108 67 . l Baraga as 55.0 Mackinth 108 52.5 Barry 93 31 . 7 Macomb 588 45 .6 Bay 858 44.3 manistee 76 43.6 Bengie 93 55.9 marquette 130_ 45.7 Berrien 344 44.6 Mason 54 38.5 Branch 130 40.3 Mbcosta 118 46.5 Calhoun 865 43.4 Menominee 90 33.3 Cass 111 50.8 Midland 157 37.8 Charlevoix 108 53.8 IMissaukee 106 59.9 Cheboygan 75 47.8 Mbnroe 893 48.8 Chippewa 145 54 . 5 Mont calm 830 58.6 Clare 100 49.3 1Montmcrency 48 60.8 Clinton 156 35.7 mushegon 337 49.8 Crawford 58 68.8 Hewaygo 176 47.1 Delta 156 50.7 Oakland 989 48.1 Dickinson 65 47.4 Oceana 188 47.5 Eaton 148 36.0 Ogemaw 97 51.7 lmmet 90 48.8 Ontonagon 180 59.6 Genesee 661 40.9 Osceola 138 49.6 Gladwin 99 46.9 Oscoda 84 46.8 Gogebic 189 63.0 Otsego 64‘ 49.1 Grand Traverse 87 45.1 Ottawa 810 30.5 Gratiot 853 47.8 Presque Isle 76 43.1 Hillsdale 147 56.9 Roscommon 33 46.0 Hough on 875 51.8 Saginaw 418 43.5 Huron 347 51 . 8 St . Clair 316 48 .9 Ingham. 485 43.8 St. Joseph 181 38.5 Ionia 165 37.7 Sanilac 865 45.8 Iosco 58 51.6 Schoolcraft 48 49.5 Iron 88 40.0 Shiawassee 198 43.8 Isabella 190 50.0 Tuscola 885 44.6 Jadkscn 885 49.5 Yan.Buren 816 45.4 Kalamazoo 880 48.9 ‘Waahtfnaw 865 43.3 Islkaska 43 37.5 Wayne 784 41.8 Kent 538 37.4 Wexford . 93 60.9 Keweenaw 35 56.9 State 15,367 44.5 1. Residence based on 1930 population. Source: Annual ' 1944, Table . Reports, Michigan Department of Health, 1935- 235 status.1 Dorn, for example, after classifying the counties of Ohio into good and poor agricultural areas (employing these classifications as indicative of economic status), found striking differences in infant death rates between the areas of good and poor economic status, the rate being more than.40 percent greater in the poor economic areas.8 It should be pointed out that the deaths in Dorn's study, as well as those in the present study, are not classified by economic status of families, thus the evidence must be regarded as an indirect corroboration of the hypothesis. Examination of Table 8.10 and Figure 8.14 showing the infant death rates for the urban segments of the counr ties of Michigan reveals a cluster having high death rates in the central part of the Lower Peninsula. These coun- ties include Isabella, which has the second highest rate in the State, namely 63.9. Urban places in the Upper Peninsula also have higher them.ayerage death rates with the exceptions of Gogebic, Iron, and Dickinson. ‘1 comparison of the infant death rates for the rural 1. See: Coombs, L.C., ”Economic Differentials in Causes of Death,7 medical Care, Vol. I, July 1941, pp. 846-855; Mott, FrederIck D. (MtD.), and Boomer, Milton I. (MtD.), Rural He th and Medic Care, MbGraw-Hill Book Company, New Tor , 9&57'p. - 3; EEgistrar General, The Be is- :rar-General's Decennial Su lament; Englandflgfig Wages, H.M§ Stationery Office, London, ngland, 1936, 701. II, p. 811, and pp. 191-810; Thompson, Warren S.,‘gp..ggt., pp e 234' 839 e 8. Born, Harold F., WMortality Rates and Economic Status in Rural ireas,’ Public as th Re orts,‘Vol. 55, No._I, January 5, 1940, pp. 3:18, eprInt No. 8186, p. 4). e‘ URBAN INFANT MORTALITY RATES {Allocated to residence) IICH GAN, AVE,, 1935-1944 = 40.7 00“ 'QUCH ’1’, N noon 5 ,, , i. ”, K easaca ‘ I K“ b ’ o’ ’ I no.» ” ' ’I ’l/ ’ ’ ’ I I, I manager? MOLCQA' tUCL I], I, u 30» LC. R uacwmac C W I], ’l / ours —\ / ‘ ' x I‘ "h ‘ - \‘Tf\ . - 9’ ‘ "1y ~ ‘ . '1 , - .' (I 1 . \\ / \ .~ «w ' \\ . ‘ ~‘s l *4 V g "(‘\ r \\ s. K‘S .\ U." _' \> O .3 ~ ~. \. , ~. .1 HA .(vou a '8! a“ vsu 9. 0 r1. 1 a ‘ l K rggx " ‘ . . son- ‘ ANYIIU 073(00 ~061sz ALPENA *" Nay I 7 mm PER 1000 r C ' LI“ 3mm ’ “"1" "" ‘ “HRS” (Iv-"om oscooA ucom 3 ,/ . w r IMAM!!! (8237):: occusv. osco f’ / new“; ’ ‘ I / , /,’ Under 38.0 g _J . ,1 A30" Lav: gstro. . CLAN ONO-“N ( (A ‘ W ' \j" 4 HUI-ON ‘ ocuu mtcosu menu "0”" MCNYCALH o ‘ 706C ; 42eO-45e9 ‘ suntwnco . 0' I ”h “ samL AC / / xclATIOT , I l,’ ’/ / 46.0-4909 '5 /”///l . ..‘IIS um (I s a '1 OTYMA ’nqu .0 (UNION wassu W. ) . n 50 .o- 54 .9 . ALLICA as v (non mg...“ ‘ , naaLau ’/ / // /’ .' / 55.0-Cver ‘Qfif ’//0' l UA‘IOO can-‘00». JACKSON wasnflasu M794! ‘ l / HON / I NC Urban POP e “2'1 u CASS ’fo’sYt/s a no. sum: LINN" . O — 1s\ Figure 8.14 (Source: Table 8.10) 236 Table 8.10 URBAN’INFANT MORTALITY RATES PER.1,000 LIVE BIRTHS IN COUNTIES OF MICHIGAN, 10 IEAR AVERAGES, BIRTHS AND DEATHS, 1935-1944 (Ex- clusive of stillbirths; by place of residence) State : Ave. : : State : Ave. : and :Number : Rate : and :Number : Eats County :Deaths : : County :Deaths : Alcona --- --- Lake --- --- Alger 55 53.1 Lapeer 48 44.5 Allegan 65 37.6 Leelana -- --- Alpena 183 45.3 Lenawee 164 54.7 Antrim. --- --- Livingston 44 51.8 Arenac --- --- Luce --- --- Baraga --- --- Mackinac --- --- Barry 47 45.5 Macomb 497 40.8 Bay 507 48.6 Menistee 79 50.8 Benzie --- --- Mhrquette 896 50.9 Berrien 344 38.9 Mason 79 48.7 Branch 74 48.8 Mecosta 46 43.9 Calhoun 631 49.0 IMenominee 136 47.3 Cass 68 60.6 Midland 143 47.9 Charlevoix 34 51.9 Missaukee --- -- Cheboygan 76 60.9 IMonroe 836 48.8 Chippewa 888 53.5 Montcalm 60 57.8 Clare ~-- --- Montmorency --- --- Clinton 51 57.4 Muskegon 886 54.8 Crawford --- --- Newaygo --- --- Delta 855 66.8 Oakland 1,866 39.3 Dickinson 168 44.6 Oceana --- --- Eaton 104 39.3 Ogemaw -- --- Emmet 77 60.8 Ontonagon --- --- Genesee 1,490 44.8 Osceola --- --- Gladwin --- --- Oscoda --- --- Gogebic 145 37.0 Otsego -- --- Grand Traverse 116 47.1 Ottawa 161 30.8 Gratiot 100 68.6 Presque Isle 41 46.9 Hillsdale 63 49.1 Roscommon --- --- Houghton 181 58.0 Saginaw 807 45.1 Huron --- --- St. Clair 411 46.9 Ingham. 733 39.7 St. Joseph 85 38.8 Ionia 111 53.0 Sanilac --- -~- Iosco --- --- Schoolcraft 64 54.5 Iron 48 37.8 Shiawassee 194 53.9 Isabella 114 63.9 Tuscola 45 61.4 Jackson. 478 43.8 van Buren 36 40.6 ‘Kalamazoo 37 89.5 Washtinaw 357 37.8 Kalkaska --- --- Wayne 13,775 38.3 Kent 1,066 33.9 Wexford 107 49.7 Keweenaw --- --- 'State 87,858 40.7 1. Residence based on 1930 population. Source: 1944, Table 8. Annug Re orts, Michigan Department of Health, 1935- 237 and urban segments of the 56 counties of Michigan which have urban places indicates that the rural segments have lower rates than their corresponding urban segments. This is true for 37 of the 56 counties. The counties with the twelve highest rates and those with the twelve lowest, are given below: Twelve Lowest Death Rates Twelve Highest Death Rates Urban Infant Deaths Urban Infant Death s Kalamazoo 89 . 5 Delta 66 . 8 Ottawa 30.; Ilabelld 63.9 St. Joseph 38. Gratiot 58.3 Kent 33. 9 Tuscola 3174' Gogebic 37 . 0 Che boygan m Allegan 3'7 . 3 Case 35:: Iron 37 . 8 Ems t A m Washtenaw 37 . 8 Clint on 37%57. Wayne 38 . 3 Montcalm , . Berrien 38 .9 Lenawee BT57 Eat on 39 . 3 Schoolcraft .5475 Oakland 39 . 5 Muskegon F478 Ten of the twelve counties with low death rates are in southern Michigan. The two exceptions are Gogebic and Iron . in the Upper Peninsula. 0n the other hand, seven of the twelve high infant rates for urban segments are also in southern Michigan. or the remaining segments, two counties are in the Upper Peninsula, namely, Delta and Schoolcraft, and three are in northern Michigan. Delta county has the - highest urban infant death rate and Kalamazoo has the low- est cne. They are 66.8 and 89.5 respectively. PART IV TRENDS OF MORTALITY CHAPTER II . MORTALITY osmorsl IN MICHIGAN Changes in the mortality conditions from one period to another is of tremendous social significance, for along with other vital phenomena they indicate the general direc- tion in which health services Should develop. In 1900 more people died from pneumonia (including influenza) than from any other cause; but today heart diseases take the great- est toll. Rupert F. Vance in his book, Allhghggg’Peo 1e, writes of man's effort to survive as taking the form.of a defense. The offense consists of all the harmful elements of environ- ment while the defense comprises both man's innate capacity to survive and the supporting environmental influences, such as medical care and health facilities, which are especially designed weapons against illness and death. He continues, "Under ideal environmental conditions the battle would be decided only when the force of inherent vitality crumbles at the end of the life span. We can measure the outcome of this conflict only in terms of performance, as in death rates, morbidity rates, and length of life. In the record . In this cha er mortality data are given, for the most 1 part, by plgte of occurrence, since relatively few States have allocated deaths to residence prior to 1938. According to Linder and Grove, “... rates for groups of registration States or for individual States would be affected very slightly, if the data were tabulated on a place-of-residence basis." Linder, Forrest E., and Grove, Robert De, _Oje Gite, De 18. 838 239 of performance, biology and environment are so intermingled that no separate accounting is undertaken, or in fact possible.'1 However, it does not seem.reasonable to assume that man's innate physical capacities are so variable that they satisfactorily explain the differing mortality rates found among the various classes of the pOpulation. A more likely assumption is that man's innate biological capacity to sur- vive varies little, and that changes in the rates at which people die during a given period reflect primarily changes in environmental conditions.2 It is the objective of this Chapter to present the changes in mortality differentials that have occurred in the population of Michigan and in certain segments of that population for selected years. This analysis will include a discussion of mortality change for the total pepulation; for age, sex, and race; for infant and maternal deaths; and for the principal causes of death.3 In order to evaluate these trends, comparable data are given for the 1e Vance, Rupert Be, _gpe Cite, Pe 335e 2. Warren S. Thompson,lgp..git., p. 240, makes a similar assumption with respect to mortality differentials between the White and colored pOpulations; compare Forrest E. Linder, 3331., pp. cit., p. 12; See also Smith, T. Lynn, 22. 2.1.1.7.” p. 1'67.- 5. the changes in.mortality for rural and urban areas cannot be determined because the data are not avail- able. Most of the States did not begin allocating deaths to place of residence until 1938. Michigan adopted this procedure in 1935. 240 death-registration States of 19101 or in some instances for the death-registration States.2 However, the death-regis- tration States of 1910 are employed almost exclusively be- cause more accurate mortality trends can be secured by using constant areas.3 In this Chapter, death rates which are ad- Justed or standardized for differences in age composition are based on the 1940 age distribution4 of the death-regis- tration States or 1900.5 During the 70 years from 1875-1944, the ten-year average crude death rate for Michigan increased from.8.8 per 1,000 population at the beginning of the period to a peak l. The death-registration States of 1910 include the fol- lowing 20 States and District of Columbia: California, Colorado, Connecticut, District of Columbia, Indiana, Maine, Maryland, Massachusetts, Michigan, Minnesota, Montana, New Hampshire, New Jersey, New York, Ohio, Pennsylvania, Rhode Island, Utah, Vermont, Washington, and Wisconsin. 2. The term ”death-registration States" refers to the ex- panding area covered by States entering the death-reg- istration system.and recording deaths by its rules. The number of States varied until 1933 when all of the 48 States were included in.the area. 3. Since the death-registration States of 1910 are em? played as the Nation in this Chapter, those death rates do not correspond to those used in Chapter VI where the death-registration States are used. 4. The choice and use of standard million pOpulations is discussed in Appendix I, p.318ff.See also Pearl, Ray- mond,'gp. cit., pp. 254-260. 5. The death-registration States of 1900 include the fol- lowing 10 States and the District of Columbia: Con- necticut, District of Columbia, Indiana, Maine, Massa- chusetts, Michigan, New Hampshire, New Jersey, New York, Rhode Island, and Vermont. 241 of 13.5 for 1905-1914,1 and it has decreased steadily since then (See Table 9.1).2 Although deaths were numerous dur- ing the great influenza epidemic of 1918-1920, the general death rate was not influenced sufficiently to equal that of the preceding decennial period, 1905-1914. Ten-year average death rates are considered better measures of mortality than annual rates, especially for computations of the occurrence of death prior to 1925 because they tend to smooth out fluctuations due to epidemics. However, the computations involved make the use of the rate impractical on a large scale. Inasmuch as rates based on decennial years such as those given in Table 9.2 are considered adequate for show- ing the general trends of mortality, these have been employ- ed in subsequent data in this Chapter. Both the crude and age-adjusted death rates given in 1. See Willcox, Walter F.,lgp. cit., Chapter 13, especially pp. 204-230. See also this dissertation,pp,29-34for a consideration of errors in reporting deaths. Presumably, the death rate in Michigan would have been much higher before 1900 if all deaths had been properly recorded. Thus some of the increases shown in these death rates up to 1910, or thereabout, may be due to inefficient registration of deaths. Willcox believes that few com- munities have absolutely complete registration of deaths, the largest omissions being those under 1 year, especial- 1y babies who live only a few hours or days, and deaths from.accidents, for example, by drowning in Which the body is not found. 2. The general decline of the death rate in western civil- ization is indicated by Thompson, Warren S.,‘gp. cit., p. 218, thus: "... the death rate has been declining in much the same manner as the birth rate during the last few decades. In some countries, indeed, it has been declining somewhat faster than the birth rate in many Western countries.” 242 Table 9 el CRUDE MORTALITY RATES (TEN-YEAR AVERAGES) IN'MICHIGAN, 1875-1944 Rates_per 1,000 Population Year Average Number of Deaths Crude .Adjusted (l) 1875-1884(3) 14,582.4 8.79 ----(3) 1885-1894‘3) 19,388.9 9.26 ----(5’ 1895-1904(2) 28,806.5 11.90 14.87 1905-1914 57,894.4 15.48 15.83 1915-1924 47,059.0 12.83 15.50 1925-1954 51,411.4 10.62 15.05 1955-1944 53,028.2 10.09 10.96 1 Based on the standard specific death rates of 1940 for the death-registration States of 1900. 2 muehigan was included in the Death-registration States of 1900 and employed her own system.prior to this date. 3 Necessary data for age-adjustement are unavailable. Source: Computed from the Sixty-ninth Annual Report, .Michigen Department of Health, 1941,-Table 2, pp. 234- 235; Annual Reports for 1942-1945, Table 16. Table 9.2 reveal that people in.Muchigan died at lower rates than those of the Nation for each decennial period since 1900 with the exception of 1920. This exception is a result of the influenza epidemic of 1918-1920. The crude death rate from influenza in Michigan for 1920 was 81.1, compared with 70.5 per 100,000 population for the Nation, a difference of 10.6. Influenza led to pneumonia in many cases and.the difference between death rates from.pneumonia in the two areas was even greater, namely 25.6, the rate for 243 the Nation being only 207.3 compared with 232.9 for Michigan. Table 9.2 CRUD] AND AGE-ADJUSTED DEATH RATES PER 1,000 POPULATION MICHIGAN AND THE DEATH-REGISTRATI 0N STATES, 1900-1940 (By place of occurrence) Crude Death Rates2 Adjusted Death Rates1 Year Michigan U.S. Difference Michigan U.S. Difference UeSe"MiChe UeSe‘-M10he 1940 9.9 10.8 1.0 10.6 11.2 .6 1930 10.7 11.3 .6 12.4 13.0 .6 1920 14e1 13e1 -1.0 15.0 14e7 -e3 1910 14.2 14.7 .5 14.8 16.2 1.4 1900 14.0 17.2 3.2 15.1 18.0 2.9 1 Adjusted to the 1940 age distribution of the death-regis- tration States of 1900 by the direct method. 3 Source: Linder, Forrest E., and Grove, Robert D., Vital Statistics Rates‘in the United States, 1900-1940, TaEIe 4, p. $54, and Table 9, p. I56. With the exception of 1920, the differences between the adjusted death rates in Table 9.2 tend to decrease from 1900 to 1940, until for the period 1930-1940 it was quite small, only .6. Both populations are growing older, and, as will be shown later, both the Nation and Michigan have reduced most of their death rates from infectious diseases to a point where they are negligible. The overall decrease of age-adjusted death rates in Michigan and the Nation during the 40 year interval was 28.8 and 37.8 respectively. In attempting to estimate the actual number of lives saved by such decreases, Willcox, studying the death-registration States of 1900, for the period 1900-1930, wrote, "It is :perhaps fair to conclude that not fewer than 4,000,000 lives have been saved since 1900."1 le WillOOX, Walter Fe, QEe Cite, p. 217s man‘— \. ti :1 t] CI 1. 244 MMchigan has an extremely low death rate at the pres- ent time, and one might expect the rate to Show increases in the near'future in view of the rather general aging of the population. Warren S. Thompson, a leading authority on pOpulation in the United States, write of this possi- bility for the country as a whole thus: ”In the near future, therefore, we may expect to see the crude death rate (the number of deaths per thousand of the total pOpulation) cease to decline and begin to in- crease in those countries where it is now below 11 or 12. The low rates of today are possible only because pOpulation has been increasing very rapidly in the re- cent past, and the proportion of young people is very high." Mortality Trends 331 Ag; The age-specific death rates in Michigan and the Nation for the decennial years, 1910-1940, are given in Table 9.3. The death rates for both populations decreased more than 80 percent at ages 1-4, all other groups having relatively lower declines. Except for the group under 1 year, death rates for all age groups under 35 in both Mich- igan.and the Nation decreased more than 50 percent during the period; but from.35 years of age upward the percentage declines are less than 50 percent. (See percentage de- crease column in Table 9.3) All age-specific death rates for both.Michigan and the United States decreased from one decennial year to the next, with the exception of 1920. The people of Michigan 1m Thompson, Warren S.,‘gp. cit., p. 219. Table 9e3 245 AGE SPECIFIC DEATH RATES: PER 1,000 POPULATION MICHIGAN AND THE DEATH REGISTRATION STATES OF 1910 FOR 1910-1940 (By place of occurrence) Age Group Year Percent Decrease 1910-1940 1916 1920 A1950 1946 _ Che eSe All ages: U.Se l4e7 13e4 11e5 llel 24e5 Mich. 14.2 14.1 10.7 9.9 30.3 ‘Under 1: UeSe 132e2 101.2 68e5 48e8 63e1 Miche 127e5 llOeO 69e7 48el 62e3 1 ‘ 4: UeSe 14e° 9e9 4e9 2.2 84e3 Mich. 11.3 10.8 5.2 3e2 80.5 5 -14: UeSe 2e9 2e? 1e? 1.0 65.5 Mich. 2e9 3.3 1.8 1.0 65.5 15-24: U.S. 4.6 4.6 2.9 1.7 63.1 M1011. 4.7 5.4 3.1 1.8 61.7 25-34: U.S. 6.6 6.6 4.2 2.6 60.6 Mich. see 7.3 4.1 2e5 56.9 55‘443 U.Se gel 8.1 6e4 4e7 48.4 Mich. 7.3 8.2 6.3 4.6 37.0 45-54: UeSe 13e9 12e2 12e0 10e3 25e9 Much. 11e6 lle5 10e8 9e5 18el 55‘64: U.Se 26e5 24e5 24e5 22.7 14e3 Mich. 22e5 23.1 22.0 21e1 6e2 65‘74: UeSe 56.0 54e3 52e6 49e9 10e9 NUOh. 52e2 52.0 50e2 48e2 7e? '75-'84: U.S. 123.1 120.5 116.1 115.8 5.9 Mich. 124.7 124.0 117 . 5 114.4 8. 3 85-Up: U.S. 252e4 246e5 235e0 230e6 8.6 Miche 264e4 263e8 237e9 236e0 10e8 Source: Linder, Forrest E., and Grove, Robert D., Vital Statistics Rates in.the United States, 1900-I940, Table 4, p.I154; Table 9, p. I 8 . 246 died at slightly higher rates in 1920 than in 1910. This is true of all age groups from 15 through 64 years, a result, undoubtedly, of the toll taken by the influenza and pneu- nonia epidemic of 1918-1920. In contrast, the death-regis- tration States of 1910 showed no increases in age-specific 1eath rates throughout the 30 year period, although for the years 1910 and 1920 the age-specific rates of the death-re- gistration States for ages 15-34 remain the same. There were slight increases in death rates for two of the older age groups in Michigan, namely, 55-64 for the years 1910 1nd 1920, and 65-74 for 1920 and 1930. There was no change, :ither increase or decrease, in the death rate of the United States for age 55-64 for 1920 and 1930. In previous Chapters, it was pointed out that for L940 certain pairs of age groups in both Michigan and the Inited States had age-specific death rates which were ap- proximately the same. These pairs were: (a) all ages and age group 45-54, (b) those under 1 year and age group 65-75, Le) age group 1-4 and 25-34. An examination of the data in Fable 9.4 indicates that death rates for the pairs are similar for the years 1930 and 1940 but are quite dissimilar for earlier years. Moreover, differences between them.grow Larger with each preceding decennial year, with the exception Of the first pair. In both Michigan and the United States for 1930 differences for the first pair were less than might be expected. Thus for these pairs of age groups, an equilib- rium.seams to have been reached which will probably persist 247 .n.0 canoe “moves space no eonoom H.» m.m m.. H. 0.0 m.m H.H n. ooooaonufin 0.0 0.0 «.0 0.0 0.0 a.» H.¢ 0.0 «0:00 a o.¢H 0.0 0.0 0.0 n.HH m.oa 0.0 0.0 e c H «.0» 0.w¢ 0.0 H.H 9.00 o.mm n.0H H. oooosounan 0.00 n.¢n 0.00 0.00 0.00 0.00 «.00 «.00 «0:00 N a.mna m.aoa 0.00 0.00 n.>ma 0.0HH 0.00 a.md .AM A pecan m. N.H m. m. 0.0 0.0 H. d. oooonouuan 0.0a «.0H o.mH n.0a o.HH m.HH 0.0H 0.0 amine H b.0H ¢.nH n.HH H.HH m.¢a H.0H b.0H 0.0 .nowo Had 0HOH om0H 000a 000a 0HOH om0H 000a 000a nopem nepopm voodoo onwaoofiz noose owd no chasm Odaanoaod mom 0HOH ho mfiadem 20H84mamHowmlm94MQ mmB Qz¢ 240HmUH2 .mfia4m me4mn OHMHommmlmcd nmeomnmm 248 until new methods of control are developed in one or the other of the contrasting age groups. Mortality Differentialgngy Sex For each decade from 1910 to 1940, the death rates of males and females decreased in both the Nation and Michi- gan. (See Table 9.6) The age-adjusted death rate of males in Michigan for the year 1940 is 23.8 percent lower than it was in 1910, the rates being 11.81 and 15.49 respectively. Over the same period, the death rates of males in the Nation decreased slightly more than in Michigan, or 28.2 percent. The rates were 12.48 and 17.37 per 1,000 population. Al- though this is a sizable improvement, the death rate of females decreased proportionately more over the same years, Michigan again having a lower percentage decrease than the Nation: ‘ Percentage decrease of age-adjusted death rates 1910-1940 (See Table ._A 9.6); Males Females Michigan 23.8 33.2 United States 28.2 . 36.4 The death rates of both males and females in Michigan for the year 1910 were lower~than corresponding rates for the. Nation, and therefore, would not be expected to show higher percentage decreases from that year to 1940. The crude death rates decrease in a similar manner, females having higher percentage decreases than males in both areas. However, the crude death rates for both males and females in Michigan decreased more than the correspond- 249 ing rates for the Nation, for the decennial years 1910-1940. This is due primarily to the more rapid aging of the population of the United States during the 30 years. In 1940 there were 7.9 percent more people 45 years or over in the United States than in.1910, whereas the corresponding figure for Michigan was 3.9 percent (Table 9.5). Table 9.5 PROPORTION OF POPULATION, NECHIGAN AND DEATH-REGISTRATION STATES OF 1910 FOR 1910 and 1940, BY AGE GROUPS I"; I; Percentage distribution ofpopulation' Age ifoups 19ZOch1gnn 1910 1940mmd St:::: 411 Ages 100.0 100.0 100.0 100.0 Inder 15 25.0 29.6 22.8 28.7 L5 - 44 48.5 47.8 48.5 50.5 15 - 64 20.2 18.9 21.5 16.0 55 - Over 8.5 5.7 7.4 5.0 total No. 5,258,108 2,810,175 70,859,155 47,250,895 SOurce: Computed from.Vita1 Statistics Rates in the United States 1900-1974OT‘PoWaBIFEI 'T—e'b‘I'e‘IIT‘ia‘.‘ 'EOETUTaHIE'IIT‘E. 1008. As for the decennial periods between the years 1910 1nd 1940, the year.1920 is exceptional in that death rates lo not follow expected trends because of the influenza epi- lemic. Females had particularly high rates in Michigan dur- .ng the epidemic years. Both the crude and age-adjusted death :ates were higher in 1920 than in 1910 or 1930. The adjusted rate increased 5.9 percent from 1910 to 1920 and the crude 250 ate 3.7 percent. In contrast, both the crude and age-ad- Isted death rates of females in the United States decreased, 16 percentage decrease being 5.8 and 5.1 respectively (See able 9.6). It is probable that women, because of their ame nursing responsibilities, were actually more exposed 18D men to influenza. Moreover, the disease seemed to be Irticularly fatal to pregnant women.1 The death rate of ales decreased less than one percent in.Michigan as com- Lred with 11.1 percent for the United States. By 1930 death rates for both males and females in .Chigan had decreased substantially. The age-adjusted rath rate of females dropped 23 percent from 1920 to 1930 Lereas the corresponding rates for males dropped only 14.1 rrcent. These percentage declines in Michigan were greater ,an corresponding decreases for the Nation, Which were 19.5 rcent for females and 9.3 percent for males. From 1930 to 1940, the age-adjusted rate for females 1 Michigan fell from 11.50 to 9.42, a decrease of 18.1 per- tnt. During the same period, the rates for females of the Lited States decreased only 16.7 percent. The percentage poreases in age-adjusted death rates from 1930 to 1940 were >nsiderably less for males than for females in both.Nuchigan Ld the United States. These decreases were 10.6 percent in [___7 The death rate from puerperal causes in the United States was 15.3 in 1910, 17.8 in 1920 and 10.6 in 1940. For the peak year of the epidemic (1918) it was 21.7. The death rate from puerperal causes in.Michigan was 16.5 in 1915 and in 1920 it was 23.2. 251 eases .oeeauooea eeeeem sesame .eea .a .n eases eae .emH .a .e mopom moapmfipopm as» P Souk mopoh Spoon oosho "eonoom .oceoov meadooonm Hobo open ounce oa owosoo pooouom m .eeeuea eeeeae one an coed no eeeeem eoaeeeeeawem-epeeo one no seapennnpeae ewe oeea on» e» eeeeefiee a Ha.ea ee.ea e.na m.nH eaesoa ee.ma ee.ea e.ea e.na tees oaea e.e x ee.ea H.n . ma.ea a.» x e.na a.» - o.na eaesee e. . mn.ma H.HH\ me.ma e.e - m.ea e.oa- e.ea sass ewes o.nm- on.aa n.ea- He.aa a.mm- e.e m.eH- e.oa tosses H.5H- Hm.na e.e . Ho.ea 5.0m- n.HH m.oa- e.~H odes . ones a.man «4.0 e.eH- on.e H.0H- e.m m.e- o.oH nausea e.oa- Hm.aa e.oa- m..ma a.» . e.oa m. - a.ma odes owes owoooo opsm ewoooo spam ewoooo spam emosno ovum Hem use» manoonom consumed misconom copmsnod mvooonom ecsno muooonom eosho sameness neeepm eoeaeo sameness ..eeem eopaep ooapdaooom ceded Hon mopom Jason Now am 63733 mom 33 mezsmo azMomom 924 .on .eoooanoooo Ho ocean hmv mo mmaeem 20HaemamHemm.aoean m aeopmom ooo.H mam mmeem madam B a: 24.38% .mmfiomn am anmambhndumu4 Q24 mnbmo 252 Michigan and 10.9 in the United States. Thus, the percentage decrease for male deaths was greater for the Nation than for Michigan although the death rates in Michigan for both years were lower. Michigan, in most instances, has lower crude and ad- justed death rates for the period under consideration than the United States. In 1920 death rates in the State for both sexes exceeded those of the Nation. This was true also for the age-adjusted death rate for females in 1930. However, it was only .09 in excess of the rate for the United States. The mortality differentials between Michigan and the United States for each sex are given in Table 9.7. Table 9.7 DIFFERENCES IN DEATH RATES BETWEEN MICHIGAN AND Tim DEATH-REGISTRATION STATES OF 1910 FOR EACH SEX, 1910-1940 Excess of Death Rates of the United States over Michigan Crude Adjusted Year Male Female Male Female 1940 1.4 1.1 .67 .08 1930 1.1 .6 .80 -.09 1920 - .3 - .9 .07 -.76 1910 .7 .4 1.88 .83 Source: Based on death rates given in Table 9.6. Death rates for males are higher than those for females in both Michigan and the United States for all the decennial years under study. The excess death rates of males over fe- 253 males is given in Table 9.8. These sex-mortality differentials increase from 1910 to 1940 with the exception of the epidemic year'of 1920 for Michigan. No satisfactory explanations of the excess of death rates of males over those of females have been given, although authorities such as Willcoxl and Pearl2 attribute this mortality differential primarily to biological differences.3 Table 9 . 8 DIFFERENCES IN SEX-SPECIFIC DEA’Hi RATES, MICHIGAN AND THE DEATH-REGISTRATI ON STATES OF 1910 FOR 1910-1940 Excess Death Rates of Males over Females Year Crude Adjusted . United States Michigan United States Michigan 1940 2.3 2.0 2.98 2.39 1930 1.9 1.4 2.60 1.71 1920 .9 .3 1.27 .44 1910 1.8 1.5 2.43 1.38 Source: Based on death rates in Table 9.6. Perhaps the most accurate estimates of the changes that have taken place during the 30 year period as a result of the force of mortality in Michigan may be obtained by ex- amining the age-specific death rates by sex, presented in Table 9.9 and their change from 1910 to 1940. From these 1. See Willcox, Walter F., 93. cit., p. 120. 2. See Pearl, Raymond, 93. cit., p. 183. 3. See Appendix IV for an elaboration of a social causation hypothesis of the sex-mortality differential; it is referred to here only as a possible explanation since it is not formally tested in this dissertation. Table 9.9 254 AGE AND SEIFSPECIFIC DEATH RATES FROM ALL CAUSES PER 1,000 POPULATION, MICHIGAN, 1910 AND 1940 (By place of occurrence) . 1910 M518 :Female: Male :Female :Male fiemale :Death rateséger 1,000 pops 1940 Difference 1910 - 1940 :100 X Diff. :1910 Speg.Rate Ages Male Female Dotal Jrude 14.9 15.4 10.9 8.9 a 4.0 - 4.5 28.8 55.8 3 - 1 142.8 111.8 55.1 40.8 -87.7 -71.0 81.4 85.5 i - 4 12.0 10.7 2.5 2.0 - 9.7 ~ 8.7 80.8 81.5 5 -14 5.0 2.8 1.1 0.8 - 1.9 - 2.0 85.5 71.4 15-24 4.6 4.8 2.0 1.5 - 2.6 - 5.5 56.5 88.8 25-54 5.8 5.8 2.8 2.2 - 5.0 - 3.6 51.7 82.1 35-44 7.5 7.5 5.0 4.1 - 2.5 - 5.2 51.5 45.8 45-54 12.0 11.2 10.9 7.9 - 1.1 - 5.5 9.2 29.5 55-64 25.9 21.0 24.1 17.9 f .2 - 5.1 f .8 14.8 55-74 54.9 49.5 55.4 42.8 - 1.5 - 8.5 2.7 15.2 75-84 129.9 119.2 122.5 108.8 - 7.4 -12.8 5.7 10.8 35-up 274.7 254.5 247.4 227.0 -27.5 -27.5 9.9 10.7 Dotal 15.49 14.11 11.81 9.42 -3.68 -4.89 23.8 55.2 idjusted (l) :1) Adjusted to the 1940 age distribution of the death-regis- tration States of 1900 by the direct method. Source: Specific Rates are from Linder, Forreit Eh, gnit d Grove Robert D. Vital Statistics Ra as n e n 0 States 1900-1940: BHreau of the Census, Table 5, p. I54. 255 data the following generalizations may be made: 1. Death rates by age and sex.have declined remarkably in.Michigan during the 30 year interval from 1910 to 1940. Both of the sexes in all but one age group (55-64) had lower death rates in 1940 than in 1910. Therefore, to the extent that age and sex mortality rates neasure the health of the pOpulation, the improvement in health conditions in Michigan during the period studied seems quite impressive. 2. Declines in death rates were greatest at the beginn- ing of the life cycle and in early youth, and smallest at the end. This is true for both of the sexes. Since the death rates are normally highest in the youngest and the oldest age groups, it is to be expected that most of the decreases would occur in these age groups. However, inas- much as the death rates of the younger segments of the popu- lation cannot continue to fall greatly some of them being extremely low at present, further improvements will have to come by speeding up decreases among the older age groups. It would seem.that no further reduction of mortality among the aged can be anticipated unless medical science learns the secret of staying the degenerative processes. Inasmuch as the social environment is a conditioning factor in these processes, some control, within biological limits, should be p08 sible e 3. Death rates of females declined more than those of 256 males for all age groups.1 In three age groups (45-74) percentage declines for females were three or more times the magnitude of corresponding declines for males. For the years 15-45, which include the childbearing period for women and the most productive working period for men, the reduction in the death rate among females is from 8 to 12 percent greater than that for males. The changes that have taken place in age-sex-mor- tality rates in the United States during the 30 year period are similar to those which have occurred in Michigan (See Tables 9.10 and 9.11). The death rates for males were higher than.thcse for females in 1910 and in 1940 for all age groups. .Although the death rates of males were higher, and thus could be expected to have had greater percentage declines from 1910-1940, the percentage decreases at all ages were lower for males than for females. This was also true for Michigan. A comparison of sex-specific death rates‘by age for the Nation and for Midhigan shows that during the 30 year . interval the percentage changes have been less for Michigan than for the United States. (See Table 9.11.)Among females, there were three age groups for which declines were slightly greater in.Michigan, namely, those 15-24, and the two age groups 75-84 and 85 years and over. When compared with the 1. Pearl, Raymond, o .‘git.,pp. 180-183, shows that during the period 1910- 930 death rates for the registration States of 1910 decreases more for females than males from birth up to the age of 20 and from 50 years upward. 257 Table 9.10 AGE AND SEXFSPECIFIC DEATH RATES FROM ALL CAUSES PER 1, OOO POPULATION, REGISTRATION STATES OF 1910 AND DISTRICT OF COLUMBIA, 1910 AND 1940 (By place of occurrence) Death Rate : r 1 000 pepulation Difference:100 X Difference 19IO : I940 1910-1940 : 9 Dee Ba 8 Age Male:Female : Male:Female MalezFemale: Male:Female Total Crude 15.6 13.8 12.3 10.0 3.3 3.8 21.2 27.5 0 - 1 146.0 118.0 55.2 42.2 90.8 75.8 62.2 64.2 1 - 4 14.7 13.4 8.4 52.1 12.3 11.3 83.7 84.3 5 -14 3.0 2.9 1.1 .8 1.9 2.1 63.3 72.4 15-24 4.9 4.2 2.0 1.4 2.9 2.8 59.2 66.7 25-34 7.0 6.2 2.9 2.3 4.1 3.9 58.6 62.9 35-44 10.0 8.0 5.4 4.0 4.6 4.0 46.0 50.0 45-54 15.4 12.2 12.2 8.3 3.2 5.9 20.8 52.0 55-64 29.0 23.8 26.8 18.4 2.2 5.4 7.6 22.7 65-74 59.2 52.7 56.4 43.8 2.8 8.9 4.7 16.9 75-84 128.5 118.3 124.3 108.3 4.2 10.0 3.3 '8.5 85-up 258.0 248.0 243.5 221.7 14.5 26.3 5.6 10.6 Total 17.37 14.94 12.48 9.50 4.89 5.44 28.2 36.4 Adjusted (1) (1) {Adjusted to the 1940 age distribution of the death-regis- tration States of 1910 by the direct method. Source: Crude death rates from.Linder, Forrest E., and Grove, Robert D., Vital Statistics inzthe United States 1900-1940, Table 9, De I88e 258 Table 9 e 11 DIFFERENCES IN AGE-SEXFSPECIFIC DEATH RATES AND PERCENTAGE DECLINE, MICHIGAN AND DEATH-REGISTRATION STATES OF 1910 FOR 1910 AND 1940 Excess of Death Rate 1910 over 1940 Percentage Decline . 1910 to 1940 Age : : 100 X Difference f 1910 SpecIfic Death Rate Male : Female : Male : Female , .s. Mich.:T§. Mich. : U.S. Mi€h.: U.S. Mich. Total Crude 3.3 4.0 3.8 4.5 21.2 26.8 27.5 33.6 0 - 1 90.8 87.7 75.8 71.0 62.2 61.4 64.2 63.5 1 - 4 12.3 9.7 11.3 8.7 83.7 80.8 84.3 81.3 5 -14 1.9 1.9 2.1 2.0 63.3 63.3 72.4 71.4 15-24 2.9 2.6 2.8 3.3 59.2 56.5 66.7 68.8 25-34 4.1 3.0 3.9 3.6 58.6 51.7 62.9 62.1 35-44 4.6 2.3 4.0 3.2 46.0 31.5 50.0 43.8 45-54 3.2 1.1 3.9 3.3 20.8 9.2 32.0 29.5 55-64 2.2 - .2 5.4 5.1 7.8 K .8 22.7 14.8 65-74 2.8 1.5 8.9 6.5 4.7 2.7 16.9 13.2 75-84 4.2 7.4 10.0 12.6 3.3 5.7 8.5 10.6 85-up 14.5 27.3 26.3 27.3 5.6 9.9 10.6 10.7 Total 4.89 3.68 5.44 4.69 28.2 23.8 36.4 33.2 Adjusted (1) (1) .Adjusted to the 1940 age distribution of the death-regis- tration States of 1910 by the direct method. Source: Tables 9.9 and 9.10. 259 Nation, males in.Michigan fared even worse. Only two age groups had greater decreases than the Nation, those 75-84 and those 85 years and over. However, these prOportionate- 1y lower mortality Changes are to be expected, because Michigan had lower age-sex-specific death rates than the United States in both 1910 and 1940 for all except three of the age groups. Mortality Changes bthace The mortality changes in the white and colored groups of’Michigan are found in Table 9.12 where they may be com- pared with corresponding improvements in the Nation. Death rates are higher for colored peOple than for white people for the four decennial years under study. This is true for both the crude and the age-adjusted death rates. The mor- tality differentials by race, based on Table 9.12 are as follows: Excess of Death Rates of Colored Year Population Over White Crude ‘Adjusted 1940 2.8 6.09 1930 6.2 10.22 1920 11.9 15.04 1910 11.0 9.85 The excess mortality of colored over white people was great- est in 1920, at which time the adjusted death rate for non- whites was more than twice that of the whites, namely, 29.84 as compared with 14.80 per 1,000 pOpulation. Moreover, the crude death rate in.Muchigan for 1920 was 86.2 percent higher for colored than for'white peOple. The largest mortality 260 Table 9.12 CRUDE AND ACE-ADJUSTED1 DEATH.RATES PER 1,000 POPULATION, MICHIGAN AND THE DEATH+RECISTRATION STATES OF 1910 FOR 1910- 1940, BY RACE (Data by place of occurrence) Death ratesgper 1,000_popu1ation Year Race Crude : Adjusted :United 2 2 United Michigan :States : Michigan : States 1940 Colored 12.6 14.2 16.46 16.89 White 9.8 11.0 10.37 10.75 1930 Colored 16.6 17.3 22.23 22.06 White 10.4 11.2 12.01 12.45 1920 COlorad 25e7 21.1 29.84 24.13 White 13e8 13.2 14.80 14.58 1910 COlored 25e1 21e9 24.85 24.45 White 14.1 14.6 15.00 15.96 1 .Adjusted to the 1940 age distribution of the death- registration States of 1900 by the direct method. Source: Crude death.rates from Vital Statistics Rates'ig the United States 1900-1940, Table 9, p. 189, TabIe 4, P. 154. 261 differential by race (15.04) occurred in 1920, but the dif- ference has become less with each succeeding decade, indicat- ing that mortality conditions among the two pepulatiois have become more nearly equal each ten-year period. Death rates in.Michigan have decreased relatively more for the colored than for the white pOpulation from 1910-1940. This is true for both the crude and the age-adjusted death rates. The percentage declines of each racial group for the period under consideration are indicated below: Percentage Decline of Death Race Rates, 1910 to 1940 Crude Adjusted White 30e5 ’ 50.9 COlOrOd 49.8 33.8 A comparison of mortality rates by age groups for 1910 and 1940 given in Table 9.13 permits a more refined saalysis of the changes which have taken place among the racial groups. Nonewhites in Michigan have Shown greater percentage declines in death rates than whites for all age groups under 25 years of age and for those 65 years and over. The specific death rate of colored infants under 1 year of age was phenomenally high in 1910,-- being approximately identical with the rate of the oldest age group presented, those 85 years of age and over. The rates were 188.6 and 188.7 respectively. By 1940 the death rate for colored infants under 1 year of age had declined to 67.2, a reduction of 64.4 percent. In 1910 the rate for White infants under 1 year of age was 127.0 and by 1940 it had dropped to 47.3 per 1,000 pOpulation, a decline 262 Table 9.13 RACE SPECIFIC DEATH RATES PER 1,000 POPULATION AND PERCENT CHANGE, MICHIGAN, 1910 AND 1940, BY AGE (By place of occur- rence) “It : Percent decline: :100 X Difference 1910-40 Age 1910 1940 : Specific Rate I910 All All : All White Other White Other : White Other Total GrUde 14.1 25.1 9.8 12.6 30.5 49.8 0 - 1 127.0 188.6 47.3 67.2 62.8 64.4 1 - 4 11.3 21.1 2.1 3.9 81.4 81.5 5 -14 2.9 8.9 1.0 1.4 65.5 84.3 15-24 4.6 14.6 1.6 5.0 65.2 65.8 25-34 5.7 15.1 2.2 6.9 61.4 54.3 55‘44 7.2 15.7 4.2 11.0 41.7 29.9 45-54 11.5 26.1 9.0 21.9 21.7 16.1 55-64 22.4 55.7 20.8 58.0 7.2 i 6.8 65-74 52.1 69.8 47.9 57.5 8.1 17.6 75‘84 124.6 134.9 114.3 122.4 8.5 9.3 85-up 265.7 188.7 238.9 126.1 10.1 33.2 TOtal 15.00 24.85 10.57 16.46 30.9 33.8 Adjusted (l) (1) .Adjusted to the 1940 age distribution of the death-regis- tration States of 1900 by the direct method. Source: Linder, Forrest E., and Grove, Rebert D., Vital ‘§tatistics Rates'in the United States 1900-1945, Tabie 4' P. 154. 263 or 62.8 percent. As expected, both the white and colored populations had relatively small declines among the age groups 45 years and over. Instead of decreasing, the death rate for one age group among the nonrwhites, those 55-64, increased from.33.7 per 1,000 pOpulation in 1910 to 56.0 in 1940. .As was true in Michigan, crude and age-adjusted death rates for colored people in the Nation were higher than for Ihit. people from 1910 to 1940, but the differences by race are less than corresponding differences in Michigan, except for the year 1940. These differences are given in Table 9.14. Age-adjusted death rates for the white segment of the Nation were slightly higher than those for Michigan except for the year 1920. (See Table 9.15.) The influenza epidemic took a larger proportion of lives in Michigan than in the United States in 1920, which probably explains this exception. Table 9.14 DIFFERENCES IN DEATH RATES, NECHIGAN AND THE DEATH REGISTRATION STATES OF 1910 FOR 1910 and 1940, BY RACE Excess of Death Rates or colored over death rates of White._l910 and 1940 Year Michigan United States Crude Adjusted Drude .Adjusted_ 1940 2.8 6.09 3.2 6.14 1930 6.2 10.22 6.1 9.61 19 20 * 11.9 15 .04 7.9 9.55 1910 11.0 9.85 7.3 8.49 Source: Computed from Table 9.12. 264 Table 9 . 15 DIFFERENCES IN AGE-ADJUSTED DEATH RATES, MICHIGAN AND THE DEATH REGISTRATION STATES OF 1910 FOR 1910 TO 1940 Adjusted death rates and differences Colored White Year Death Rates : : Death Rates: Tich. : U.S. : Difference: Mich.: U. 8.: Difference 1940 16.46 16.89 .43 10.37 10.75 .38 1930 22.23 22.06 .17 12.01 12.45 .44 1920 29.84 24.13 5.71 14.80 14.58 .22 1910 24.85 24.45 .40 15.00 15.96 .96 Source: Computed from Table 9.12. In contrast to the trend for the white race, age-ad- Justed death rates for the colored segment of the pOpulation in the United States were lower than those in Michigan except for 1940. .A contributing factor might be differences in re- porting procedures. Michigan's non-white pOpulation was lo- cated chiefly in Detroit where reporting techniques were well developed in contrast to areas of the United States in which the registration system.has operated inefficiently un- til recently. Differences between the age-adjusted death .rates of the non-white population in.Muchigan and the United States together with the corresponding differences for the white population in these two areas are given above. This difference for colored persons was excessive for Michigan in 1920. It is probable that both general levels of living and lTailability of facilities were contributing factors to a high death rate among the non-whites during the influenza 265 epidemic. Crude death rates have decreased relatively more for the non-white than for the white pOpulation in the Nation which is comparable to the trends in Michigan.1 However, when the death rates for the United States are adjusted for age, the situation is reversed, the white population de- creasing at a greater rate than the non-whites. This is not true for Michigan. The percentage decreases are pre- sented below: Percentage decline of death rates, 1910 to 1940 Michigan United States Crude Adjusted Crude Adjusted White 30.5 30.9 2497 32.7 Colored 49.8 33.8 35.2 30.9 Source: Computed from Tables 9.13 and 9.16. .Age-specific death rates by race for 1910 and 1940 and percentage declines for the Nation are presented in Table 9.16. White and colored segments have approximately the same percentage decreases for those under 15 years, but above this age the whites have a definite advantage for all age groups.2 This is contrary to the corresponding differentials 1. See Mott, Frederick D'a._E°.§A's.EE' ci ., p. 56; and Kemp, Louise, 23.31.,.gp. c t., p. , for comparable findings. 2. 0n the basis of life-table death rates Warren s. Thompson arrived at similar conclusions for the death-registration States of 1920 for the period 1920-1940,‘gp. cit., p. 140. 266 Table 9.16 RACE SPECIFIC DEATH RATES PER 1,000 POPULATION AND PERCENT CHANGE, DEATH-REGISTRATION STATES OF 1910 for 1910 AND 1940, BY AGE (Data by place of occurrence) : Percent decline: :100 X Difference 1910-4O Age 1910 1940 : Specific Rate 1910 A11 A11 : All White Other White Other : White Other Total Crude 14.6 21.9 11.0 14.2 24.7 55.2 0 - 1 129.7 241.9 47.0 87.4 63.8 63.9 1 ’ 4 13.7 28.6 2.1 4.3 84.7 85.0 5 '14 2.9 5.7 .9 1.7 69.0 70.2 15'24 4.4 10.4 1.6 5.0 63.6 51.9 25‘34 6.4 12.3 2.3 7.3 64.1 40.7 55’44 8.8 17.3 4.4 11.7 50.0 52.4 45-54 13.6 25.0 9.8 21.9 28.0 12.4 55-64 26.2 39.0 22.2 37.5 15.3 3.9 65'74 55.8 65.0 49.6 62.3 11.1 4.2 75-84 123.5 104.5 115.8 115.1 _5.2 w f 10.4 85-up 254.6 177.5 252.2 178.2 8.8 { .5 Total 15.96 24.45 10.75 16.89 32.7 50.9 Adjusted (1) (1) Adjusted to the 1940 age distribution of the death-regis- tration States of 1900 by the direct method. Source: Linder, Forrest E., and Grove, Robert D., Vital Statistics Rates in the United States 1900-1945, Table 9, P. 189. 267 in.Michigan, the colored population having higher declines than the white for ages 65 and over. A comparison of mortality changes by sex and race in Michigan is presented for the years 1910 and 1940 in Table 9.17 and corresponding data for the Nation in Table 9.18.1 In general, these data show that the death rates of both males and females in.the two racial categories have decreas- ed substantially during this 30 year interval. In Michigan the actual differences in the rates for both sexes are greater for the colored than the white groups, although the percentage decreases are greater for the White segments. (Table 9.17) The excess of the age-adjusted sex specific death rates by race in the Nation given in Table 9.18 below over corresponding rates for Michigan in Table 9.17 indicate that the differences were much greater in 1910 than in 1940. For example, colored males in Michigan and the Nation had death rates of 27.77 and 33.43 respectively in 1910, a difference of 5.66; while in 1940 corresponding death rates were 20.33 and 20.87, giving a difference of .54. The mortality dif- ferential for colored females in Michigan and the Nation de- creased from 5.30 in 1910 to approximately equal rates in 1940. The differentials for males and females of the white 1. Adjusted death rates in Tables 9.17 and 9.18 are computed by the indirect method, which employs standard specific death rates as a standard rather than a standard million population. It was necessary to use the indirect method in adjusting the death rates by sex and race because the age-specific death rates, required for the direct method, were not available for Michigan. 268 Table 9 .17 AGE-ADJUSTEDl DEATH RATES PER 1,000 POPULATION AND DIFFERENCES MICHIGAN, 1910 AND 1940, BY RACE AND SEX I! J Death Rates per 1,000 population Year Colored White Male Female Male Female 1910 27.77 25.58 17.13 15.56 1940 3.91.9.9. 12.-.29. 1119.9. .2.-.19. Difference 7.44 7.92 5.53 6.37 100 X Difference 26.8 31.0 32.3 40.9 1910 Death Rate 1 Based on the standard specific death rates of 1940 for the death-registration States of 1900. Table 9 . 18 AGE-ADJUSTED1 DEATH RATES PER 1,000 POPULATION AND DIFFERENCES DEATH-REGISTRATION STATES OF 1910, 1910 AND 1940, BY RACE AND SEX Death ratesgper 1,000 population Year . Colored White Male Female Male Female 1910 33.43 30.88 19.68 16.63 1940 M 311.292 $3.213}; .2.-3’2?! Difference 12.56 13.25 7.47 7.30 100 X Diffaraaaa 37.6 42.9 38.0 45.9 1910 Death Rate 1 Based on the 1940 standard specific death rates of the death-registration States of 1900. 269 race follow the same trend, although in 1910 the differ- ences between.Michigan.and the Nation were not so great for either males or females. Thus, it appears that mor- tality differentials between Michigan and the Nation by sex and race are gradually being leveled. However, the percentage decline for each sex by race is still greater for the white than the colored population. Mortality flange 111 Cause The causes of death presented in Table 9.19 reveal that in Michigan death rates from most of the leading diseases in 1910 had been reduced considerably by 1940. Among the ten principal causes in 1910 only two, heart diseases and cancer, had higher death rates in 1940. These two causes of death increased primarily because of the aging population in.Michigan. Diabetes mellitus, another degenerative disease, increased 80.4 percent, and by 1940 was among the ten principal causes.1 This cause ranked fifteenth in 1910 and I“ ninth in 1940.. Both suicide and homicide are among those causes of death which increased between the years 1910 and 1940, their percentage increases being respectively 2.2 and 20.0. Death rates from.motor vehicle accident multiplied at a phenomenal rate, namely 1,722.2 percent, or from 1.8 per 100,000 in 1910 to 32.8 in 1940. This trend coincides 1. This corroborates the findings reported by Bortz. See Stieglitz, Edward J.,.gp. cit., pp. 226-227. 270 Table 9.19 DEATH RATES PER 100,000 POPULATION BY CAUSE, MICHIGAN, AND PERCENT CHANGE, 1910 TO 1940 (By place of occurrence) Death rates_per 100,000 population Cause of death Michigan Diff- $00 x‘DIfft 1910 1940 erence Death rate 1910 Diseases of the heart 167.0 293.6 126.6 { 75.8 Pneumonia.& influenza 109.7 52.8 56.9 51.9 Tuberculosis (all forms)- 97.2 33.6 63.6 65.4 Diarrhea, enteritis, etc. 103.0 4.6 98.4 95.5 Intracranial lesions of 104.7 89.5 15.2 14.5 vascular origin Nephritie 68.3 56.7 11.6 17.0 Accidents (non-vehicle) .74.1 42.1 32.0 43.2 Cancer (all forms) 74.6 119.4 44.8 / 60.1 Premature births 39.3 24.0 15.3 38.9 Bronchitis 26.3 3.2 23.1 87.8 Typhoid, paratyphoid fever 23.5 ._;§ 23.3 99.1 Diphtheria 17.5 .4 17.1 97.7 Diabetes mellitus 14.8 2677 11.9 f 80.4 Puerperal causes 16.7 5.5 11.2 67.1 Suicide 15.6 15.9 .5 # 2.2 Congenital malformation 15.9 13.9 2.0 12.6 Syphilis 12.6 11.5 1.1 8.7 Cirrhosis of the liver 12.2 9.2 3.0 24.6 measles 9.0 ‘_;§ 8.7 96.7 Hernia & intestinal obs. 13.8 9.2 4.6 33.3 WhOOping cough 10.9 1.1 9.8 89.9 Scarlet fever 10.7 .8 9.9 92.5 Appendicitis 12.7 9.8 2.9 22.8 Dysentery 6.4 .3 6.1 95.3 Alcoholism 3.6 172' ‘ 2.4 66.7 Homicide 2.5 5.0 .5 J 20.0 Ulcer of the stomach 5.5 6.0 2.5 J 71.4 Diseases of prostate 4.0 5.7 1.7 f 42.5 Biliary calculi 2.9 5.7 .8, f 27.6 Accidents, vehicle 1.8 32.8 51.0 f1722.2 Goiter, exOphthalmic 2.4 5.7 1.5 / 54.2 malaria 1 .7 .0 .7 100.0 Cerebrospinal meningitis _;§ .4 .2_ / 100.0 Senility 51.5 6.4 44.9 87.5 Ill-defined causes 32.6 4.1 28.5 87.4 All Gausas 1408.3 991.4 416.9 29.6 Source: Linder, Forrest E., and Grove, Robert D., Vital Statistics Rates lg the United States 19003I940 TabIe 20, pp. 344-38I. 271 with the growth of the automobile industry. The death rate from accidents other than motor vehicle, one of the ten principal causes of death in 1910 as well as in 1940, decreased 43.2 percent during the 30 year period. Since this was an era of industrial expansion when high acci- dent rates could be expected, one might assume that the public has learned its lessons in safety well. Pneumonia, including influenza, was the second principal cause of death in 1910, having a rate of 109.7 per 100,000 population. By 1940 it had decreased 51.9 percent and ranked fifth as a principal cause. Bronchi- tis and tuberculosis were both among the ten leading causes of death in 1910. By 1940 rates for bronchitis were reduced from 26.3 to 3.2 per 100,000 pOpulation. Although rates from tuberculosis decreased 65.4 percent during the period under consideration, it is still among the ten principal causes of death. Thus, in Michigan those causes of death which increased during the 30 year interval were chronic diseases, with the possible excep- tion of cerebrospinal meningitis, whereas mortality rates from.communicable diseases showed substantial declines. To assist in evaluating the change of mortality classified by cause in Michigan corresponding data are 272 presented for the Nation1 in Table 9.20. Comparison of Michigan data with that of the Nation in Table 9.21 Shows that the State had lower death rates than the Nation for 20 of the 35 causes in 1910. Although both pOpulations show increases and decreases for the same causes of death, except for homicide, (Table 9.21 the last two columns), the number for which the death rates are higher in Michigan than in the Nation decreased from 15 in 1910 to 10 in 1940. The ratios of death rates which were 1. As pointed out earlier, the Nation is defined in this Chapter as the death-registration States of 1910, in order that more accurate comparisons of mortality change might be obtained. How do mortality rates of Michigan and the death-registration States of 1910 compare as to efficiency of classification and report- ing of deaths? This question may be answered partial- ly by comparing the death rates for senility and ill- defined causes of death in the two populations. (See Chapter I, pp. 29-34, for discussion of sources of error which influence original data on which cause- specific death rates are based.) Such a comparison is presented as follows: Death rates per 100,000 population 1910 1940 :Ill-Defined: :Ill-Defined Senility : Causes : Senility: Causes Michigan 51.3 32.6 6.4 4.1 D.R.S. 1910 25.5 22.0 4.0 3.6 Differ6n09325.8 10.6 2.4 .5 (Source: Linder and Grove, 22. cit., Table 18, pp. 312 and 320, and Table 20, p. 381. It would appear that on the basis of this test the D.R.S. of 1910 have more accurate mortality data than Michigan.for the years 1910 to 1940, although the differences of death rates for these two causes in 1940 are small. 273 Table 9.20 DEATH RATES PER 100,000 BY CAUSE, D.R.S.l OF 1910, AND PERCENT CHANGE, 1910 TO 1940 (By place of occurrence) (Death rates per 100,000 gonulation D.R.S. D.R.S. Diff“ X Diff. Cause of death 1910 1940 erence'”Dgath rate 1910 Diseases of the heart 158.9 343.9 185.0 / 116.4 Pneumonia and influenza 155.9 60.6 95.3 61.1 Tuberculosis (all forms) 153.8 42.5 111.3 72.4 Diarrhea, enteritis, etc. 115.4 5.7 109.7 95.1 Intracranial lesions of 95.8 92.9 2.9 3.0 vascular origin Nephritls 94.8 76.8 18.0 19.0 Accidents (non-vehicle) 82.7 47.7 55.0 42.5 Cancer (all forms) 76.2 142.4 66.2 f 86.9 Premature births 37.7 21.4 16.3 43.2 Bronchitis 23.9 3.1 20.8 87.0 Typhoid, paratyphoid fever 22.5 .5 22.0 97.8 Diphtheria 21.1 .4 20.7 98.1 Diabetes mellitus 15.5 52.6 17.5 / 115.1 Puerperal causes 15.3 4.9 10.4 7 68.0 Suicide 15.5 16.7 1.4 / 9.2 Congenital malformation 15.2 10.8 4.4 29.0 Syphilis ‘ 13.5 13.5 0.0 0.0 Cirrhosis of the liver 13.3 10.4 2.9 21.8 MGESIOB 12.4 .2 12.2 98.4 Hernia & intestinal obs. 12.1 9.5 2.6 21.5 Whooping cough 11.6 1.3 10.3 88.8 Scarlet fever 11.4 .5 10.9 95.6 Appendicitis 10.8 9.9 .9 8.3 Dysentery‘ 6.0 .5 5.5 91.7 AlOOhOliBm 5.5 2.3 3.27 58.2 Homicide 4.6 3.2 1.4 30.4 Ulcer of the stomach 4.0 7.5 5.5 f 87.5 Diseases of prostate 3.7 7.0 3.3 / 89.2 Biliary calculi 2.7 4.4 1.7 x 63.0 Accidents, vehicle 1.8 26.2 24.4 {1555.6 Goiter, exOphthalmic 1.2 3.3 2.1 f 175.0 Malaria 1.1 .1 1.0 90.9 Cerebrospinal meningitis .5 .5 .2 # 66.7 Senility 25.5 4.0 21.5 84.3 Ill-defined causes 22.0 3.6 18.4 83.6 All causes 354.9 24.2 1468.0 1113.1 1 D.R.S. of 1910 is abbreviation for death-registration States of 1910. Source: Linder, Forrest F., and Grove, Robert D., Vital Statistics Rateslgg‘ghg United States 1900-1940, TabIe 18, pp. 306’321. 274 higher in Michigan for the year 1910 are listed below with the corresponding ratios in 1940. Causes of death with higher death rates Gauge. in Michigan than in the Nation in 1910 - 1910 1940 Heart diseases 95.1 117.1 Intracranial lesions 91.5 103.8 Premature births 95.9 89.2 Bronchitis 90.9 96.9 Typhoid fever 95.7 250.0 Puerperal causes 91.6 89.1 Congenital malformation 95.6 77.7 Hernia 87.7 103.3 Appendicitis 85.0 101.0 Dysentery 93.8 166.7 Diseases of Prostate 92.5 122.8 Biliary calculi 93.1 118.9 Goiter 50.0~ 89.2 Senility ' 49.7 62.5 Ill-defined causes 67.5 87.8 Source: Table 9.21. Of the 15 death rates Which were higher in Michigan in 1910, eight were higher in the Nation for the year 1940. Two causes which had lower rates in Michigan in 1910 had higher rates than the Nation in 1940. They are measles and scarlet fever, mortality in the Nation being 66.7 and 62.5 percent as high respectively, for thee. causes as it was in IMichigan. The death rate for motor vehicle accidents was identical in the two pOpulations in 1910, namely, 1.8, but by 1940 it had increased to 32.8 in Michigan as compared with 26.2 for the Nation. In.the United States, death rates for 10 of the 35 causes increased during the 30 year interval, but in Michigan rates for 11 of them.increased. These causes are indicated 275 by pluses in Table 9.21. Of the causes that increased in both areas, the percentage changes were higher for eight of the ten in the Nation. 0f the 23 that decreased in both groups, fourteen had greater percentage declines in.Michi- gan. Thus of the causes that increased during the 30 year period, higher percentage increases occurred in the Nation, but of those Which decreased, percentage declines were greater in Michigan. Change 2; Infant Mertality Infant mortality rates for Michigan and the birth- registration States1 of the United States are given in Table 9.22, for the years 1915-1940 inclusively. During these 26 years Michigan has made remarkable improvements in saving the lives of infants. The rate in 1915 was 86.0, which is more than twice that of 1940, namely, 1. The birth-registration States of 1915 would havebeen a more apprOpriate group with which to compare infant and maternal mortality rates of Michigan over a number of years, because a majority of these nine States were also in the death-registration States of 1910. However, the data is not available. The "birth-registration States", Which are used as a comparable group here, comprise the expanding group of States entering the national birth- registration system. Michigan was among the birth-re- gistration States of 1915, which included also Connect- icut, District of Columbia, Maine, Massachusetts, Minn- esota, New HampShire, New York, Pennsylvania, and Ver- mont. By 1920 the birth-registration States had in- creased to 22 States and the District of Columbia, while in 1933 all of the States were included. To be admitted to the birth-registration States, a State must demon- strate that at least 90 percent of all births are pro- perly registered. For a discussion of completeness of registration of births and deaths, see Chapter I, pp. 29-34. 276 Table 9.21 RATIO or DEATH RATES AND PERCENT CHANCE FOR THE DEATH-REGIS- TRATION STATES OF 1910 AND MICHIGAN, BY PRINCIPAL CAUSE or DEATH, 1910 AND 1940 Ratio of Death EPercent Change Rates I 100 : 1910 to‘194g_ D.R.S.1916: D.R.S.19I6' M10hi'3D.R.S. Cause of Death Michigan.: Michigan : gan : of 1940 : 1910 : : 1910 Diseases of the heart 117.1 95.1 / 75.8 {116.4 Pneumonia and influenza 114.8 142.1 - 51.9 - 61.1 Tuberculosis (all forms) 126.5 158.2 - 65.4 - 72.4 Diarrhea, enteritis, etc. 123.9 112.0 - 95.5 - 95.1 Intracranial lesions of 103.8 91.5 - 14.5 - 3.0 vascular origin ’ Nephritis 155.4 138.8 - 17.0 - 19.0 Accidents (non-vehicle) 113.3 111.6 - 43.2 - 42.3 Cancer (all forms) 119.3 102.1 / 60.1 f 86.9 Premature births 89.2 95.9 - 38.9 - 43.2 Bronchitis 96.9 90.9 - 87.8 - 87.0 Typhoid fever 250.0 95.7 ‘ 99.1 - 97.8 Diphtheria 100.0 120.6 - 97.7 ' 98.1 Diabetes mellitus 122.1 105.4 / 80.4 {115.1 Puerperal causes (all) 89.1 91.6 - 67.1 - 68.0 SUlOidfi 120.1 112.5 % 2.2 / 9.2 Congenital malformation 77.7 95.6 ' 12.6 ‘ 29.0 Syphilis 117.4 107.1 - (8.7 0.0 Cirrhosis 113.0 109.0 - 24.6 - 21.8 Measles 66.7 137.8 - 96.7 - 98.4 Hernia & intestinal obs. 103.3 87.7 - 33.3 - 21.5 Whooping cough 118.2 106.4 - 89.9 - 88.8 Scarlet fever 62.5 106.5 - 92.5 - 95.6 Appendicitis 101.0 85.0 - 22.8 - 8.5 Dysentery 166.7 93.8 - 95.3 - 91.7 AlCOhOlism 191.7 152.8 - 66.7 ' 58.2 Homicide 105.7 184.0 % 20.0 - 50.4 Ulcer of the stomach 125.0 114.5 f 71.4 { 87.5 Diseases of the prostate 122.8 92.5 / 42.5 / 89.2 Biliary calculi 118.9 95.1 / 27.6 55.0 Accidents, vehicle 79.9 100.0 722.2 355.6 Goiter, exOphthaLmic 89.2 50.0 54.2 175.0 Malaria --- 157.1 -100.0 - 90.9 Cerebrospinal meningitis 125.0 150.0 100.0 # 66.7 Senility 62.5 49.7 - 87.5 - 84.3 Ill-defined causes 87.8 67.5 - 87.4 - 83.6 All causes 112.3 194.2 ' 29.6 - 24. Source: Tables 9.19 and 9.20. 277 Table 9 . 22 INFANT MORTALITY RATES BY RACE, BIRTH-REGISTRATION STATES AND MICHIGAN, 1915-1940. (By place of occurrence for 1915- 1938; by place of residence for 1939 and 1940) Number of deaths under 1 year per 1,000 live births Birth" Ngchigan Registration States Year Total White Colored Total White Colored 1940 40.7 40.0 56.7 47.0 43.2 73.8 1939 41.8 41.2 57.2 48.0 44.3 74.2 1938 44.6 43.9 61.5 51.0 47.1 79.1 1937 47.9 46.9 75.6 54.4 50.3 83.2 1936 50.7 49.9 71.0 57.1 52.9 87.6 1935 47.7 47.1 64.2 55.7 51.9 83.2 1934 52.0 51.3 71.7 60.1 54.5 94.4 1933 50.5 49.7 71.0 58.1 52.8 91.3 1932 54.0 53.4 67.6 57.6 53.3 86.2 1931 57.0 55.6 93.5 61.6 57.4. 93.1 1930 62.7 61.5 91.9 64.6 60.1 99.9 1929 66.4 64.9 109.0 67.6 63.2 102.2 1928 69.4 67.5 126.3 68.7 64.0 106.2 1927 67.7 66.5 101.7 64.6 60.6 100.1 1926 77.2 75.7 124.4 73.3 70.0 111.8 1925 75.3 73.3 148.9 71.7 68.3 110.8 1924 72.3 70.6 126.4 70.8 66.8 112.9 1923 80.3 78.7 147.2 77.1 73.5 117.4 1922 74.5 73.5 126.8 76.2 73.2 110.0 1921 78.6 77.7 125.3 75.6 72.5, 108.5 1920 91.7 90.3 179.5 85.8 82.1 131.7 1919 89.7 89.0 147.0 86.6 83.0 130.5 1918 89.1 88.4 134.7 100.9 _ 97.4 161.2 1917 88.3 87.9 157.8 93.8 90.5 150.7 1916 96.1 95.5 199.1 101.0 99.0 184.9 1915 86.0 85.6 159.9 99.9 98.6 181.2 Source: Linder, Forrest E., and Grove, Rebert D., Vital Statistics Rates in the United States 1900-I940, "Ta'b"1"e" '28“, p—‘S’. 57 ,Tp—. 589-590 . 278 40.7.1 Though the trend of the rate in general has been down- ward, the rate has not decreased consistently year by year. The highest rate was in 1916, while the next highest occurred in 1920. From.1916 through 1920 infant death rates were higher than the rate for 1915, but from.1921 through 1940 they were lower. Michigan had lower infant death rates than the Nation for 17 of the 26 years, including 1915-1918, the year 1922, and 1929-1940 inclusively. The highest rate for the two populations occurred in 1916, and the lowest for both was in 1940. The percentage decrease of infant death rates during the 26 year interval 1915-1940 and the average annual de- creases for Michigan and the Nation are given in Table 9.23. .According to these computations the rate in the Nation had a higher percentage decline and average annual decrease than occurred in Michigan. However, comparison of the rates for each year do not appear to bear out this conclusion. Data corresponding to those given in Table 9.23 were computed for the 25 years from.1916-1940, and the results Show Midhigan to have slightly higher percentage decreases as Shown immedi- ately'below: 1. This trend is generally true for the United States and for most other Western countries. See Pearl, Raymond, 22. cit., Table 25, p. 192, and Thompson, Warren S., o . 353., pp. 221-222. See, also, Kumlien, w. F., 92. .2.—E” p. 11. 279 Table 9.23 INFANT MORTALITY RATES (DEATHS UNDER ONE YEAR PER 1,000 LIVE BIRTHS), PERCENT DECREASE, AND AVERAGE ANNUAL DE- CREASE, MICHIGAN AND THE BIRTH-REGISTRATION STATES, 1915 TO 1940, BY RACE : (1): ; Percent : Average Race : B.R.S. : Infant Death Rates : Decline: : Annual :Muchigan: : : : : Decline : : : : :100 X Diff.: 1915- : : 1915 : 1940 : Diff. :1915 Rate : 1940 TCtal B.R.S. 99.9 47.0 52.9 53.0 2.03 Michigan 86.0 40.7 45.3 52.7 1.74 White B.R.S. 98.6 43.2 55.4 56.2 2.13 Michigan. 85.6 40.0 45.6 53.3 1.75 Colored B.R.S. 181.2 73.8 107.4 59.3 4.13 Michigan 159.9 56 .7 105.2 54.5 5.97 (1) B.R.S. is the abbreviation for Birth-registration States of the United States. Source: Infant death rates from Table 9.22. 280 Percent decrease of Average Annual Infant Death Rate 1916-1940 Decrease B.R.S. 53.6 2.16 Michigan 57.6 2.22 Infant death rates of the white population in Michigan exceeded the corresponding rates in the Nation for the years 1919 through 1932, except for the year 1931, while the rates of colored persons exceeded the rates for the Nation from 1917 to 1929, except for the year 1918. The infant death rates of the colored populations in both areas decreased more than those for their corresponding white Segments. Changes in Maternal Mortality "Until recently almost no success had been achieved in preventing death from complications of pregnancy and child- birfih if the recorded maternal mortality rate is accepted as the measure of progress."1 This statement, written by Dorn, pertains to the United States as a whole, but it describes Just as well the corresponding trend in.thhigan. From 1915 to 1929 the maternal death rate decreased only from.6.7 to 6.6, fluctuating slightly above or below these rates during the intervening years (Table 9.24).2 Although a slight down- ward trend is observable from 1929 to the present, the rate had decreased only 21.2 percent by 1936, contrasted with a 44.2 percentage decrease during the last four years, 1936- 1940. The rate for the Nation was lower than for Michigan 1. Born, Harold F., _92. Cite, P. 2. 2. Kbmlien, W. F., (22. cit., p. 10) describes a similar fluctuating pattern in South Dakota for these years. 281 Table 9.24 MATERNAL MORTALITY RATES BY RACE, BIRTH-REGISTRATION STATES AND MICHIGAN, 1915-1940. (By place of occurrence for 1915- 1938; by place of residence for 1939 and 1940) Number of deaths of mothers per 1,000 live births Birth Michigan Registration States Year Total White Colored Total White .Colored 1940 2.9 2.8 .QLQ 3.8 3.2 7.7 1939 3.1 3.1 3.6 4.0 3.5 7.6 1938 5.7 5.5 ‘82? 4.4 3.8 8.5 1937 3.6 3.5 7.2 4.9 4.4 8.6 1936 5.2 5.0 11.1 5.7 5.1 9.7 1935 5.3 5.1 9.8 5.8 5.3 9.5 1934 5.7 5.6 8.8 5.9 5.4 9.0 1933 6.1 5.9 10.8 6.2 5.6 9.7 1932 6.0 5.9 8.6 6.3 5.8 9.8 1931 6.0 5.9 9.8 6.6 6.0 11.1 1930 6.2 6.2 7.4 6.7 6.1 11.7 1929 6.6 6.4 10.3 7.0 6.3 12.0 1928 6.6 6.5 11.6 6.9 6.3 12.1 1927 6.8 6.6 12.3 6.5 5.9 11.3 1926 6.7 6.6 11.0 6.6 6.2 10.7 1925 6.4 6.2 12.7 6.5 6.0 11.6 1924 6.5 6.4 11.6 6.6 6.1 11.8 1923 7.0 6.9 12.2 6.7 6.3 10.9 1922 6.9 6.7 12.5 6.6 6.3 10.7 1921 6.9 6.7 14.5 6.8 6.4 10.8 1920 9.3 9.3 11.8 8.0 7.6 12.8 1919 7.7 7.6 13.5 7.4 7.0 12.4 1918 8.6 8.5 7.3 9.2 8.9 13.9 1917 7.4 7.4 10.3 6.6 6.3 11.8 1916 6.8 6.7 9.9 6.2 6.1 11.8 1915 6.7 6.7 .7 6.1 6.0 10.6 Source: Linder, Forrest E., and Grove, Robert D., Vital Statistics Rates in the United States 1900-1945, 282 in 1915, but by 1940 the positions had reversed, Michigan having a rate of 2.9 compared with 3.8 for the Nation. It may be concluded, therefore, that Midnigan has shown slightly greater improvements than the Nation in reducing the death rate from puerperal causes. Table 9.25 shows the maternal death rates and per- centage changes from 1915-1940, classified by race. The rate for the white pOpulation of Michigan decreased only slightly more than for the colored, but in the Nation the decrease for the colored group compared unfavorably with that of the white population.1 1. The colored population of Michigan is predominantly urban, while in the Nation it is largel rural. Accord- ing to Dorn, Harold F., (gp.‘g;§., p. 8 , it is probable that the maternal death rate is decreasing more rapidly in urban than in rural areas. The death rates of the urban colored in Midhigan are generally much lower than for the rural colored population. 283 Table 9.25 MATERNAL MORTALITY RATES (DEATHS OF MOTHERS PER 1,000 LIVE BIRTHs) , PERCENT DECREASE, AND AVERAGE ANNUAL DECREASE, MICHIGAN AND THE BIRTH-REGISTRATION STATES, 1915 TO 1940, BY RACE (By place of occurrence 1915-1938; by place of residence 1939-1940) 3 3 Maternal 3 3 : (1) : Death Rate : Percent : Average Race : B.R.S. : : Decline: : Annual :Michigan: : : : : Decline : : : : :100 X Diff.: 1915- : : 1915 : 1940 : Diff.:I9I5 Rate : 1940 TOtal B.R.S. 6.1 3.8 2.3 37.7 .088 Michigan 6.7 2.9 3.8 56.7 .146 White B.R.S. 6.0 5.2 2.8 46.7 .108 Michigan 6.7 2.8 3.9 58.2 .150 00101396. B.R.S . 10.6 7 .7 2.9 27.4 .112 Nuchigan 10.7 4.6 6.1 57.0 .235 (1) 3.3.8. is the abbreviation for Birth-registration States of the United States. Note: Rates based on less than 20 deaths are underscored. Source: maternal death rates from.Tab1e 9.24. CHAPTER.X SUMMARY AND CONCLUSIONS Objectives and Methods The general objectives of this study are as follows: first, to describe systematically the mortality character- istics of the population of Michigan for the year 1940 and to indicate the major changes which have taken place from 1910 to 1940; and, second, to determine whether certain generalizations about mortality are true in Michigan, to record the exceptions, and to make any new generalizations Which may be observed during the analysis. Death rates have been analyzed for several classes of the population, namely, residence, race, age, sex, and cause of death, and the results compared with corresponding data for the Nation. In addition, ten-year average rates (1935-1944) have been computed for each county to show the incidence of death in localities of the State. The measures of mortality employed throughout the study include the crude death rate, the age-adjusted death rate, the infant mortality rate, and the maternal mortality rate. The sources of data.have been the official reports and publications of the State and National Bureaus of Vital Statistics. Findings 1. The rural pepulation has a lower death rate than the 284 285 urban population. This proposition is generally true for the country as a Whole for both the crude and the age-ad- justed rates. In Michigan, however, it is true only when the rate is adjusted for age, The crude rate being higher for rural than for urban people. This is a reflection of the larger proportion of aged people in.the rural pOpular tion. Moreover, rural people tend to have higher death rates than the urban group for the younger ages (1-24 years) and for the older ages (75 years and over). Crude death rates for most of the principal causes of death are higher for rural than for urban residents, but when.adjusted for differing age distributions, they are lower in rural areas. 2. ,Among urban residents of'Mflchigan, those living in small cities (2500-10,000) have the highest crude death rate, while people living in large cities (100,000 and over) have the lowest rate. This condition apparently re- sults from.the tendency of the retired farmer to settle in nearby small cities. In the United States all of the various size cities have both crude and age-adjusted rates in excess of cor- responding rates in.Michigan. Both series of death rates vary inversely with the size of the city in the State, but in the Nation this is not true. 3. Crude death rates are higher for the colored than for the White population. The colored people in Michigan die at a rate 28.6 percent higher than that for white persons. When adjusted for age, the rate for non-whites 286 exceeds that for whites by 63.3 percent. This pattern of mortality, favorable to the white segment,persists when the data are classified by residence (including various size cities), by age and sex, and by cause of death. This is, also, generally true for the Nation. The rural colored people of Midnigan have crude and age-adjusted death rates more than twice as high as those of the rural white group. Moreover, greater mortality differences occur between whites and non-whites in rural areas than between the same groups in urban places. Rural death rates for the colored population are about twice the corresponding rates in urban areas. Mortality patterns for the United States vary in some of the details. Although both crude and.age-adjusted death rates for the colored segment are higher than for the finite, the differentials between the racial groups are not as great in the Nation as in.Michigan. Moreover, urban rates for nonswhites are higher than rural rates (whether crude or adjusted), the reverse relationShip being true in Michigan. 4. Males have higher death rates than females. This generalization is true for both.Michigan and the Nation for data classified by residence, for both racial segments, and for nearly all age groups. Exceptions to the proposition occur at ages 5-24 and 35-44 for colored persons. .A part of such deaths is usually attributed to lack of care given expectant mothers during pregnancy and childbirth. 287 5. Heart diseases are the chief cause of death in Mddhigan, in all counties of Michigan, and in the United States. Death rates from cancer are second highest in the State and the Nation, and in.most of the counties of Michi- gan. 0f the ten principal causes of death, two are communi- cable diseases and the remaining eight, non-communicable. In Michigan rates from.tuberculosis are higher in the urban than in the rural population. This rate is ex- aggerated by Detroit as well as the Upper Peninsula. Colored residents die of this disease six times as often as white people. Rates from syphilis are higher in the country than in the city. Furthermore, they are about six times as high among non-whites as Whites. Deaths from.this cause are more frequent among males. Suicide is far more prevalent among rural white males than any other class. It is twice as prevalent among whites as non-whites. In contrast to the white population, urban colored males are more prone to commit suicide than rural colored males. Homicide is found more often among non-whites than Whites. For’those diseases which afflict both of the sexes, males have higher death rates than females except for diabetes mellitus which proves fatal more often to women. 6. .A larger number of infants die annually in rural areas for each 1,000 live births than in urban areas. Among urban residents, those living in small cities have 288 the highest infant death rate and those in large-size cities the lowest. Michigan has lower rates than the Na- tion for all residence classes. Within the Nation's urban population, infant death rates vary inversely with the size of the city. The same relationship, however, is not true of Michigan. Proportionately more colored infants succumb before reaching their first birthday than white infants. In the State infant death rates are much higher in rural than in urban areas but the reverse is true for the Nation. ,7. In Michigan fewer mothers per 1,000 live births die from childbirth in rura1.than in urban areas. However, the differential between the urban and the rural rate in the Nation is small. In both Michigan and the United States the highest death rate occurs in small cities. Hewever, the lowest rate in the State is found among rural women; whereas in the Nation, women living in cities of 100,000 or more occupy the most advantageous position. . 8. A declining death rate characterizes all segments of the population of Michigan for the period 1910-1940. As is generally true, the greater decreases occurred among the younger ages and only slight declines took place among the older ages. Death rates of females declined more than those of males in all age groups. The colored population has made greater proportional gains than the white popular (tion. This is partially a function of the high rates which prevailed at the beginning of the period. Among the major 289 causes of death in 1910, only diseases of the heart, cancer, and diabetes mellitus show an increase for the period under consideration. In 1910 four communicable diseases were listed among the ten principal causes of death in Michigan, but by 1940 only two of these remained on the list, and their rates had dr0pped considerably. Pneumonia and influ- enza dropped from second position to fifth, and tuberculosis from.fifth position to seventh. All communicable disease rates in Michigan decreased for the period with the exception of cerebrospinal meningitis, which increased slightly. IMPLICATIONS Some of the implications of the findings of this study are discussed in more detail below: Michigan leads the Nation in the control of communi- cable diseases. This is no accident. The urgency of the need for such control was imperative. During the early years of Michigan history the State with its many swamps and swales was known as the "home of malaria." In 1910 four of the leading causes of death were from communicable diseases. Even as late as 1920 the death rate from pneumonia and influenza, and from bronchitis were substantially higher _in Michigan than the Nation. But hard work and well-directed effort, accompanied by a rising level of living, brought re- sults. By 1940 only pneumonia and tuberculosis remained among the ten leading causes of death. Moreover, by this time, the Nation's rate for both causes was over 40 percent 290 higher than.that in.Michigan. However, the State does not have a spotless record. Tuberculosis rates are excessive in Wayne county and in the western half of the Upper Peninsula. While it is recogniz- ed that the rates in the Upper Peninsula are probably as- sociated with the generally low level of living of that area, this excuse can hardly be given.for Wayne county. Perhaps the high rates found among colored people is the explanation. In any event, it is to be hoped that the in- tensive health education and the increased facilities for care now being directed into these areas will do much to equalize the situation throughout the State. Even as mortality studies of a few decades ago in- dicated a need for programs to eliminate the communicable and preventable illnesses, so mortality studies today are focusing attention on the need for research and health programs in the field of degenerative diseases. Five of the ten principal causes of death are degenerative in nature, and they account for 59.3 percent of all deaths in Michigan. They result in part from an aging population. That public health programs today are not directed against these diseases, which are taking the greatest toll of death, is common knowledge. .Aside from the humane principles in- volved, there are at least two basic reasons why we Should be concerned with these problems: 1. The increasing economic and social loss from.the toll of chronic illness and death among the work- ing population. 291 2. The cost of social adjustment for the older aged group, retired from productive work because of the onset of degenerative diseases. .A part of the poor showing that has been made in the control of degenerative diseases when contrasted with the great inroads that have been made upon certain infectious and contagious diseases in.Michigan can be attributed to the characteristic attitudes of both lay and professional people. Stieglitz, in his collection of works oaneriatric medicine, deplores the common attitudes of the medical fraternity, the public health officials, and lay leaders, who assume that the Chronic diseases are of such a nature that little can be done to prevent or stay their course. He observes that "biologists and physicians have been strangely content to take the phenomena of aging as a mat- ter of course."1 In his Chapter on ”Orientation,” he says, "The study of aging as a process has been conspicuously neglected until very recently... But a change in attitude will come and it must come soon, for there is true urgency in the need to know more about aging and the aged... The apparent boon of great longevity may become a curse, a terrible danger. Long life without health is not only an individual, personal tragedy, but a social evil seriously threatening national economy."2 With this quickening concern about the problems of 1. Stieglitz, Edward, ed., Geriatrig_Medicine, W.B. Saunders Co., Philadelphia, Pennsylvania, 1944, p. 3 ff. 292 chronic illness has come also the realization of our lack of knowledge about them. Mr. Stieglitz continues: "It is impossible in our present ignorance to say what aging is, what it does, or why or how it does it. Very little is known; much of what we think we know is merely superstition. Some is more purely wishful thinking. But the consciousness of ignorance is distinctly encouraging."l One might easily contend that in the past_our attitude toward the control of degenerative diseases was one of pessi- mism.and fatalism. The point of View which seemed to moti- vate these early efforts was that of helping those to live who had their lives before them. There is nothing pessi- mistic with this attitude unless it is emphasized to the point where people are not concerned with the possibility that death can be stayed. Modern physicians approach the hazards of degenerative disease with more hope. Further- more, these more optimistic Views are not limited to a con- sideration of one or two more tractable disease entities but seem to apply to many of them. The following excerpts by leading authorities will serve as illustrations: Walter S. Priest, Associate in medicine, Northwestern University Medical School, and Attending physician at the Cardiac Clinic of Evanston.Hospital has this to Say regarding cardiovascular-renal disease: "It is common knowledge that cardiovascular-renal diseases are first in the list of causes of death, exceeding by four times the second on the list, 1. Stieglitz, Edward, ibidg p. 5. 293 cancer .... When the death rate is adjusted for a changing age distribution of the population, the picture is not so black, however.... There is actually a decline in all groups, slight in the group seventy to seventy-four, but increas- ing with each lower five-year period untillforty- five to forty-nine, where, in 1935, the rate was 'slightly less than 400 per 100,000, as against 550 in 1911.... The fact of the improved rate is the important thing and Should stimulate re- search calculated to discover the reasons, with the hope that further decline may be brought about within the next twenty-five to fifty years." Regarding the nephropathies, John P. Peters, School of medicine, Yale University, and Associate Physician, New Haven Hospital states: "Whether the irreducible remnant of idiopathic senile nephrOpathies prove to be large or neg- ligibly small, the most immediately favorable approach to the problem of renal gerontology is a concerted effort to prevent or cure the dis- eases of earlier life that contribute to the total .... These ends can be achieved only if both public and physicians abandon their cav- alier attitudes toward the relatively trivial disorders in which these conditions originate. It will require also that medical practice be oriented toward prevention rather than_cure; that it be organized to follow persons in and out of disease, the apparently healthy as well as the evidently ailing."2 Concerning diabetes mellitus, which takes its larg- est death toll of females, Edward L. Bortz, Graduate School of Medicine, University of Pennsylvania, writes: "It is true that large numbers of individuals with diabetes succumb before they attain the later years of life. However, knowledge avail- able concerning the nature of diabetes and methods for treatment is of such practicable 1. Stieglitz, Edward, ibid, pp. 441-442. 2. Stieglitz, Edward, ibidg pp. 653-654. 294 value that for every diabetic who dies someone is probably to blame. The doctor*himse1f may be at fault in his lack of understanding of pro- per diet regulation, the use of exercise and the prescription of insulin. The modern treatment of diabetes today offers highly effective control measures to stay the révages that occur within the body in the uncontrolled diabetic condition. Or it may be the patient who is at fault in fail- ing to follow the physician's instruction, or, as too often happens, by sneer neglect and disinter- est in his own welfare. It was Mbntaigne who once.iaid 'people don't die, they kill themsel- ves. ,As for cancer, one needs only to pick up the morning paper to learn of the fight for prevention and control of this plague of mankind. Obviously, the millions-of dollars spent annually in research is testimony to our belief that this second most destructive disease can be conquered if we Will. And finally, with our increasing interest in degener- ative disease, has come the realization that we have inad- equate treatment facilities. Quoting from the report of the Kellogg Foundation entitled Hospital Resources and Needs, we find this statement: "For instance, the great decrease in deaths from communicable diseases in recent years has lessened the need for isolation hospitals. 0n the other hand, increases in deaths due to cancer, heart disease, and other diseases of mature and old people emphasize the geed for increased facili- N ties for such cases. It is interesting to assume that the health depart- ments of the Nation, both state and local, will at some time 1. Stieglitz, Edward, ibié5 p. 226. 2. Report of the Michigan.Hospita1 Survey,'gp. cit., p. 52. 295 in the future, be given the official function and full re- sponsibility of reducing our casualties from.degenerative diseases and from accident, as well as those from communi- cable diseases. There should be active health programs directed to the problems of each age group, giving due emphasis and importance to the adult population and es- pecially to the older age groups in proportion to their numbers in the population. This does not imply any slackening interest in the control of communicable dis- eases. It is rather the problem of keeping the gains al- ready made while reaching out to conquer other enemies. This would entail provisions for the control of all types of activity Which are known to be major threats to the lives of citizens, some of which are not considered as within the sphere of a health department program, for example, the pre- vention of accidents. In.the meantime mortality studies are among the chief indicators of these problems and their relationship to each other. Rural counties of Michigan have greater need for the expansion of health facilities than urban counties. The average toll of death levied on the rural population is 17 percent greater than in urban areas. This mortality con- dition occurs primarily because elderly people are rela- tively more numerous in rural communities throughout the State. For the same reason, largely, the population of rural cities (2,500-10,000) is decimated at a rate approxi- mately 25 percent higher than that of the total urban popu- 296 lation. As would be expected, the degenerative diseases take a far greater toll in rural areas than in urban. The four leading causes of death, heart diseases, cancer, intra- cranial lesions of vascular origin, and nephritis, take the lives of rural residents at a rate of 119 per 100,000 popu- lation in excess of the urban rate which is 521.4. But what percent of the really modern and well equiped clinics for heart disease would one expect to find in cities of 10,000 or under, though they might be serving a large trib- utary area? It is true that the rural pOpulation has a lower age- adJusted death rate than the urban, and that this rate is considered an acceptable measure of the environmental con- ditions conducive to good health. But it should not be used by society to rationalize away its responsibility to the people Who are actually living and dying in a given community. Because of the relatively greater number of rural residents under 15 and over 55 years of age, the de- mand for health facilities is relatively greater. Where there is much death there is ordinarily many days of ill- ness. Furthermore, there is often considerable imbalance between the nature and extent of facilities for care and treatment, and the nature and extent of illnesses. The highest incidence of death from chronic diseases occurs in the rural counties of the State where the special facilities required for diagnosis and treatment of these ailments are probably least available. 297 If mortality rates are taken as measures of health, colored people in Michigan are underprivileged. The age- adjusted death rate for this group is more than a third higher than for the white population. Communicable dis- ease death rates, considered among the better indicators of the health level of a community, are notoriously higher among non-whites than whites. Among the colored popula- tion, syphilis is the second highest cause of death in rural areas and tuberculosis is the second highest in urban areas. Among corresponding segments of white people, syphilis ranks thirteenth and tuberculosis, eighth. The situation of the Negroes in Michigan (who con- stitute 96.3 percent of the nonewhite group) is ecologically quite different from.that of their Southern neighbors. In Michigan they constitute an invading minority group. Not- withstanding equal rights principles expressed in free ac- cess of non-whites to the public institutions of the come munity, freedom of intercourse between the racial groups is restricted by the customs and mores of the people. Health progress depends not only on access to health facilities, but also on the building up of a set of attitudes and habits which bring the people to the facility. Because the educa- tional and economic levels of the colored pOpulation are ' generally inferior to that of the whites, and because of the conflict situations in which the colored person often finds himself, Negroes suffer from the kind of intellectual and emo- tional blocs Which keep people from forming constructive 298 health attitudes and habits. Inasmuch as such blocs, under present conditions, will probably be levelled more rapidly by enlightened members of their own kind, one might expect the urban colored of Nflchigan, where the Negro population is numerous, to respond more readily to health education and health opportunities, and hence to have higher levels of liv- ing and lower death rates than the corresponding group in rural areas. An examination of the age-adjusted death rates indicates this is likely to be true. Rural colored rates are 67.3 percent higher than urban rates, and communicable dis- ease rates are also higher in rural areas. Apparently, non-whites make use of health instruction and health facilities When opportunity and incentivads given them.although the quality of their performance may not be quite as high as that of the White population. This is best illus- trated by examining death rates from syphilis. Programs for the control of veneral disease in.Michigan are more comprehen- sive in urban than in rural areas. If colored people were to make use of the opportunities which they have, one would expect to find syphilis rates for the urban colored population con- siderably 10wer than those for the corresponding segment of the rural population. These rates are 56.4 and 504.7 respec- tively. One task facing those interested in.the promotion of public health among colored people is the extension of health education and health facilities to rural areas. Females have a definite survival advantage over males. This fact itself is attracting considerable attention. Many 299 attribute this differential to the biological superiority of the female (especially for those under five years of age). Others quite frankly say they have no explanation. The open- ing sentence in a popular article entitled "How to Keep Your Husband Healthy?"1 begins: "No one has ever discovered Why men refuse to wear galoshes, resist long underwear, and are repelled by practically any edible that isn't steak and pota- toes." The same authors quote Dr. Edgar V. Allen at the mayo Clinic in a description of the American.husband as follows: ”His blood pressure is high, his arteries hardening, his temp per short. He is irritable, nervous and melancholic. His brain is weary, his muscles tired, his stomach acid." The authors then summarize as follows: "It is a portrait Which your husband will hardly recognize, much less acknowledge. It is largely this capacity for self-delusion on the part of the married American male that makes him.the least healthy member of the family unit today." It is the contention of the author of this disserta- tion that females live longer chiefly because of the differ- ing behavior roles of the sexes, and that the individual male is not as deluded as the above authors would have us think. Consciously or unconsciously, he is forced into the position he takes by male roles already defined by the group to which have been assigned status values. The American male is pay- ing dearly for the privilege of occupying his particular star 1. Robbins, Jhan and June, "How to Keep Your Husband Healthy?" Magazine Section, The Sunday Star, Washington,D.C., Sep- tember 26, 1948, p. 4. 300 tus in our society. It is probable that many of the undesir- able characteristics of the male role from a health standpoint could be eliminated,--for example, the idea that it is "sissy" to go to a doctor. These concepts are not necessarily essen- tial to the male role as now structured. Attitudes and values inherent in.this role Which are health hazards Should be sought out through scientific research and their implications taught as part of the health training of youth. Infant and maternal death rates are lower in Michigan than in the Nation. Within.Michigan, more infants die in the country but more mothers during childbirth and pregnancy die in the city. This appears to be associated with the level of living. The infant death rate, considered by many as the most sensitive index of health care, is_beginning to mirror the superior advantages offered urban mothers in child care and training. These programs should be extended to rural areas. The maternal death rate, on the other hand, reflects the strain of urbanization. Urban.mothers undergo childbirth at later ages than rural women, and tend to limit the size of their families, sometimes at great personal risk to theme selves. The tempo of the city, its system.of values, and its way Of life do not seem to be as conducive to the physical and mental well-being of prospective mothers as are the rural patterns. RECOMMENDATIONS On the basis of the findings of this study and the 301 analyses that have been made, the following recommendations are presented in the hope that they will assist the State in improving its health programs and in extending its present position of leadership. 1. That a systematic analysis of the State's mortality characteristics be made every ten years. 2. Inclusion of either five or ten-year average annual age-adjusted death rates for the counties of Michigan by residence and leading causes of death in the State Health Department Reports, or in special publications. 3. The initiation of local studies of health facilities and attitudes of the public toward health needs, preferably by local institutions through State aid. 4. The development of research on the unhealthy aspects of role behavior, especially occupational and sex roles, and the incorporation of findings in public health education prOgrams. 5. The expansion of preventative and research programs especially designed to increase our knowledge and control of the degenerative diseases. 6. More intensive health education programs for colored people, especially for those in rural areas. 7. A survey of all health programs and activities in the State as a first step in organizing a complete_preventative program for chronic and other diseases which may be under- emphasized at present. 8. The expansion of both private and public programs to 302 include diagnosis and some treatment of non-infectious dis- eases, especially in rural areas. BIBLIOGRAPHY 1. 3. 4e 5. '7. 8. 9. 10. 11. 12. 13. BIBLIOGRAPHY Anderson, Gaylord W., ”The Political Impact of Modern Science on Public Health,“ _'I_'_h_9_ Annals 2; the American Academy of Political 313 Social Science, January, I917, pp. 26'fi4e Annual Re orts of the State De artment of Health, Num- """""bers ath mag—an D‘spartm-fi—nt eel-EE,T_Ians ng, Michigan, ices-1945. Beegle, J. Allan, Michi an Po ulation, Special Bulletin 342, Michigan Agricu tur figeriment Station, East Lansing, Michigan, 1947. Bixler, Genevieve Knight, "Nursing Resources and Needs in Michigan," Michigan Council on Community Nursing, Lansing, Michigan, 1946. Brown, J. F., Psychology and the Social Order, McGraw- Burke, Bertha 8., "Nutrition and Its Relationship to the Complications of Pregnancy and the Survival of the Infant,“ American Journal of Public Health, Vol. 35, No. 4, AW , 945*, pp."'331=3_—39. _— Byrd, Oliver 15., Health Instruction Yearbook, 1945, Stanford University Press, Stanford University, Cali- fornia, 1945. Gavan, Ruth Shonle, Suicide, University of Chicago Press, Chicago, Illinois, 1953. Commission on Hospital Care, "State Hospital Study Manual," Book IV, (preliminary draft), Chicago, Illinois. Coombs, (L. 0., "Economic Differentials in Causes of Death," Medical Care, Vol. I, July, 1941, pp. 246-255. Culver, Elizabeth M., ”Women in Service ," The Annals g; the American Academy of Political and Social—Science, September, I943, pp. '53-53. Densen, Paul M., "Family Studies in the Eastern Health District: II The Accuracy of Statements of Age on Census Records," The American Journal 3; H iene, XXIII, No. 1, Soc. A, 1925',- pp. I-SE. Dorn, Harold F., "Maternal Mortality in Rural and Urban 28, 1959—“, pp_. am. 303 1‘ 14. 15. 16. 17. ' 18. 19. 20s 21. 22. 23s 24. 25. 26. 27. 304 Dermal-wold F.,. goitality Rates and Economic Status :1 Areas, b is Health Re orts Vol. 55 No. 1 January 5, 1940, pp. 3-12, (ReprInt N i 2126). ’ ' Dublin, Louis 1., "Recent Changesin Negro Mentality,a Megapclitan Life Insurance Company, New York, New York, 9 . Dublin, Louis I., and Bunzel, Bessie, To Be Or Not 193 _B_e_, H. Smith and R. Haas, New York, New York,_I9'3_3. Dublin, Louis I., and Lotka, Alfred J., Len th of Life, The Ronald Press Company, New York, New or , 1936. Duncan, Otis Durant, Social Research _qg Health, Social Science Research CouncII', New York, New York, 1946. Freedman, Ronald, “Health Differentials For Rural--Urban Migration,” American Sociological Review, Volume 12, No. 5, 1947, pp. 536-54I. Gist, Noel P., and Halbert, L. A., Urban Socie , Thomas Y. Crowell Company, New York, New York, 1 . Gover, Mary, and Sydenstricker, Edgar, "Mortality among Negroes in the United States," Public Health Bulletin 174, Government Printing Office, Wash'Ington, D.C., June, I927. Graunt, John, Natural and Political Observations Made Upon the Bills of Mortflity, (Edited with—Introduction by Water I. fillcox , The John Hopkins Press, Baltimore, Maryland, 1939 . Hagood, Margaret Jarman, "Development of a 1940 Rural-Farm Level of Living Index for Counties," Rural Sociol , Vol. 8, No. 2, June, 1943, pp. 171-180. Hagood, Margaret Jarman, uRural Level of Living Indexes, 1.940," m Sociolo , September, 1943, pp. 292-293. Hagood, Margaret Jarman, Rural Level _o_i; Living Indexes for Counties 93 the United States, I940, Bureau Agricultural Economics, United States Department of Agriculture, Washington, D.C., 1943. Hagood, Margaret Jarman, Statistics for Sociolo ists, Reynal and Hitchcock, Inc., New York, New York, I9ZI. Hitt, Homer L., and Bertrand, Alvin L., Social As cts of Hos ital Planni , Louisiana Study Series No. I, HEaIth and HospItEIn Division, Office of the Governor, Baton Rouge, Louisiana, 1947. 28. 29. 30s 31. 32. 33. 37. 38. 39. 40s 41. 305 Hoffer, Charles R., "Health and Health Services for Michigan Farm Families," Department of Sociology and AnthrOpology, Michigan State College, East Lansing, Michigan, 1947. (Mimeographed) Hoffer, Charles R., "Medical Needs of the Rural Popu- lation in Michigan," Rural Sociolo , Vol. 12, No. 2, Kemp, Louise, and Smith, T. Lynn, Health and Mortality _i_g Louisiana, Agricultural Experiment Staiion, ou s ana State University, Bulletin 390, 1945. Kleinschmidt, Earl R., "Major Problems in Sanitation and Hygiene in Michigan 1850-1900," Michi an Histog M_a_ga- zine, Vol. 28, No. 3, 1944, pp.-.42 - . Kleinschmidt, Earl R., "Prevailing Diseases and Hygienic Conditions in Early Michigan," Michigan Histogy Maine, v01. 25, NOe 1' 1941, PP. 57-9 0 . Kumlien, W. F., Basic Trend _o_f_ Social Chan e _i_g South Dakota; Public H__ealth '—_I£Faci ities_—_i, Agr c_1—_B-u tural smut Station, South Dakota State CoIIege, Bulletin 334, Brook- ings, South'Dakota, 1940. Landis, Paul H., Population Problems, American Book Company, New York, New York, I949. Linder, Forrest E., and Grove, Robert D., Vital Statis- tic; Rates _13 the United States 1900-1940, Bureau of ensue, Washington, D. 9., I915. Loomis, Charles P., “Measuring Rural Medical Needs," First Annual Mich an Rural Health Conference, Michigan Staie CoIBge, ast ansing,"MiEFigan, I917, p. 20. Mangus, A.R., Health ing Human Resources _i_g Rural Ohio, Mimeograph BulIetin No. 176, Department of RuraI Econom- ics and Rural Sociology, Ohio State University and Ohio Agricultural Experiment Station, Columbus, Ohio, 1944. MetroPolitan Life Insurance Company, "Why Do People Kill Themselves?" Statistical Bulletin, 26 , No. 2, February, 1945. Report of the Michigan Hospital Survey, Hos ital Resources and Needs, W.K. Kellogg Foundation, BattIe Creek, Micfii- gan, 19460 Molyneaux, J. Lambert, "Differential Mortality in Texas," American Sociological Review, Vol. X, 1945, pp. 17-25. Holyneaux, J. Lambert, Gilliam, Sara K., and Florant, L. 0., "Differences in Virginia Death Rates by Color, Sex, Age, and Rural or Urban Residence," American Sociolo ical Review, Vol. 12, No. 5, October, 1947, pp. 525-535. ("I r1 fa 42.. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 306 Mott, Frederick D. (M.D.), and Roemer, Milton I. (M.D.), Rural Health and Medical Care McGraw-Hill Book Company New York, New York, I949. , ' Newsholme, Arthur, The Elements of Vital Statistic}, D. Appleton and Co., N_ew Ymork, New vars—r. 924. O'Brien, Robert W., Readi s'ip General Sociology, Pacific Books Publi ers, alo AIto, Ca forn a, 1947. Oosterhof, Willis M., fiMedical Cars Under Public Assis- tance in Michigan," Michigan Welfare Review, Vol. 3, No. 4, 1946, pp. 1-4. Oosterhof,'Willis Ms, "Medical Care Under Public Assis- tance in.Michigan,” Michigan Welfare Review, Vol. 4, NOe 4. 1947, pp. 1'21e Palmer, Carroll E., Tuberculosis ip the United States Medical Research Committee, National Tubercqusis .Association, New York, New York, Vol. 4, 1946. Pearl, Raymond, Introduction ngMedical Biometr and Statistics, W.B. Saunders 90mpany, PEiIa e p a, PEnn., 94 . Registrar General, The Registrar General's Decennial Supplement; England and Wales, H.M. Stationery Office, Robbins Jhan and June, “How to Keep Your Husband Healthyi" Magazine Section, The Sunda Star, Washington, D.C., September 26, 1948, pp. 1-5. Roosevelt, Franklin D., The Governor's Foreword, from Public Health in New York State, Department of Health, KI—bany,—-'7New o'EE,T—932"—. '— Smiley, Dean Franklin, and Gould, Adrian Gordon, Com- munity H iene, The MacMillan Company, New York, New erk, 1946. Smith T. Lynn The Sociolo of Rural Life, Harper and Brothers, New Yo—rk, T—‘flew York, REF—Tia. s'dfii'on, 1947. Stieglitz, Edward, ed., Geriatric Medicine, W.B. Saunders 00., Philadelphia,'PSnnsyIvania, 1944. Sydenstricker, Edgar, Health and Environment, McGraw-Hill Book Co., Inc., New York, New York, I933. Tate Leland B. The Health and Medical-Care.Situatian in Rural Vir inia, BuIletin 363, Virginia AgricuIiuraI 'kkperimeni 9tation, Blacksburg, Virginia, 1944. 307 57. Thompson, Warren 5., Population Problems, McGraw-Hill Book Company, Inc., NSW'YOIK, New York, 1942. 58. Thompson, Warren S., and.Whelpton, P.K., Po ulation Trends in the United States, McGraw-Hill Book Company, Inc., New York, New York, I933. 59. Vance, Rupert B., and Danilevsky, Nadia, All These People, The University of North Carolina Press, Ekapel l , North Carolina, 1945. 60. Vit%l Statistics 2; the United States, 1940, Part II, Res dance, United States Government Printing Office, Washington, D. G., 1943. 61. Bureau of the Census, "Age-Adjusted Death Rates in the United States, 1900-1940," Vital Statisticpr-Special Reports, Vol. XXIII, No. 1, I94 . 62. “Deaths and Death Rates for Selected Causes United States, Each Division and State, 1945," Vital Statis- tics--Special Reports, 1947, Volume 27, Number 3. 63. Walters, J.E., ”women in Industry," The Annals'gg_the American Academy of Political and Soaikl Science, September, 943, pp. 56-62. 64. Whelpton, P. K., Needed Population Research, The Science Press Company, Lancaster, Penn., I938. 65. Wiehl, Dorothy G., "Some Recent Changes in the Mortality Among Adults,” Journal pgugpeventive Medicine, Vol. IV, NOe 3’ May, 19 O 66. Willcox, Walter F., Introduction pp‘Vital Statistics ' of the United States I999 90 I930, Bureau e fin‘mm'. ,_Washing——_to n, D""". c . ,‘Isflas. 67. Willcox, Walter F., Studies ip American Demoggaphy, Cornell University Press, Ithaca, New York, 9 . 68. ‘Yerushalmy, J., "The Age-Sex Composition of the Popu- lation Resulting From Natality and Mortality Conditions,” The Milbank Memorial Fund Quarterly,‘Vol. XXI, No. I, 'Jkkuary, I934, (Reprint). APPENDICES 308 APPENDIX I MORTALITY RATES: DEFINITIONS, COMPUTATIONS , AND EVALUATI one The interpretation of mortality rates, especially their implications in the local community, is usually sociological in nature. However, there are rathcr'precise definitions of various vital statistic rates which have been carefully formulated by statisticians, especially biostatisticiane, which inculd serve as guides in the interpretation of vital data. In the following pages, a brief discussion is presented of the basic mortality rates employed in this study. ‘Various authorities are relied upon for concepts, formulas, and evaluations. The rates described are the crude, specific, and adjusted (standardized) death rates. The material is organised in outline form.as follows: 1. Definition 2. Nomenclature 3. Data required for computation 5. Advantages and disadvantages 309 CRUDE RESIDENT DEATH RATES Defines: "The crude rate for an area which most nearly approx- imates the total probability of death is a rate computed by dividing the deaths occurring to that pOpulation regardless of where those deaths occur by the resident population of that area. This rate is called the 'resident death rate' and is based on deaths tabulated according to place of re- sidenceo'l “The crude resident death rate is a direct measure of the effects of mortality on the population which actually is living’ in an area, and therefore has value in a histori- cal sense of recording the number of people per 1,000 mid- year pepulation who actually died."8 , Igcrmation Reguired tp Compute cm. s Regident Death Rates: 1. The number of deaths of residents in an area dur- ing a given period of time. 2. The total resident population of a given area and period of time. Agvantgges of the Crude Resident Death Rate: 1. It is the most common, simple, and perhaps the most valuable, death rate. 2. It is easy and simple to compute. 1 Linder, Forrest E. and Grove, Robert D., pp. cit., p. 38. 3 Hagood, Margaret Jarman, Statistics for Sociolo ists, Reynal and Hitchcock, Inc., New York, N.Y., I9ZI, p. 823. 3. 4. 5. 310 Information required for its computation is usually readily available. Crude death.rates may be interpreted both as a tell at deaths in an area and as a probability number, if the exact number exposed to the risk of death is used as the denominator. Crude death rates are valid numbers, "...and comparisons can Justifiably be made in spite of any differences in the population composition of compared groups of people, or in the conditions under which they are exposed to death.'1 Dipadvantggep pf. thg Crude Deatpl M: 1. 2. 3. It is difficult to obtain the exact number of persons exposed to»the risk of death in an area for a given period of time because of the con- stantly changing population base.8 There is some uncertainty that the numerator (number of deaths) represents the actual deaths occurring to the population group used in the denominator. Death rates computed from the place of occurrence of death do not have a probability meaning showing the chances of death. The resident death rate does‘hare the possibility of showing the probabil- ity of death in an area. I ”Mel-I9 Forrest E. and Grove, Robert D.,‘pp. cit., p. 40. 2 Linder, Forrest r. and Grove, Robert D., ibid, p. 37. 311 4. Crude death rates are questionable when consid- ered as measures of the force of mortality. According to Pearl, ”So many variables besides those essentially lethal can (and do) influence the stated values of crude death-rates as to make them.rather untrustworthy for why but the broadest and roughest conclusions and estimates."1 This criticism is based primarily on differences in the age characteristics of the pepulation, thougi in certain areas sex and race may also influence the crude rate. ..5. Crude death rates must be used with utmost caution and reservation in comparisons of one locality or one time with another-"unless it be shown that the age and sex constitutions are similar in the areas compared. SPECIFIC DEATH RATES agined: Specific death rates may be defined as the number of deaths in a specific class of the pOpulation divided by the population in that specific class multiplied by 1,000. goal-.1. Raymond, _220 eitO, p. 177. 3MB Nomenclature : Rates may be made specific for many characteristics of the pepulation, (i.e., specific rates), It is customary and preferable to add the name of the pepulation character- istic for identification and clarity. Examples: 1. Age-specific death rates 2. Age-sex-speoific rates 3. Sex-specific rates 4. Race-specific rates 5. Age-SexPRace-Specific Rates Advantages: 1. 3. 4. Specific death rates by age are considered by some authorities as the best measures of the real force of mortality.1 Specific rates represent the chance or probabil- ity of death more nearly than crude rates, since they are more refined. Age-specific death.rates are the basis for adjust- ing rates by the direct method. Age-specific death rates serve as the standard for standardizing rates by the indirect method. Disadvantages: 1. ‘Mortality data is not tabulated for a large numv ber of socio-psychologioal traits. Data is available for age, sex, residence (rural, urban), and race. Information on residence and race is. frequently not tabulated and presented for public ISearl, Raymond, ibidg p. 179. 313 use. 2. Age-specific rates require the population by age groups, ihich sometimes is not available for intercensal years, especially for small areas. DEAIH RATIO Qafined: The death ratio.may be defined as an index showing the numerical relation between one cause of death (or group of related causes) and total deaths, or between deaths at a given age with deaths for all ages. Naaanclature: l. “PrOportionate mortality." (Linder, Forrest E., and Grove, Robert D.) H 2. “Death Ratio." (Pearl, Raymond) Since Professor Pearl has been theprincipal advo- cate of this index, his terminology will be used, namely, “death ratio." Ingarmatian Necessary for Computation: l. The number of deaths in a specified class for a given time. 2. The total number of deaths for all classes during a given time. Adzantagaa: 1. an a cause basis, a.death ratio "measures the pro- bability that in a given total number of deaths from.all causes a particular one will be from.one 2. 3. 4. 314 particular cause, say tuberculosis of the lungs.'1 Likewise, this index measures the pro- bability that in a given total number of deaths in all age groups a particular death will be in one particular age group, say 65-69. The death ratio has validity and usefulness when employed in studies concerned with the kind of probability described in Advantage number 1. This index has been effectively employed in re- search on diseases and the construction of life tables.8 According to Pearl, "... death ratios are caning more and more into use as an effective statis- tical technique to bring out important changes in the mortality pattern of a pOpulation (see Welling, for example)."3 ("The Changing Aspect of Age at Death." gaxaa. §aa_1_t_a EEEH Vol. 52, pp. curses, Nov. 1938.) final, Raymond, ibide, p. 202. 3 See: a. Greenwood, M., and 'I‘ebb, A.E., ”An Inquiry into the Prevalence and Etiology of Tuberculosis Among Industrial Workers, 'ith Special Reference to Female Munition Workers," Med. Res. Comm., Spec. Rept. Ser. No. 22, London, 1919. 1'1 721—33?“ b. Fisher A. An Elementagr Treatise on Fregueu Curves 31.1.51. Their A, lication n 1;h Ana sis _o__an Death res and Iif___e_ TabEe s. Tran-51am efrom tED iusson, New —_York (Macmillan), 1922, pp. xv fo 240). WHY EeAe 5 Pearl, Raymond, an. cit., p. 205. 315 5. The pepulation of the area for a given time is not required for computation of death ratios. 6. In cases where the population is not available to compute other death rates, the death ratio is sometiles substituted for death rates.1 7. Death ratios “may also have a certain rhetorical value in discussions listing or ranking in order of importance the principal causes of death." 1 8. Death ratios ”can be used to compare the relative importance of different causes for the same area.'1 9. "Whenever the pepulation base is the same for a group of specific rates, the corresponding death ratios will be in direct proportion to the rates.'1 Was: 1. Death ratios, like most other mortality rates, are percentage figures, and thus they are subject to the some limitations and possible misinterpretae tions as percentages. 2. Since both the numerator and denominator of the death ratio are deaths, and thus variable quanp tities, they are affected by the same biological forces. For this reason, death ratios cannot be highly precise measures, especially, since one ‘iilinder, Forrest E., and Grove, Robert D.,.aa. c;t., p. 51. 316 can never be sure as to what portion of the death ratio is due to the numerator or the denominator. Pearl claims that the some criticism applies to any death rate, for the pepulation figure in the denominator, or any other more precisely defined part of population, is not an invariable quan- tity.1 3. When substituted for death rates, death ratios have two defects: (a) changes in the death rate for a cause will not-be reflected by the ratio if the death rates for other causes are changing proportionally, (b) changes in a death ratio may indicate either (1) an increase or decrease in mortality for that'eause, or (8) an opposite change in the mortality for the (other causes.8 These are the possible misinterpretations of the death ratios. Obviously, these possible misin- terpretations are not likely to be made when the data is studied carefully and additional analyses made in cases of doubt. 4. Death ratios, like certain other mortality indices, are not in general use. Evayiation: The death ratio may be of value in a study where the objective is to present the relation of principal causes, I Pearl, Raymond, aa. cit., p. 802. 3 Linder, Forrest E. and Grove, Robert D., an. cit., p. 50. 317 or their ranks, within a particular area in an effort to analyse and point out the health problems found there. They cannot be compared with ratios of another area. ADJUSTED own more (on STANDARDIZED)1 Dafinaa: In general, adjustment of death rates may be defined as the modification of crude death rates of areas to show what they would have been had the pepulation in the areas been constituted as to age, sex, or age and sex, (or any other characteristic of the population) like the population chosen as a standard. Nomenclature: Names of the various methods of adjusting death rates will be presented when the methods are discussed. Adjusted death rates may differ in two ways: (1) in.method of adjustment employed. (2) in pepulation chosen as the standard. Objectives of Adjusted Death Rates: 1. sAge-adjusted death.rates are attempts to "express the mortality of a pepulation of changing or ab- normal age distribution by a single figure calculat- 1 The Committee on forms and Methods of Statistical Practice of the American Public Health Association recommends the use of the term.'adjusted' in preference to ”standardized" as the latter term.has been used to cover other widely different definitions by statisticians. See American Publia Hea;th Association Yearbook, 1937-1958, p. Ida. 2. 318 ed in such a way that the changes or abnormalities in constitution do not appreciably affect it.'1 “...the problem.is to suggest as substitutes for the crude death rate the simplest practicable rates, such as might be used by a health officer who is asked at a meeting how the city death rate compares with the rural, or whether it ks true that mortality in the state is rising." Spandard Population and Use of Standard Million 1. 2. 3. 4. 5. 6. 11 standard population is required for adjusting the death rate by the direct method.- The population chosen as a standard must be sel- ected with the type of comparisons and purposes of the research being conducted clearly in mind. The standard population used should be fairly typical but different from.the pOpulations in the communities whose mortalities are to be compared. Only those adjusted death rates standardized on the same population are comparable. Actual pOpulations at a specific time are used as standards by expressing them.on a per-million or per-thousand basis. Examples of standard million pepulationszs (1) England and Wales 1901 . (2) ‘A population proposed by the International IfThe Registrar-General's Statistical Review‘ag,§agland aha macs 3933a; yaar 1957. New Annual Series, No. , Tex . .M. Stationery Office, London, 1940, p. 10. . 3 Report of the Committee on Forum and Methods of Statistical Practice of the American Public Health.Association, American gablic Health Association Year Book, 1959-1940, p. 12 . 3 Linder, Forrest E. and Grove, Robert D., ap. cit., p. 80. 319 Statistical Institute 1917 (3) United States 1930 and 1940 (4) United States life table (whites and negroes) I 19 29-31 . Standard Specific Death Rategz 1. Standard specific rates are required for adjust- ing the death rates by the indirect method. 2. There are specific mortalities that can logically be set up as yardsticks, against which the rates of the community can be measured. 3. National specific rates are frequently taken as a standard for comparisons of subdivisions, or ideal specific rates might be used.1 Age-sexeAdjusted Death Rates: Some authorities believe that for the majority of pur- poses for which death rates may be adjusted, age variation is the only factor important enough to cause the crude rate to be significantly biased. When adjustment for sex differ- ences as well as for age differences is required, the rates ‘must be obtained specific for sex as well as for age, and these must be multiplied by the standard pepulation of the given age and sex. In most cases where it is desirable to adjust for age differences, it is theoretically desirable also to adjust for sex differences, but the added refinement usually makes little difference in the result. For this rea- son, in areas where the sex-ratio is fairly well balanced, 1:Dublin, Louis 1., and Lotka, Alfred J., 22. cit., p. 191. 320 it is of little advantage to adjust for both age and sex. For purposes of showing the computations necessary for age-sex-ad— justed death rates, an example will be presented using the standard white population of the United States, 1930, and amp playing the direct method. (See Table III) Male plus female expected Age-sex-adjusted death a deaths Michigan 1940 X 1,000 rate, direct method cpu at on of Standard Community (ihites U.S.,1930) Males plus females e 1,000,000 It should be pointed out that this formula does give an ad- justed rate independent of varying sex proportions and vary- ing age distributions, but does not permit comparison of male and female rates, specific rates being referred to different standards (the male to male and female to female).1 IMEEHODS OF ADJUSTING DEETHIRLTES There are three methods for adjusting mortality rates in common use. Each method has its advantages and disadvane tages, and each may be preferred at times, depending upon the purposes of the particular study. These methods are known by various names, but the following terms are preferred: Adjusted death.rates, indirect method .Adjusted death.rates, direct method Equivalent average rates method. The life table death rate (or the reciprocal of the 'expec- tation of life at birth“) is also in common.use; however, this rate, requiring the construction of life tables, is so 1 Linder, Forrest E., and Grove, Robert D., pp, cit., p. 68. 321 laborious to compute that it will not be illustrated. The three methods listed above will be discussed and an example of each presented in detail. ADJUSTED DEATH RATES, INDIRECT METHOD Defined: . An adjusted death rate, indirect method, "... is an gbstract or theoretic figure derived by applying the spe- cific death-rates of the general pOpulation, or of some standard imaginary population, to the actually existing age and sex distribution of the living pepulation of a particular locality to determine what would be the number of deaths in that locality if the specific death-rates of the standard pOpulation prevailed there, and then dividing the number of deaths so obtained by the actual total living population of the locality. The figure so obtained is then employed... in the computation of an adjustment factor ihich is applied to the observed crude death-rate of the locality to give finally the adjusted death-rate (indirect method) of the locality.'1 Nomenclature:. 1. Age-adjusted death.rate, indirect method: ‘Used by Linder, Forrest E., and Grove, Robert D., and recommended by the Committee on terms and fPearl, Raymond, _Qne Gite, pp. 270‘271e 2. 3. 4. Methods of Statistical Practice of the American Public Health Association.1 (4, p. 165) .Lge-adjusted death-rates (1), Raymond Pearl.2 Indirect method of standardization, Sir Arthur N’ewsholme.3 Standardized death rates, Indirect method, Hagood.4 ngormation.Reguireg: l. 2. gvgt ages : l. 2. The specific-age group death rates of the popula- tion being used as the standard. (Standard spe- cific rates) The population for which death rates are to be adjusted by age groups. The indirect method is used currently in the.1nnual Volumes of The Registrar-General's Statistical Be- view for.nngland and Wales, which gives the method some prestige and authority. The method is equally as reliable as other methods (Direct).5 I Committee on Forms and Methods of Statistical Practice of the American.Public Health Association, American Public Health Assogiatign Yearbook, 1937-1938, p. 135. 3 pg. cit., p. 270. 3 Newsholme Sir Arthur The Elements of Vital Statistics in Their Bearing on Social and fiinc 33am Problems, 5. Appleton and Company, New Yerk, N.Y., 1924, p. 222. 4'IHagood, margaret Jarman,‘gp. cit., p. 822. 5 Pearl, Raymond,‘gp. cit., p. 270. 3. 4. 5. 6. Standard specific death rates of the United States are easily available. The indirect method does not require the nume ber of deaths in the locality to be adjusted by age groups. Thus, pOpulations or small areas ihich do not tabulate deaths by age can be adjusted. Less labor and time is required than for the direct method. The indirect method is more useful in adjust- ing local death rates in intercensal years. Newsholme says, “The factor of correction (standardizing factor) once ascertained, can be employed, with approximate accuracy during ' an entire decennial period, provided that no great change occurs in the character of the pop- ulation.'1 Disadvantages: l. The indirect method is considered less refined {Adjustigg the Death Rate of’Michigan.by thegIgdirect method: Applying the standard specific death rates for the by some authorities, who regard it as an approx- imation of the direct method. (This is not generally considered to be true. Note Pearl's statement in Advantage 2 above.) United States, 1940, to the pOpulation age groups of Michigan, T Newsholme, su- Arthur, pp. it., p. 222.. 324 the “expected deaths" are obtained which would have occurred had the United States had the same population age distribu- tion as Michigan. Deaths occurring in Michigan have not yet entered the computations. The sum.of the expected deaths in each age group divided by the population of Michigan, 1940, and the result in turn divided into the crude rate of the standard population gives an adjustment factor. This adjust- ment factor'multiplied by the crude death rate of Michigan gives the standardized death rate of Michigan using the in- direct method. An example of the indirect method worked out in de- tail for Muchigan, 1940, is presented in Table I. The re- sults obtained by employing the indirect and direct methods to adjust the death rate of Michigan are: Indirect method: 10.08 Direct method: 10.19 325 TABLE I CONDE’UTATIONS ma STANDARDIZING Tar DEATH RATE or MICHIGAN, 1940, FOR AGE, USING m AGE SPECIFIC DEATH Rims 03‘ THE UNITED STATES, 1940, as THE STANDARD: INDIRECT METHOD. Standard Michi an W Specific Death 13363080; Deaths , opu- POpula ti on 1 x 2 Age lation 1940 '1U&:8U% (U.s . 1940 L 001. 1 Col. 2 501. 5 0-~--- 4 1,287 451,384 5,552 5----- 9 108 418,855 452 10-----14 99 485,917 459 15-----19 172 478,220 885 20.----24 240 457 ,151 1,097 25-----54 506 848,848 2,597 45-..--- 54 1,060 654,347 6 ,936 55-----64 2,220 408,897 9,108 65-----74 4,799 229,842 11,510 75---Over 12,910 101,012 13,041 Total 1,075 5,258,106 55,540 Note: POpulation enumerated as of 4-1-40. Population of unknown ages, if any, excluded. Formula: Set up for death rates per 1,000 population. TLD Du wh re 1 . mm ' ‘SuTn o; ganja x (k) "—51—"— 0 I. MUm.-.Age-adjusted death rate .indirect method Ds - Deaths in standard p0pu ation Ps = Standard population Msa 3 Standard specific rate Pua = POpulation per age group in area to be standardized Du 3 Deaths in area to be standardized Pu 3 Population in area to be standardized k ' 1,000. . Mm i 10e75 x 9e91 10e75 I gegl 5,256,153 ‘ 10.75 x 9.91 ' 1.0170 x 9.91 3 10.08 f. 325 ADJUSTED DEATH RATE, DIRECT METHOD Defined: “In its simplest form.the direct method relates the age-specific death rates of a given community to some popu- lation taken as a standard, to find for each age group that the expected number of deaths in the standard population would be if the age-specific mortality of the community ap- plied. This is accomplished by multiplying the specific rate for each age group by the population for the correspond- ing age group in the standard population. The adjusted death rate, direct method (Mum), is formed by adding the expected deaths for each age group and dividing the sum by the total population that was taken as a standard.*1 Nomenclgturg: The direct method has been referred to by various terms: ' l. "Corrected death rates.“ The Registrar-General of England and Wales named the method 'corrected death rates" originally. a. "Standardized rates." When adopted by the United States Bureau of the Census in 1911, the method was called "standardized rate." 3. ”Adjusted rates.“ Professor 0. C. Whipple of Harvard, and other authorities on vital statis- tics prefer the term "adjusted rates." fLinder, Forrest r. and Grove, Robert D., pp. cit., p. es. 4. 327 “Age-adjusted," 'Age-sezeadjusted rate.“ These terms are recommended by the Committee on Forms and Methods of Statistical Practice of the Amer- ican.Public Health Association. It is prefer- able also to state somewhere the particular method employed, such as "Age-adjusted death rate, direct method." Information Engaged in Use of Dlpeot Mothpq: 1. 3e 3. 4. Advgptgggs: 1. The standard million age distribution based on a community suitable to the purposes of the re- search being conducted. The population by age groups of the locality for which death rates are to be adjusted. The number of deaths by age groups for the le- cality to be adjusted. The age-specific death rates of the locality to be adjusted. The direct method is employed by the United States Bureau of the Census. Because it is the official method of the government, it gains prestige and authority. The United States Department of Health recommends the method to State Health Departments. The direct method is equally as reliable as other methods. In fact, it is claimed by some authori- ties to be a little more accurate. "328 Disadvantages: l. The direct method requires deaths by ages for the locality to be adjusted. 2. Computations are more laborious and time-consume ing then for the indirect method. 3. It requires a "standard million“ which is an ad- ditional computation if a suitable one is not available. Adjustigg the Death Rate of Michigan by the Direct method: Applyingjthe age-specific death.rates of Michigan to the age groups of the chosen standard population of 1,000,000 (United States, 1940), one obtains the "expected number of deaths,“ that is, the number of deaths which.would occur if the population of Michigan had the same age distribution as that of the United States in 1940. The sum.of the expected number of deaths divided by the population in the standard community, (1,000,000) and this quotient multiplied by 1,000 gives an age adjusted death rate for Michigan per 1,000 popue lation. Age-adjusted death .‘§ugrg£_§§p%gpggfgumpgg_g%gphgix 1,000 rate, direct method opu at on o standar ‘ community (1,000,000) An example of this method of adjusting the deathrate for age is worked out in detail in Table II. Michigan data for 1940 is used. Table III illustrates the mechanics necessary for ad- _justing the white death rate of Michigan, 1940, for both age and sex as based on.the age distribution of whites in the 329 TABLE II COMPUTATIONS FOR STANDARDIZING THE DEATH RATE FOR MICHIGAN, 1940, FOR.AGE DISTRIBUTION, USING THE TOTAL POPULATION OF THE CONTINENTAL UNITED STATES, APRIL 1, 1940, ,AS THE STANDARD POPULATION: DIRECT METHOD, PER 1, 000 PEOPLE. Specific Expected' Standard Death Rates Deaths in Million (MICHIGAN per 100,000 Standard for the PopuIEtion DeatEs Po ulation Million Age U.S. 1940 1940 1940 3 1 2) 4 x 1 EOIe 1 001. 2 EOIe E EOle { COle 3 0--- 4 80,100 431,384 4,783 1,109 888 5--- 9 81,100 418,855 418 100 81 10--—l4 89,200 463,917 453 98' 87 l5---19 93,700 478,220 718 150 141 20---24 88,000 457,151 914 200 176 25---34 162,100 848,848 2,101 248 402 35---44 139,200 763,633 3,472 455 633 45---54 117,800 654,347 6,255 956 1,126 55---64 80,300 408,897 8,649 2,115 1,698 65---74 48,400 229,842 11,077 4,819 2,332 75—0ver 20,100 101,012 13,227 13,094 2, 632 Total 1,000,000 5,256,106 52,092 991 10,196 Unknpwn pp 25 Note: Death and population of unknown ages, if any, are ex- Source:Sixt t-ninth Mflm. Mdm lua Psa Specif k 8 1, 000 Mdm (Michigan) - min—ratfitrm Sum.of Mua.Psa Sum.of Psa Age-adjusted death rate, direct method c death rates of a given community . Population by age groups in standard mil ion, U.S. 1940 Sum of (5) prected deaths in a standard million Sum.of‘Tl) standard million for the U .S e 1940 W10 196 x 1,000 9 9 ' .010196 x 1,000 10.20 ' 10.2 cluded from Columns 2 and 3, but are included in totals. ual Report, Michigan Department of chigan, 1941, Table 13, p. 264. k where, x 1,000 r1 330 TABLE III COMPUTATIONS FOR STANDARDIZING THE WHITE DEATH RATE FOR MICHIGAN, 1940, FOR AGE AND SEX, USING THE WHITE POPULATION OF THE CONTINENTAL UNITED STATES, APRILil, 1930, AS THE STANDARD POPULA- TION: DIRECT METHOD. J_ MALE _ White Age Expected Standard Specific Rates in Million Rates,l940,Standard for The per - Million U.S. 1930 MICHIGAN'WHITE 100,000 (4! x El) PopfiIation Deaths Population , 0 Age 1940 (SH-(2) x 100,000 ‘ COIe I COle 2 COle 3 COle‘l 001; 5 0--- 4 46,434 212,100 2,632 1,241 576 5--- 9 51,099 204,242 230 113 59 10---14 49.090 226,837 245 108 53 15---l9 46,552 232,192 409 176 82 20---24 42,894 218,346 460 211 91 25---29 39,042 208,867 481 230 90 30---34 37,316 196,949 548 278 104 35---39 38,345 192,696 725 376 144 40---44 34,322 182,292 1,013 556 191 45---49 30,279 181,243 1,496 825 250 50---54 25,880 157,247 2,055 1,307 338 55---59 20,455 119,032 2,375 1,995 408 60---64 16,446 90,703 2,621 2,890 475 65---69 12,161 67,035 2,922 4,359 530 70---74 8,580 46,374 3,118 6,724 ,577 75-0ver 7,827 48,304' 6,655 13,777 1,078 Total 506 722 2 584 459 28 005 1 084 5 045 Note: THenty’fiiie deaths incIuded In the totaI, of Hitfiown age, which are excluded from.the age groups. Source:Standard Million: from Hagood, Margaret J., Statistics for Sociologists, Reynal and Hitchcock, Inc., ew or 1 Po 4 ‘ 4 e Deaths: from Vital Statistics of the United States, 1940, Part II, dbl—_Tge c, p. oZ'IeI". """'" """""" Formula: (Set up for death rates per 1,000 population.) Age-sex adjusted death rate: Direct method: (Mdm) mum,. Sum.of {mMua x mPsa plus fMua x fPsa) where Sum.of sa plus f as mMua 8 Specific male death rate of a given community than ' Specific female death rate of a given community mPsa =Male pOpulation by age groups in standard million, U.S. 1930. . fPsa 8 Female pOpulation by age groups in standard 1111111011, U.S. , 19300 r! (9 (I I. I. 1‘ IV I. f. I. I. II 331' TABLE III (Cont inued) - COMPUTATIONS FOR STANDARDIZING THE WHITE DEATH RATE FOR MICHIGAN, 1940, FOR AGE AND SEX, USING THE WHITE POPULATION OF THE CONTINENTAL UNITED STATES, APRIL 1, 1930, AS THE STANDARD POPULA- TION: DIRECT METHOD t m f White Age Expected Standard Specific Rates in Million Rates,l940 Standard for the per 100,000 Million United MICHIGAN WHITE Population Age States Population Deaths 8 . 100,000 56! 2 £9] Iago £940 1940 7f COIe 6 COle 9 COle 8 0010 9 COte 15 5--- 9 49,613 197,186 165 84 42 10---14 47,855 , 218,641 171 78 37 15---19 46,396 228,474 226 99 46 20---24 44,121 221,291 359 162 _ 71 25---29 39,770 - 208,837 362 173 7 69 30---34 37,232 192,417 428 222 83 35---39 36,881 178,965 548 306 113 40---44 31,849 165,760 728 439 140 45---49 27,844 156,743 _923 589 164 50---54 23,841 133,763 1,262 \943 225 55—--59 19,009 104,639 1,539 1,471 280 60---64 15,515 83,828 1,759 2,098 326 65---69 11,692 65,150 2,223 3,412 399 70---74 8,316 46,154 2,528 5,477 .455 75-0ver 8,523 50,995 6,338 12,429 1,059 Total 493,278 2,455,184 21,446 873 3,924 Note: Twelve female deaths of unhnown age included in the total which are excluded from.the age groups. Source:Standard Million: JHagood, Margaret J., Statistics for Sociologists, Reynal and Hitchcock, nc., New YOI‘K, 94 , P. 40-841. Deaths: from.Vital Statistics of the United States, 330, Part II, a 1e , p. 02'1ch Age-Sex Adjusted Death a Rate of White Race, . Sum.Col 5 plus Sum.Col 10 :x 1,000 NHchigan, 1940 um Co p us Sum Co , 3 5045 lus 3924 A a 506,722 plus 493,278 8,969 9 9 = 0008969 X 1,000 3 89969 I 9.0 x 1,000 l‘ I. f. 332 United States in 1930. The adjusted death rate is obtained for age and sex as follows: Age-Sex-adjusted Death Rate; of White Race I‘_§um.Col. 5 plus Col. 10 X 1,000 Michigan, 1940 Sum of Col. 1 plus 001.76 The Standard Million: White population of the United States, 1930, (Col. 1 and Col. 6) in Table III is the number of persons in each age-sex group in a population of 1,000,000 which has the same age-sex distribution as the white population of the United States in 1930. This standard million can be used for any age-sex-adjustment of death rates for Which the Ihite population of the United States in 1930 is an appro- priate choice. EQDIVILENT AVERAGE RATE METHOD Defined: This method is defined as "...an arithmetical mean of the rates of quinquennial groups of ages up to some con- venient lbmit of age such as 65, this being equivalent to calculating a standardized death-rate at ages under 65 based upon a pOpulation equally distributed over the 13 age groups."1 Raymond Pearl first suggested this method in 1923. G. U. Yule preposed and adapted the method to the study of I Linder, Forrest E. and Grove, Robert D.,‘gp. cit., p. 81. See also.ghg_Registrar-General's Statistical Review'of W e f r 1 figgland and g s g Jhg_Year.I937. New AnnuaI SerTEs, c. 7, Text. . . Stationery Office, London, 1940,p. 11. 333 differential occupational mortalities in 1934 and used it extensively.1 . Ipfgrmation Reguired in Adjusting Death Rates: 1. The specific-age death rates for the locality to be adjusted. (up to age 65) a. Weights for the size of the age group: (Weights consist‘ of 1 for each five-year age group, 2 for each ten-year age group, etc.) 3. The weighted specific rate must be obtained by multiplying the weights by the ageespecific death rates of the community to be adjusted. An example of the computation of an adjusted death rate for muchigan, 1940, by this method is given in Table IV. Advantagegz 1. It is independent of any standard population. 2. The method is useful in comparing international death rates where pOpulations by age distribution are uncertain. Yule has used the method extensive- ly for studying occupational mortality there the population over age 65 is not important. 3. The Equivalent Rates method of adjusting death rates is the easiest and the quickest to compute of the three methods discussed. IiIule, G. U., “On Some Points Relating to Vital Statistics, Mere Especially Statistics of Occupational mortality," , Journal‘gf the Royal Spatistical Societ , vol. 97, 1934, T1084. 334 TABLE IV COMPUTATIONS FOR STANDARDIZING THE DEATH RATE PER 100,000 POPULATION, MICHIGAN, 1940, FOR AGE, USING THE.EQUIVALENT AVERAGE DEATH RATE METHOD. Michigan WSW Specific Rates Weights Specific _ége ger iOOiOOO for Size of Rats ) opu at on e Grou s 2 x 1 ##7 GOIe I OI. 2 0 e 0--- 4 1,109 1 1,109 5--- 9 100 1 100 10---l4 98 l 98 15---l9 150 1 150 20---24 200 l 200 25---34 248 2 496 35---44 455 2 910 45---54 956 2 1,912 55---64 2,115 2 4,230 65 plus 7,346 Note: Deaths and population of unknown ages, if any, excluded. E uivalent average death rates Sum of §M¥g . "91) .w q Sum.of We . here Mus I Wci ' Weight of class interval k = 100 Equivalent average death rate- Age-specific death rates of given community Sum.of weighted specific rates (Column 3) Sum of weights for class interval (Column 2) times 100 9 205 W 9 205 19366 7.08 335 Digadventggeg: 1. 2. 3. l. 2. The equivalent average rates method cannot be safe- ly employed in studying general mortality for all ages. According to Linder and Grove, 'The requirement that the storage (or sum) be cut off at 'some cons venient limit of age such as 65' is necessary for two reasons: first, because without such a limit it would be difficult to assign a weight to the open-end age group (for example, '65 years and over'): and second, because if the limit were set too high, the index would be over-ihelmingly dominated by the specific rates of the later ages.'1 The weighting systan must be kept up-to-date. » ADVANTAGES OF ADJUSTED DEATH RATES Age-adjusted death.rates of local areas can be compared without being influenced by abnormal age distributions. Since age is held constant, any differences that may appear in the rates of local areas will not be due to abnormal age dis- tributions. There is economy of expression in the use of ad- justed death rates. 7A single figure may measure ‘I—Ldnder, Forrest E. and Grove, Robert D.,‘gp. cit., pp. 81-82. 3. 4. 5. l. 356 the deaths of an area, holding constant age, sex, race, residence, or any combination of these which available data suggests may be unduly in- fluencing crude rates. This single figure may be compared to a similarly computed rate for other areas, greatly reducing the number of tables and comparisons while at the same time being more precise and reliable. Adjusted rates are truer measures of the real force of mortality in areas or regions of a na- tion or state than are crude rates. The principle employed in adjusting rates may be used to estimate crude death rates in future years. The adjustment of death rates now is not im- practicable because of the time and labor re- quired for computation, and the cost involved, if modern calculating machines are available. DISADVANTAGES 0F ADJUSTED DEATH RATES Adjusted death rates eliminate any variation present due to age, and by so doing also elimi- nate any variables which may be associated with the age of the pepulation. In communities where health problems may be determined largely by the peculiar characteristics of the age distribution, 337 age-adjusted death rates may not reflect the health problems characteristic of the age groups which are predominant in the pepulation. 2. Adjusted rates may exaggerate the differences in crude rates rather than reducing the amount of dispersion between them. The Registrar-General of England and Wales states, "The correction for differences in age distribution accentuates the contrasts between the mortalities of the northern industrial towns and the residential and agri- cultural towns rather than diminishing them.'1 He explains that “rai- from accounting for part of the wide differences in mortality rates be- tween.individua1 county boroughs, the peculiari- ties in age distribution tend in general to mask these differences, the more favorably circums stanced towns having larger prOportions of old people.'3 3. tAdjusted death rates are limited for use in public health data due. to the lack of power of the human mind (untrained in statistics) to grasp the full meaning of a number of characteristics combined in a single figure. 4. The requirement of population by age groups hinders ITIdnder, Forrest E. and Grove, Rebert D.,.gp. cit., p. 87. 3 Linder, Forrest E. and Grove, Robert D., ibid, p. 87. l. 2. 3. 4. 5. 338 the use of the methods for adjusting death rates in local areas between decennial years. Estimated populations by age groups are not usually avail- able in local areas. MORTALITY FORMULAS Crude resident death rate: Number of deaths of residents in area for s ecified eriod Total residents of given area for the same specified time Recorded death rates: (by place of occurrence) Number of deaths occurring in area for specified eriod Tfifial population 5? area for same speci fied time Age-Specific Death Rates: Deaths occurring to persons in given age group for specified time ‘ 'Persons exposed to risk of death in same age group for same peri od Sethpecific Death Rates: Deaths occurring to persons in iven sex group for given time ersons exposed to risk of death in same sex group for same period Cause-Specific Death Rate: Deaths from specific cause for iven area and eriod Total population exposedtc risk of death, same area and period I 1,000 x 1,000 x 1,000 x 1,000 6. 7. 8. 9. 10. ll. Age-Adjusted Death Rate: (Direct Method) Age-specific Sum. death rates 11 by age groups iven area in stand mil.) Population - Standard m.ll on Age-Adjusted Death Rate: (Indirect Method) Crude death rate of standard 0 ulation (Standard) I (Population by, Sum specific; I (age group in 1) of rates (area to be 3 adjusted POpulation of area to—be adjusted Death Ratio: Deaths in specified class for given.period of time otal deaths during the same period Infant Death Rate: Deaths under 1 year of age except stillbirths, specified riod Total live births during same period Maternal Mortality Death Rate: ‘A11 deaths from puerperal causes _fgr given area and time Number of live births forsame area and peri od Under 5-Year Death Rate: .All deaths at ages 0-4 except stillbirths, given area and time ving population at ages Tot 0-4, same area and time 339 X 1,000 (Crude death) rate in area to be adjusted x 1,000 X 1,000 I 1,000 X 1,000 x 1,000 340 HH O.O a O O.O e O O O m H onooaoao 0.0H a O O O.OH N O O a O H noonHHO HH O a O OH O n m a O H omeHO HH OH O a O O O O a m H esoooHnO HH O O s OH O O a O m H nemaonono HH OH O a O O O a n m H aHoeoHaenO HH O a O.O O.O a O m m m H emeO HH OH O.O a n.m O O 4 N O H asonHeO 0.0H 0.0 a 0.0 0.0H O O a m n H nooenm HH OH O.a O.a O O O a O m H aoHnnom HH a O O OH O O n e m H oHnnom HH OH O a O O n e n m H gem OH O O O HH O o e m n H annem O HH OH O O O m N O e H omenem OH HH O O a O m m e n H onsets HH a 0.0 O.O OH e O O N O H anosH HH O s OH O O O a m n H enoaHe HH O a O OH O O a O m H nemoHHs OH HH O O a O a N O n H noMHH OH HH O a O O O a m n H onoeHH HH OH O O a O m e O m H opoom e fiaHm oOHo mHmonoHom mopop snap meHH mHmoa apnea hHon Moo uHsm uoHnopnd uan umaoam .m.e Ignaz isoqm nHoom load ammo endow hpssoo mammoo zoom ma omzmo hp mopmm spoon ho Edam . eeOH-OnOH mOamm>e .zeOHmOHs ewapoO am zOHBaHOmOm OOO.OH mam mmOHOHOm one manna so nmmOHO OzHaHmH zme aO Hana H oHoea HH sHoneaaH 341 Nwssoo HOOM sH Omsmo Np nepOm momma yo xssm 0.0H O.OH O O a O O O O O H seneHeoH HH O O OH O O O O O O H nooaeH O OH O .HH O O O O O O H OOOH OH O a HH O O O O O O H senooaem HH O a O.O O.O O O O O O H pang HH O OH O O O O O O O H OOOOOHOO HH O.O O.O O OH O O O O O H ooOesOHOO HH a O O OH O O O O O H nonaeeO HH O OH O O O O O O O H OHHOOOOH HH 0H m a O O O n O m H soaH HH O OH O O O O O O O H onOH HH a O O OH O O O O O H OHOOH OH O O a HH O O O O O H _sOnOnH HH O O O OH O O a O O H aoasm OH HH O O O O a e O . O H soonOsom O a. O HH OH O O O O O H OHOOOHHHO HH a O O OH O O O O O H poHpOnO HH OH O.O O.O O O O O O O H omaoeeae OOOOO OH HH O O O a O O O O H OHOOOoO HH OH O a O O O O O O H nHeOOHO HH 0 OH O O O m n w m H memosow HH O.O O.O O a O O O O.O O.O H posse OH O O O HH O O O O O H noses HH OH O O O O a O O O H OOOOHOOHO HH OH O O a O O O O O H OOHOO OBOE OOHO OHOonOHom moves one» meHn OHmQa mmeO OHOHH poo :Hsm IoHHOPH4 uOHm Inseam .m.a inmoz locum IHood toad ammo phmom handoo Ooqupsoo H manna 342 HH O O.O O OH O O.O O O O H OOOOOO O.O O.O HH O O O O O O O H OOOOOO HH O O O O O O O O O H Oeoeeo HH O O O OH O O O O O H eHooomo OH HH O O O O O O O O H OoOeaoonO HH OH O O O O O O O O H seseOO O.O O O OH O.O O O O O O H OOOOOO .2 OH O O O O O O O O H Eadeo HH O O O OH O O O O O H OOOOOOZ HH OH O O O O O O O O H OOOoOOOO O.OH O.OH O.O O O.O O O O O O H Oenoaosenoz HH O O O OH O O O O O H sHeeonom HH OH O O O O O O O m H condom OH HH O O O O O O O O H eeOeOOOHH O.OH O.OH O O O O O O O O H OOOHOHO HH OH O O O O.O O.O O O O H eoaHsonea HH O O O OH O O O O O H eemoeoz HH m m OH O O O O n m H momma OH. HH O O O O O O O O H eopesOOOH HH O O OH O O O O O O H OOOOHOOH HH OH O O O O O O O O H Osoeea HH OH O O O O O O O O H onHOeOO 0H HH O O O O O N O O H oosH 0.0H O O O O.OH O O O O O H OOOOOOHOHH HH O m m 0H O O O m n H ooBOmoH OOOOHO OOHo OHOOHOHOO nepop ens» meHH «Haas mpsou OHOHQ Moo :Hdm oHHOOHO iOHQ imaonm .m.B unmoz. {zoom :Hood :094 undo pHOOm hammoo handoo OOOM mH Omsmo Op wepOm npwom yo mqwm OOOOHOOOO H eHOOO 343 .O.O-O.O OOHHOO "eonsom HH 0H 0.0 0.0 O O O O O m H unease: OH HH O O O O O O O O H .923 HH O OH O O O O O n m H endowment HH O O 0.0 0.0 O O O n w H sensm qe> HH OH O O O O O O n m H sHoonsa HH O O OH O O O O O O H Hence .3 HH OH O O a O O O O m H uHuHo .am HH O O O OH O O O O m H eenmsannm 0H HH O O O O O n O O H pushoHooHom HH OH O O a O O O a O H oeHHssm HH OH O O a O O O O a H nssHOnm 0.0 0.0 O HH O O O n v m H aossoonom HH OH O O O O O O O O H .HOH eeueonm . anvaHm eOHo aHnonOHon aepep ens» erHn eHsos apnea OHeHm nee [Ham ioHnepnd idHn inseam .m.O unnoz -sosm 3Hoc4 .094 undo venom Ownsoo qusoo moon nH ensue Ho aevsm queen no Edam eons HosoO H .33. APPENDIX II Table 3 m or rm PRINCIPAL “ms or 13mm IN Bum nuns. 3! com, MICHIGAN 1935-1944 (Rum 901: 10, 000 pOpulation) { w State and z z : . : County :Hbart :Canoortlpoplozy : Accident :NUphritil Stat. 8 3 £ (i U Alcona Alger Allogun Alpena Antrim Aronaol Baraga Barry Buy Danni. Borrion Branoh Cglhoun Cass Charlevoix Cheboygan Chippewa Clare Clinton Crawford Bolt. Diokinlon Baton lmmot Gonoloo Clnawin Gogebic Grand Tioyurao Gratiot Hilladalo D GGflN§FOO§’ flflfiflflfifififlfi o a Houghton Enron Ingham. Ionia Iosoo Iron Isabella Jackson Kalamazoo Inlknokn HPPHPPPPPH HHPHPPHPPH PPHPHPPHPH PPFPHPPHHP H uuuuuuuaun uuuauuuuu? uaafiuuuuuu uaubu a uaau’uuuuu mumbapuuuh arruauuuuu oouaruyur aauuaauauu aaouuauaaa ruaaaa§aaa auauuauaau a OOOOOOOOOO OOOOOOOOO Table 2 Continued State and County Kent Keweenaw Lake Lapeer Leelanau Lenawee1 Livingston Luce Mackinac1 Macomb Mani stee Marquette Mason Mecosta Menominee Midland Missaukee Monroe Montcahm Montmorency Muskegon Newaygo1 Oakland Oceana Ogemaw Ontonagon Osceola Oscoda Otsego Ottawa Presque Isle Roscommon Saginav Ste Clair Ste Joseph Sanilac Schoolcraft Shiawassee Tuscola Van Buren ‘Washtinal Wayne Wexrord 1. Contains one or more urban places which were incifi He HHH HHHHHHHHHH HHHHHHHHPH HHHHPHHHHH HHHHHHHHHHf: t E Cancer1Apoplexy 010110 53106903191910” NNNGGPGNGN NGNNGNGGGN NIPGNGNGUIGN NIPOI GGNNNA$IPO§IF auuwmmmbmu GNPNUGNNNG GGFGNGNNNO‘I O. O. 345 Accident gNephritis OOO OOOOOOOOOO OOOOOOOOOO OOOOOOOOOO OOOOOOOOOO Cl #001 OIUIFUIIF’UIOIGOI mtuaumucnmm FUIOIUNOIOIUIUIO‘ 001000010010 ded with the rural in the Reports or the Michigan Department or Health, 1935-1940. These places are included with the rural throughout the ten year period, 1935-1944. APPENDIX II 346 Tabla 3 BANK OF FIVE PRINCIEIL CAUSES OF DEETH.IN URBAN BY COUNTY, MICHIGAN 1935-19“ (Rates per AREAS 10,00 _L population) gtate and County State Alger Allegan ‘Llpena Cass Charlevoix Cheboygan Chippewa Clinton Delta Dickinson Baton Emmet Genesee Gogebic Grand Traverse Gratiotl Hillsdale Houghton Ingham Ionia Iron Isabella Jackson Kalamazoo Kant Iapeer 1 Lenawee Livinglton Maconh Manistea Harquette lason Nbcosta anominea Midland O. .0 HPPHPHHHPP PHHPPPPHPH HHPPPHHPHH HHHHPPPPHH H Heart ; C 080030003000” NNNflNGUUNN 0000030000” 00000000000 Ouuhuuurau uuufipuauuu uabuubaOpu uuhuuuuuua ancerglpOplexy 3 ‘lccident 4 OOOOOOOOOO OOOOOOOOOO OOOOOOOOOO OOOOOOOOOO : : nephritis ”000000000 000000000. 0000000000 0000000000 347 TflWeSMmumnd State and : : : County :Heart : CancerzApOplexy : Accident : Nephritie Monroe Montcalm Muskegon (mane CWum Presque Isle figmur St e 01‘12 St. Joseph Sdhoolorart Shialaesee Tuscola WmBum: Waahtznaw Mm» Wexrord HHPPPP HHHHPPPHHH manna» uuuuuuuuuu vacuum puuauufiuno OOOOOO OOOOOOOOOO aaaaua aoaapammaa 1. Does not include one or more urban places which were. included with the rural in.the Reports or the Michigan Department or Health, 1935-1940. These places are in- cluded in the rural throughout the ten year period, 1935-1944 for purposes of computing the ten-year ayerage death rates. Source: Table Table 4 TOTAL NUMBER OF DEATHS FROM.ALL CAUSES, BY COUNTY, MICHIGAN, 1935-1944 State and Total Number State and Total Number County of Deaths County of Deaths State 530,282 Alcona 636 Lake 640 Alger 1,138 Lapeer 3,566 Allegan 5,140 Leelanau 949 Alpena 2,160 Lenawee 7,023 Antrim. 1,401 Livingston 2,681 Arenac 1,006 Luce 761 Baraga 1,036 Mackinac 1,041 Harry 2,981 Maconb 9,489 Bay 8,227 Manistee 2,466 Benzie 1,075 marquette. 5,185 Berrien 10,315 mason 2,356 Branch 3,769 Mecosta 2,289 Calhoun 11,470 menominee 2,747 Cass 3,244 Midland 2,261 Charlevoix 1,730 Missaukee 849 Cheboygan 1,666 ‘Monroe 5,944 Chippewa 3,116 Montcahm 3,995 Clare 1,193 Montmwrency 408 Clinton 3,101 Muskegon 9,648 Crawford 452 Newaygo 2,290 Delta 3,778 Oakland 20,744 Dickinson 2,725 Oceana 2,025 Eaton 4,414 Ogemaw 1,016 Emmet 2,082 Ontonagon 1,288 Genesee 19,341 Osceola 1,777 Gladwin 1,000 Oscoda 227 Gogebic 3,424 Otsego 646 Grand Traverse 2,696 Ottawa 5,773 Gratiot 3,921 Presque Isle 1,139 Hillsdale 3,998 Roscommon 375 Houghton 5,947 Saginaw 13,581 Huron 3,526 Sanilac 3,590 Ingham. 12,707 Schoolcraft 1,054 Ionia 4,475 Shiawassee 5,010 Iosco 939 St. Clair 9,163 Iron 1,847 St. Joseph 4,281 Isabella 2,737 Tuscola 4,318 Jackson 10,546 Van Buren 5,126 Kalamazoo 10,814 Washtenaw 8,756 Kalkaska 585 Wayne 179,008 Kent 25,693 Wexford 2,258 Keweenaw 488 Source: Computed from.the Annual Reports 2; the Michigan ‘Qgpartment‘gg’flealth, Lansing, Michigan, 935- 944, Table 16, pp. II /‘ IV 1‘ 349 Table 5 TOTAL NUMBER OF BIRTHS, BY COUNTY, MICHIGAN, 1935-1944 State and Total Number State and. Total Number County of Births County, of Births State 1,028,519 Alcona 1,063 Lake 807 Alger 2,135 Lapeer 5,325 Allegan 7,742 Leelanau 1,582 ‘Alpena 4,393 Lenawee _ 10,294 Antrim. 2,301 Livingston 3,620 Arenac 1,670 Luce 1,521 Baraga 1,714 Mackinac 2,058 Barry 3,970 Macomb 23,622 Bay 16,251 manistee 3,318 Benzie 1,665 Marquette 8,664 Berrien 16,566 Mason 3,508 Branch 4,741 Mecosta 3,458 Calhoun 18,982 Menominee 5,573 Cass 3,336 Midland 7,132 Charlevoix 2,684 Missaukee 1,896 Cheboygan 2,816 Monroe 11,845 Chippewa 6,808 Montcalm 5,419 Clare 2,027 Montmorency 790 Clinton 5,257 Muskegon 22,002 Crawford 762 Newaygo 3,734 Delta 6,896 Oakland 55,683 Dickinson 5,135 Oceana 2,696 Eaton 6,589 Ogemaw 1,878 Emmet 3,410 Ontonagon 2,015 Genesee 49,398 Osceola 2,660 Gladwin 2,111 Oscoda 519 Gogebic 5,962 Otsego 1,303 Grand Traverse 4,391 Ottawa 12,218 Gratiot 6,899 Presque Isle 2,639 Hillsdale 5,264 Roscommon 717 Houghton 7,635 Saginaw 27,491 Huron 6,775 Sanilac_ 5,786 Ingham. 28,169 Schoolcraft 2,145 Ionia 6,470 Shiawassee 8,044 Iosco 1,645 St. Clair 15,228 Iron 3,318 St. Joseph 6,317 Isabella 5,584 Tuscola 7,116 Jackson 16,538 Van Buren 5,640 Kalamazoo 19,232 Washtenaw 15,576 Kalkaska 1,147 Wayne 377,137 Kent 45,800 Wexford - 3,682 Keweenaw 615 Source: Computed from.the Annual Reports or the MHchi an Department of’Health, Lansing, Michigan, Table 16, pp. 935-1944, Pu 350 Table 6 TOTAL NUMBER OF DEATHS FROM HEART DISEASE, BY COUNTY, MICHIGAN, 1935-1944 State and Total Number State and Total Number County of Deaths County of Deaths Stats 149,925 Alcona 192 Lake 182 Alger 261 Lapeer 996 Allegan 1,612 Leelanau 318 Alpena 693 Lenawee 1,924 Antrim 409 Livingston 865 Arenac 305 Luce 286 Baraga 217 Mackinac 279 Barry 855 Macomb 2,386 Bay 2,279 MMnistee 823 Benzie 309 Marquette 1,398 Berrien 3,056 {Mason 676 Branch 1,149 Mecosta 671 Calhoun 3,446 menominee 766 Cass 854 Midland 546 Charlevoix 527 Missaukee 247 Cheboygan 483 Mbnroe 1,727 Chippewg 934 Montcalm 1,141 Clare 355 Montmorency 104 Clinton 816 muskegon 2,570 Crawford 116 Newaygo 726 Delta 997 Oakland 5,170 Dickinson 780 Oceana 539 Eaton 1,371 Ogemaw 278 Emmet 687 Ontonagon 310 Genesee 4,932 Osceola 518 Gladwin - 277 Oscoda 62 Gogebic: . - 819 Otsego 207 Grand Traverse , 811 Ottawa 1,748 Gratiot 1,150 Presque Isle 282 Hillsdale 1,121 Roscommon 112 Houghton 1,647 Saginaw 3,855 Huron 915 Sanilac 1,010 Ingham. 3,422 Schoolcraft 307 Ionia 1,322 Shiawassee 1,320 Iosco 288 St. Clair 2,555 Iron 478 St. Joseph 1,200 Isabella 768 Tuscola 1,337 Jackson' 3,142 Van Buren 1,627 Kalamazoo 3,041 Waahtenaw 2,535 Kalkaska 176 Wayne 50,101 Kent 8,389 Wexford 706 Keweenaw 114 Source: Computed from.the Annual Repw of the Michim De artment of H__e___a1th, Lansing, tMiChlgSn, 935- 944, Tr2_—_—__-_ able 16, pp. 351 Table 7' TOTAL NUMBER OF DEATHS FROM CANCER, BY COUNTY,.MICHIGAN, 1935-1944 w 3‘ 1 State and Total Number State and Total Number County of Deafihs County of Deaths State 60,752 Alcona 6 1 Lake ‘ 66 Alger 100 Lapeer 382 Allegan 572 Leelanau 104 Alpena 210 Lenawee 735 Antrim. 155 Livingston 301 Arenac 97 Luce . 51 Baraga 85 Mackinac 97 Barry 314 Macomb . 1,072 Bay 982 menistee 305 Benzie 99 Marquette 601 Berrien 1,158 Mason 256 Branch 379 mecosta 211 Calhoun 1,196 menominee 339 Cass 374 Midland 239 Charlevoix 204 Missaukee 68 Cheboygan 196 Monroe 667 Chippewa 275 Montcalm. 402 Clare 109 Mbntmorency 46 Clinton 313 muskegon 1,001 Crawford 57 Newaygo 237 Delta 441 Oakland 2,335 Dickinson 317 Oceana 217 Eaton 525 Ogemaw 84 Emmet 223 Ontonagon 126 Genesee 2,052 Osceola 198 Gladwin 109 Oscoda 28 Gogebic 362 Otsego 67 Grand Traverse 271 Ottawa 801 Gratiot 433 Presque Isle 127 Hillsdale 450 Roscommon 40 Houghton 759 Saginaw 1,483 Huron 409 Sanilac 364 Ingham 1,485 Schoolcraft 125 Ionia 505 Shiawassee 615 Iosco 123 St. Clair 1,059 Iron 238 St. Joseph 478 Isabella 294 Tuscola 488 Jackson 1,203 Van Buren 625 Kalamazoo 1,218 7ashtenaw 1,066 Kalkaska 57 Wayne 21,262 Kent 3,315 Mexford 210 Keweenaw 49 Source: Computed from.the Annual Reports of the Michi Department or Health, Lansing, Michigan, 935- 944, Table 16, pp. 352 Table 8 TOTAL NUMBER or DEATHS IRON 190nm, BY COUNTY, MICHIGAN, 1933-1944 State and Total Number State and Total Number County or Deaths County of Deaths State 43,270 Alcona 70 Lake 73 Alger 65 Lapeer 421 Allegan 336 Leelanau 98 Alpena 217 Lenawee 732 Antrim 161 Livingston 238 Arenas 91 Luce 46 Baraga 86 Mackinac 100 Barry 382 Macomb 925 Bay 688 Hanistee 294 Benzie 98 Marquette 438 Berrien 938 Mason 829 Branch 434 Mecosta 248 Calhoun 1,239 Menaninee 868 Case 339 Midland 139 Charlevoix 174 Missaukee 108 Cheboygan 168 Monroe 303 Chippewa 247 Montcalm 436 Clare 103 Montmorency 23 Clinton 304 Muskegon 800 Crawford 23 Newaygo 238 Delta 398 Oakland 1,690 Dickinson 831 Oceana 236 Baton 442 Ogemaw 130 Emmet . 228 Out onagon 131 Genesee 1,498 Osceola 211 Gladwin 103 Oscoda 18 Gogebic 330 Otsege 42 Grand Traverse 329 Ottawa 391 Gratiot 348 Presque Isle 70 Hillsdale 471 Roscommon 88 Houghton .608 Saginaw 1,169 Huron 332 Sanilac 430 Ingham 1,368 Schoolcraft 98 Ionia 467 Shiawassee 480 Iosco 101 St. Clair 890 Iron 140 St . Joseph 484 Isabelle 236 Tuscola 439 Jackson 1,002 Van Buren 376 Kalamazoo 1,087 Washtenae 872 Ialkaska 73 lane 11 , 322 Kent 2,480 lexrcc'd 208 Keweenaw 63 Source: Computed from the Annu Re ts or the Mich en Hartman; g Realm—fins ng, chigan, 9 44, a l. 6’ O Table 9 TOTAL NUMBER OF DEATHS FROM ACCIDDTTS, B! W State and Total Number State and Total Number County at Deaths County of Deaths State 7,399 Alcona 39 Lake 38 Alger 138 Lapeer 299 Allegan 428 Leelanau 67 Alpena 169 Lenawee 368 Antrim 83 Livingston 233 Arenas 103 Luce 79 Baraga 126 Mackinac 114 Barry 236 Macomb 864 Bay 647 Manistee 139 Benzie 97 Marquette 393 Berrien 873 Mason 186 Branch 283 Mecosta 163 Calhoun 846 Menominee 193 Cass 283 Midland 197 Charlevoix 134 Missaukee 38 Cheboygan 138 Monroe 610 Chippewa 269 Montcalm 337 Clare 189 Montmore nay 44 Clinton 226 Muskegon 637 Crawford 43 Newaygo 133 Delta 330 Oakland 1,848 Dickinson 238 Oceana 163 Eaton 353 Ogemaw 93 Emmet 133 Ontonagon 123 Genesee 1,392 Osceola 143 Gladwin 92 Oscoda 20 Gogebic 338 Otsego 38 Grand Traverse 200 Ottawa 377 Gratiot 292 Presque Isle 87 Hillsdale 304 Roe common 36 Houghton 407 Saginaw 1,060 Huron 309 Sani lac 274 Ingham 1,030 Schoolcraft 96 Ionia -‘ 303 Shiawassee 395 Iosco 78 St. Clair 801 Iron 193 St. Joseph 306 Isabella 237 Tuscola 341 Jackson 807 Van Buren 412 Kalamazoo 880 Washtenaw 683 Kalkaska 40 Wayne 11,632 Kent 1 , 79 2 Next ord 160 Keweenaw 34 Source: Computed from the ual Re arts the Michi De artment 3; Healfising, Mic gan, 1935-1544, atria-7n. 354 Table 10 TOTAL NUMBER or DEATHS FRCI PNEUMONIA, BY COUNTY, MICHIGAN, 1935-1944 State and Total Number State and Total Number County of Deaths County of Deaths State 30,029 Alcona 86 Lake 40 Alger 77 Lapeer 189 Allegan 193 Leelanau 41 Alpena 94 Lenawee 358 Antrim 78 Livingston 134 Arenae 39 Luce 40 Baraga 34 Mackinac 73 Barry 186 Macomb 389 Bay 396 Manistee 104 Benzie 42 Marquette 873 Berrien 309 Mason 83 Branch 183 Mecosta 98 Calhoun 613 Menominee 188 Cass 144 Midland 119 Charlevoix 72 Missaukee 41 Cheboygan 77 Monroe 348 Chippewa 136 Montcalm 153 Clare 83 Montmorency 38 Clinton 140 Muskegon 668 Crawford 80 Newaygo 183 Del ta 232 Oakland 1 , 832 Dickinson 139 Oceana 74 Baton 181 Ogemaw 36 Emmet 93 Ontonagon 71 Genesee 1,413 Osceola 63 Gladwin 39 Oscoda 16 Gogebic 191 Otsego 33 Grand Traverse 142 Ottawa 173 Gratiot 168 Presque Isle 63 Hillsdale 198 Roscommon 88 Houghton 241 Saginaw 787 Huron 804 Sanilac 173 Ingiam 353 Schoolcraft 38 Ionia 183 Shiawassee 829 Iosco 36 St. Clair 441 Iron 106 St. J oseph 232 Isabella 133 Tuscola 206 J aekson 431 Van Buren 193 Kalamazoo 331 lash tenaw 438 Kalkaska 43 Wayne 18,844 Kent 1,004 Wexford 127 Keweenaw 17 Source: Computed from the Annual Re 9; Health, Ens ng, e PP. De rtment Taege IE rts out the Mich 18811, 44. V???) - 355 Table 11 TOTAL NUMBER or DEATHS FROM NBPHRITIS, BY COUNTY, MICHIGAN, 1935-1944 State and Total Number State and Total Number County of Deaths County of Deaths State 28,707 Alcona 46 Lake 33 Alger 68 Lapeer 198 Alle gan 868 Leelanau 39 Alpena 110 Lenawee 407 Antrim 89 Livingston 104 Arenae 34 Luce 33 Baraga 34 Mackinac 47 Barry 807 Maconb 374 Bay 389 Manistee 87 Benzie 66 Marquette 230 Berrien 681 Mason 184 Branch 888 Mecosta 167 Calhoun 368 Menominee 182 Case 888 Mid land 293 Charlevoix 98 Missaukee 33 Cheboygan 64 Monroe 848 Chippewa 138 Montcalm 238 Clare 73 Montmorency 34 Clinton 344 Muskegon 498 Crawford 32 Newaygo 132 Delta 180 Oakland 1,083 Dickinson 93 Oceana 80 Eaton 278 Ogemaw 43 Emmet 83 Ontonagon 63 Genesee 1,053 Osceola 99 Gladwin 69 Oscoda 21 Gogebic 184 Otsego 30 Grand Traverse 118 Ottawa 337 Gratiot 240 Presque Isle 31 Hillsdale 218 Roscommon 84 Houghton 308 Saginaw 778 Huron 196 Sanilac 189 Ingham 774 Schoolcraft 37 Ionia 883 Shiawassee 343 Iosco 37 St. Clair 644 Iron 184 St. Joseph 323 Isabella 117 Il'uscola 849 Jackson 611 Van Buren 316 Kalamazoo 780 Washtenaw 379 Kalkaslna 23 Wayne 8 , 888 Kent 1,271 Wexford 133 Keweenaw 33 Source: Computed from the e artment g Heglt ale 9 PP. an %,ual Eenpgrts 0 3.1.1.9. c igan, m°§15.f§4,, ‘A WT: 356 Table 12 row. NUMBER or DEATHS mom mmcuosrs, BY com, MICHIGAN, 1935-1944 ' W State and Total Number State and Total Number iounty of Deaths County of Deaths Stats 18,838' Alcona 9 Lake 14 Alger 36 Lapeer 68 Allegan 79 Leelanau 29 Alpena 34 Lenawee 131 Antrim 88 Livingston 40 » Arenae 30 Luce 23 Baraga 32 Mackinac 31 Barry 28 Macmb 866 Bay 136 Mani stee 72 Benzie 88 Marquette 814 Berrien 833 Mason 78 Branch 48 Mecosta 33 Calhoun 833 Menominee 129 Case 30 Midland 32 Charlevoix 3O Missaukee 11 Cheboygan ‘ 33 Monroe 143 Chippewa 79 Montcalm 38 Clare 80 Montmorency 9 Clinton 42 Muskegon 880 Crawford 8 Newaygo 34 Delta 139 Oakland 499 Dickinson 163 Oceana 87 Baton 32 Ogemaw 81 hmnet 36 Ontonagon 67 Genesee 684 Osceola 86 Gladwin 13 Oscoda 4 Gogebic 233 Otsego 10 Grand Traverse 73 Ottawa 96 Gratiot 38 Presque Isle 43 Hillsdale _ 44 Roscommon 10 Houghton 331 Saginaw 374 Huron 71 Sanilac 70 Imam 196 Sch oolcraft . 33 Ionia 37 Shiawassee 69 Iosco 20 St. Clair 190 Iron 89 St. Joseph 70 Isabella 69 Tuscola 80 Jackson 189 Van Buren 81 Kalamazoo 191 Washtenaw 808 Kalkaska 13 Wayne 10 , 664 Kent 323 Wexford 33 Keweenaw 83 Source: Computed from the ual Re arts of the Mich an 33%?tment p; Hegfié'finsflng, Michigan, 44, ‘ 0 6: PP“ 35'? ("w I! I: Table 13 TOTAL mm 01' DEATHS IROM PREMATUBI BIBBS, BY COUNTY, MICHIGAN, 1933-1944 State and Total Number State and Total Numer County of Deaths County of Deaths State 14,196 Alcona 19 Lake 3 Alger 48 Lapeer 32 Allegan 94 Leelanau 24 Alpena 36 Lenawee 163 Antrim 23 Livingston 63 Arenae 87 Luce 30 Baraga 87 Mackinac 37 Barry 33 Macomb 869 Bay 862 Menistee 37 Benzie 23 Marquette 135 Berrien 838 Mason 37 Branch 76 Meco sta 46 Calhoun 326 Menominee 88 Cass 30 Midland 76 Charlevoix 49 Mi ssaukee 36 Cheboygan 33 Monroe 178 Chippewa 104 Montcalm 87 Clare 28 Montmoren cy . 14 Clinton 69 Muskegon 391 Crawford 14 Newaygo 34 Delta 107 Oakland 787 Dickinson 77 Oceana 81 Baton 76 Ogemaw 38 me 31 Ontonagon 36 Genesee . 766 Osceola 32 Gladwin 86 Geo oda 6 Gogebic 87 Otsego 13 Grand Traverse 76 Ottawa 119 Gratiot 97 Presque Isle 33 Hillsdale 36 Res common 3 Houghton 98 Saginaw 383 Huron 104 Sanil so 66 Ingham 389 Schoolcraft 28 Ionia 7O Shiawassee 187 Iosco 86 St. Clair 207 Iron 46 St. Joseph 64 Isabella 79 Tuscola 86 Jackson 239 Van Buren 79 Kalamazoo 832 Wamtenaw 206 Kalkaska 16 lane 3,066 Kent 323 Wexford 65 Keweenaw 9 Source: Computed from me u R to of 1h; chi an ga'fins ng, Eigan, 44 , De ai‘tm6 t 1 Egg; , TH . , , 358 Table 14 Tom. NUMBER or moss non ensues, BY scum, MICHIGAN, 1935-1944 * _; ‘— _‘ ____—: State and Total Number State and Total Number County of Deaths County of Deaths State 13,719 Alcona 18 Lake 18 Alger 23 Lapeer 110 Allegan 143 Leelanau: 88 Alpena 63 Lenawee 198 Antrim 23 Livingston 30 Arenae 83 Luce 10 Baraga l4 Mackinac 23 Barry 71 Maccmb 268 Bay 289 Manistee 74 Dennis 84 Marquette 137 Berrien 283 Mason 62 Branch 98 Mecosta 74 Calhoun 238 Menominee 92 Case 76 Midland 30 Charlevoix 37 Mi ss aukee 84 Cheboygan 36 Monroe 134 Chippewa 66 Mont calm 89 Clare 33 Montmorency 9 Clinton 86 Muskegon 261 Crawford 13 Newaygo 37 Delta 103 Oakland 470 Dickinson 87 Oceana 33 Baton 99 Ogemaw 83 trust 44 Ontonagon 83 Genesee 408 Osceola 41 Gladwin 20 Oscoda 4 Gogebic 78 Otsego . 6 Grand Traverse 76 ' Ottawa 173 Gratiot 113 Prosque Isle 83 Hillsdale 94 Ros semen ’ 8 Houghton 190 Saginaw 370 Huron 78 Sanilac 80 Ingham 327 Schoolcraft 81 Ionia 113 Shiawassee 130 Iosco 13 St. Clair 287 Iron 38 St. Joseph 120 Isabella 39 Tuscola 103 Jackson 290 Van Buren 101 Kalamazoo 230 Washtenaw 196 Kalkaska 6 Wayne 4,766 Kent 713 Nexford 63 Keweenaw 18 Source: Computed from the ual Re orts of the Mich an Department g H a1 , 8 ng, m‘éfiigan, 3 944. 8b 0 6. p e 359 Table 15 TOTAL m or DEATHS FROM ARTERIOSCLEROSIS, BY COUNTY , MICHIGAN, 1933-1944 State and Total Number State and Total Number County of Deaths County of Deaths State 9,783 Alcona 10 Lake 4 Alger 13 Lapeer 81 Allegan 124 Leelanau 10 Alpena 60 Lenawee 216 Antrim 38 Livingston 113 Arenae 10 Luce 3 Baraga 13 Mackinac 14 Barry 89 Macomb 98 Bay 104 Manistee 76 Benzie 41 Marquette 77 Berrien 179 Mason 71 Branch 73 Mecosta 97 Calhoun 839 Menominee 48 Cass 59 Mudland 27 Charlevoix 86 Missaukee 6 Cheboygan 30 Monroe 98 Chippewa 37 Montcalm. 133 Clare 86 Montmorency 3 Clinton 113 Muskegon 148 Crawford 8 Newaygo 30 Delta 34 Oakland 418 Dickinson 38 Oceans 84 Baton 134 Ogemaw 14 hmet 44 Ontonagon 13 Genesee 306 Osceola ‘ 48 Gladwin 18 Oscoda 4 Gogebic 43 Otsego _ 7 Grand Traverse 38 Ottawa . 146 Gratiot 187 Presque Isle 18 Hillsdale 174 Roscommon 6 Houghton 7O Saginaw 318 Huron 99 Sanilac 36 Ingham 813 Schoolcraft 7 Ionia 174 Shiawassee 174 Iosco 32 St. Clair 184 Iron 13 St. Joseph 163 Isabella 108 Tuscola 69 Jackson 303 ‘Van Buren. 107 Kalamazoo 830 Wash tenaw 329 Kalkaska 18 Wayne 1,730 Kent 688 Wexford 87 Keweenaw 18 Source: Computed from the us Reo of the an gggartmsnt ggne _Lfiffinfifsfiig‘fi,m 44. 8. 0 pp. 360 Table 16 TOTAL NUMBER or DRATHS FROM SUICIDBS, BY CGINTY, MICHIGAN, 1953-1944 State and Total Nunber State and Total Number County of Deaths County of Deaths Stats 6,617 Alcona 18 Lake 10 Alger 14 Lapeer 36 Allegan 33 Leelanau 10 Alpena 24 Lenawee 118 Antrim 10 Livingston . as Arenae 18 Luce 7 Baraga 83 Mackinac 18 Barry 48 Macomb 138 Bay 79 Manistee 16 Benzie 9 Marquette 82 Barrien 163 Mason 89 Branch 48 Mecosta 30 Calhoun 188 Menominee 88 Case 49 Midland 87 Charlevoix 22 Missaukee 8 Cheboygan 13 Monroe 83 Chippewa 41 Montcalm 3O Clare . l2 Montmorency 3 Clinton 44 Muskegon 93 Crawford 7 Newaygo 28 Delta 43 Oakland 279 Diekinscn 38 Oceans 27 Baton 71 Ogemaw 10 Met 13 Ontonagon 83 Genesee 232 Osceola 18 Gladwin 13 Oscoda 3 Gogebic 36 Otsego 7 Grand Traverse 30 Ottawa 37 Gratiot 46 Presque Isle 11 Hillsdale 6O Roscommon 6 Houghton 80 Saginaw 134 Huron 37 Sanilac 58 Ingham 139 Schoolcraft 11 Ionia 32 Shiawassee 87 Iosco 10 St. Clair 93 Iron 39 St. Joseph 48 Isabella 27 Tuscola 48 Jackson 143 Van Buren 73 Kalamazoo 144 Uashtenaw 128 Kalkaska 1 Iayne 2,814 Kent 839 Wexford 83 . Keweenaw 10 Source: Computed from the Annual Reports of the Mich an Department 3 Hgal—Enth, sing, 1551851533, - 44, ab e 6, pp. 361 Table 17 TOTAL NUMBER OF DEATHS J'BOM SYPHILIS AND GONOCOCCUS, BY COUNTY, MICHIGAN, 1933-1944 State and Total Number State and Total Number County of Deaths County of Deaths Stats 4,832 Alcona 3 Lake 10 Alger 8 Lapeer 10 Allegan 17 Leelanau 1 Alpena 11 Lenawee 24 Antrim 8 Livingston 13 Arenae 4 Luce 3 Baraga 6 Mackinac 6 Barry 10 Macomb 73 Bay 66 Manistee l3 Benzie 3 Marquette 23 Berrien 83 Mason 23 Branch 80 Mecosta 16 Calhoun 123 Menominee lO Cass 13 Midland 18 Charlevoix 13 Missaukee -- Cheboygan 8 Monroe 41 Chippewa 88 Montcalm 23 Clare 3 Montmorency 2 Clinton 9 Muskegon 93 Crawford 4 Newaygo 13 Delta 16 Oakland 183 Dickinson 8 Oceana 10 Baton 28 Ogemaw 3 Emmet 9 Ontonagon 8 Genesee 180 Osceola . 1 Gladwin 8 Oscoda 8 Gogebic 84 Otsego 3 Grand Traverse 80 Ottawa 17 Gratiot 83 Presque Isle 2 Hillsdale 13 Roscommon 3 Houghton 31 Saginaw 144 Huron 7 Sanilac 13 Ingham 112 Schoolcraft 6 Ionia l9 Shiawassee 26 Iosco 8 St. Clair 47 Iron 14 St. Joseph 84 Isabella 18 Tuscola . 18 Jackson 89 Van Buren 88 Kalamazoo 80 Washtenaw , 3O Kalkaska 3 layne 2, 491 Kent 192 Wexford _ 10 Keweenaw 2 Source: Computed from the ual Re orts of the Michi an De tment 9; 39:5, Lansing, Mi'c'Higan,'T535§N44, a e 6, T%—PD e 362 Table 18 NUMBER or URBAN AND RURAL DEATHS mom ALL omens PER 1,000 POPULATION, MICHIGAN COUNTIES, 1935- 1944 (By place of residence) State :Number :Number : State :Number :Number and : Rural : Urban : and : Rural : Urban County :Deanis : Deaths : County :Deaths :Deaths Alcona 636 --- Lake 640 --- ‘Alger 631 487 Lapeer 2,928 638 Allegan 4,048 1,098 Leelana , 949 --- Alpena 780 1,440 Lenawee 5,047 1,976 Antrim 1,401 --- Livingston 2,066 615 Arenae 1,006 --- Lucel 761 «- Baragal 1,030 --- Mackintcl 1,041 -- Barry 8 , 803 776 Macomb 5 , 806 4 , 283 Bay 2,528 3,699 Manistee 1,249 1,217 Benzie 1,073 --- Marquette 1,490 3,693 Berrien 3,346 4,969 Mason 1,818 1,138 Branch 2,311 1,238 Mbcosta 1,524 763 Calhoun 3, 783 7 ,747 Menominee 1, 358 1, 389 Case 8, 359 883 Midland l ,404 837 Charlevoix 1 , 356 374 Mi ssaukee 849 --- Cheboygan 860 , 806 Monroe 3,893 8,051 Chippewa 1 , 300 1 , 816 Montcalm 3 , 254 741 Clare 1 , 19 3 --- Montmcreney 408 ,--- Clint on 8 ,439 642 Musics gen 3, 088 6 , 620 Crawford - 438 , --- Newaygol 2 , 290 _ --- Delta 1,317 2,261 Oakland 9,160 11,584 Dickinson 738 1,967 Oceana 2 .025 -... Baton 8,382 1,892 Ogemaw 1,016 -- 18mm t 1 , 228 834 Out onagon 1 , 288 --- Genesee 3,903 13,438 Osceola 1,777 «- Gladwin 1,000 --- Oscoda 287 «- Gogebic 1 , 887 2 , 137 Ct sego 646 -- Grand Traverse 1,135 1,361 Ottawa 3,129 8,644 Gratiotl 2,997 924 Presque 1.1. see 271 Hillsdale 3,096 902 Roscommon 375 --- Hough on 4,101 1,846 Saginaw 4,464 9,117 Huron 3,326 -- St. Clair 4,062 5,101 Ingham 4,153 8,532 St. Joseph 2,603 1,676 Ionia 3,027 1,448 Sanilac 3,390 --- Iosco 939 ~-- Schoolcraft 397 ,637 Iron 1,070 777 Shiawassee 2,838 2,132 Isabella 1,848 889 Tuscola 3,745 573 Jackson 4,053 6,491 Van Buren 4,369 737 Kalamazoo 4 , 283 6 , 531 Washtinaw 4 , 033 4 , 721 Kalkaska 585 --- Wayne 3,979 173,029 Kent 7 , 191 18 , 502 Wexford 1 , 043 1 , 213 Keweenaw 488 --- State 191,833 338,449 1. Residence according toith: 193g pepulation; H alth 1935 Source: Annual Repgts, M ch gan spar men 6 e , - 1944’ a 1. O I1 I" I‘. I‘D (I 1' 1‘ I. 1‘ a“ 363 TablaIHD RURAL DEATHS FROM THE FIVE PRINCIPAL CAUSES OI DEATH PER 10, 000 POPULATION, MICHIGAN COUNTIES, AVERAGE 1935-1944 (Based on 1940 pOpulation; by place of residence.) State NUmbg£_g§_§g£al Deaths and 1Hea1t Cancer ,A Oplexy Accidents Nephritis County 1(90-95) (45-35) 83ab) (169-195) (130-132) .Alcona 192 61 7O 59 46 ,Alger 144 49 36 86 33 Allegan 1,266 447 442 344 213 Alpena 205 74 74 71 28 Antrim 409 153 161 83 89 Arenacz 305 97 91 103 34 Baraga 217 85 86 126 34 Barry 653 235 272 173 163 Bay 672 286 192 255 287 Benzie 309 99 92 97 66 Berrien 1,654 597 490 450 393 Branch 774 236 291 , 184 155 Calhoun 1,164 328 480 292 153 Cass 633 260 282 209 284 Charlevoix 434 . 153 126 107 74 Cheboygan 267 83 80 74 25 Chippewa 406 103 106 115 38 Clare 353 109 105 129 73 Clinton 655 248 242 . 186 867 Crawford 116 57 23 43 32 Delta 415 166 148 166 70 Dickinson 227 81 80 71 23 Eaton 756 305 241 805 143 Emmet 415 137 130 97 54 Genesee 1,609 380 302 524 310 Gladwin 277 109 103 92 69 Gogebic 306 118 108 166 58 Grand T§averse 335 102 129 92 57 Gratiot 881 332 271 229 174 Hillsdale 869 359 384 238 166 Houghgon 1 , 114 501 414 892 884 Huron 915 409 338 309 196 Ingham. ' 1,131 488 471 389 811 Ionia 924 362 313 810 164 Iosco 288 183 101 78 57 Iron 874 137 81 189 78 Isabella 503 203 188 160 76 Jackson 1,220 448 408 381 866 Kalamazoo 1,227 458 438 378 878 Kalkaska 176 57 73 4O 25 Kent 8,309 857 724 522 357 Keweenaw 114 49 63 34 33 364 T8316 19( Continued) RURAL DEATHS FROM THE FIVE PRINCIPAL CAUSES OF DEATH PER 10, 000 POPULATION, MICHIGAN COUNTIES, AVERAGE 1935-1944 SBased on 1940 popular tion; by place of residence. 1% State m of Rural Deaths and 1 sort ancer A 6p exy cc dents Nep r County (co-95) (45-55) 83ab) (169-195) (loo-132) Lake 182 66 73 38 33 Lapeer 846 303 355 252 163 Leelana a5 318 104 98 67 59 Lenawee 1,413 324 361 411 274 uggston 691 241 192 168 83 Luce 886 51 46 68 33 Mackingca 279 97 100 114 47 ,Macomb 1,317 386 527 493 336 Nhnistee 415 144 132 98 49 Marquette 360 170 117 185 61 Mason 305 138 144 94 61 Mecosta 417 143 183 111 103 Menominee 406 174 123 . 100 43 thlend 373 147 84 188 185 NHssaukee 847 68 102 58 35 Mbnroe 1,190 436 375 418 172 Mbntcalm 900 339 360 291 190 Montmorency 104 46 23 44 34 MUskegog 784 309 856 244 136 Newaygo 726 237 838 q 133 138 Oakland 2,277 1,018 787 898 480 Oceana 639 817 836 165 80 Ogemaw 278 84 150 93 43 Ontonagon 310 126 151 123 63 Oseeola 318 198 811 145 99 Oscoda 68 88 18 , SO 81 Otsego 807 67 42 58 50 Ottawa 950 442 338 . 809 178 . Presque Isle 819 96 38 63 38 Roscommon 112 40 88 36 84 Saginaw 1,839 507 401 411 836 St. Clair 1,181 468 413 _ 403 289 St. Joseph 744 890 894 186 188 Sanilac 1,010 364 430 274 189 Schoolcraft 107 38 44 43 84 Shiawassee 768 336 273 820 288 Tuscola 1,158 417 381 897 286 Van Buren 1,379 538 495 354 264 Nashtgnaw 1,054 479 434 a 361 178 Wayne 1,610 590 362 718 812 Wezford 380 98 108 67 74 Stats 34,698 20,909 19,154 A 16,534 10,861 1. International List Number, 1939. 2. Residence acccrding to the 1930 pOpulation. Source: Annual Re orts Michi an De artment of Health 1935- "mrz 11P——’ g p ’ 104‘. 365 Table 20 URBAN DEATHS PROM THE FIVE PRINCIPAL CAUSES OF DEATH PER 10,000 POPULATION, MICHIGAN COUNTIES, ANERAGE 1935-1944 (Based on 1940 pOpulation; by place of residence.) State Number of Urban Deaths and 1 e ancer .4 Op exy Acci ants Nephritis County (90-93) (43-55) 83ab) (169-195) (130-132) .Aleona --- --- ' --- --- --- Alger 117 51 89 32 83 Allegan 346 185 114 84 53 ‘Alpena 488 136 143 98 88 Antrim. --- --- --- --- --- Arenae --- --- --- --- --- Baragaz --- --- --- --- --- Barry 202 79 110 63 44 Bay 1,607 696 496 398 368 Benzie --- --- --- f --- -- Berrien 1,408 561 448 485 888 Branch 373 123 143 101 73 Calhoun 2,882 868 819 554 415 Cass 881 114 77 74 64 Charlevoix 93 51 48 27 84 Cheboygan 216 113 88 64 39 Chippewa 588 178 141 154 86 01‘" --- --- --.- --- --- Clinton 161 65 62 4O 77 Crawford --- --- --- --- --- Delta 582 273 830 184 110 Dickinson 553 836 151 167 70 Eaton 615 820 201 150 135 Emmet 272 86 98 58 31 Genesee 3,323 1,472 990 1,068 743 Gladwin --- --- --- --- --- Gogebic 513 244 828 178 126 Grand TEaverse 436 169 800 108 81 Gratiot 269 101 77 63 66 Hillsdale 232 91 87 72 58 Hough on 333 238 194 115 78 Huron ; --- --- --- --- ... Ingham. 8,871 1,037 891 641 563 Ionia 398 143 154 95 39 10300 --- --- --- _D-n --- Iron 804 101 39 ~64 46 Isabella 863 89 68 77 41 Jackson 1,988 755 600 _ 486 345 Kelamasoo 1,814 760 653 308 448 KamalkO --- --- --- --- --- Kent 6,080 8,458 1,756 1,270 914 KO'OOiial """' --- --. seq-es --- 366 Table 20 (Continued) URBAN DEATHS FEW! THE FIVE PRINCIPAL CAUSES OF DEATH PER 10,000 POPULATION, MICHIGAN COUNTIES, AVERAGE 1935-1944 (Based on 1940 popu- lation; by place or residence.) State Number of Urban Deaths and 1 eart ancer A op exy .Accidents NEEEEIFIs County (90-95) (45-55) f83ab) (169-195) (130-132) lake --- --- --- --- --- Lapeer 150 79 66 47 33 “918118 --- --- --- --- --- Lenawee 509 211 191 157 133 Liviggston 174 60 40 67 21 1.1100 . OI.- -.- --— can. an... Mackingcz --- --- --- --- --- Macomb 1,069 486 398 369 838 anistee 408 161 162 67 38 Marquette 1,038 431 321 868 189 Mason 371 118 83 92 63 Mecosta 234 66 63 32 64 Menominee 360 163 137 93 79 Midland 173 92 53 69 110 Missaukee --- --- --- --- --- Monroe 537 231 130 198 76 Montcahm 241 63 76 46 42 Montmorenoy --- --- --- .--- --- Muskegog 1,786 692 544 413 336 Newaygo --- .--- --- --- --- Oakland 8,893 1,317 903 936 603 Oogana --- --- nun- --- --- Ogemaw -- --- --— --- --- Ontonagon --- --- --- --- --- Osceola --- --- --- --- --- Oscoda --- --- --- --- --- Otsego --- --- --- --- --- Ottawa 798 359 853 168 159 Presque Isle 63 31 12 24 13 Roscommon --- --- --- --- --- Saginaw 8,616 976 768 649 536 St. Clair 1,434 591 477 396 353 St. Joseph 456 188 190 180 135 Sanilac --- --- --- --- --- Schoolcraft 200 87 48 53 33 Shiawassee 552 859 807 175 123 Tuscola 179 71 58 44 83 Van Buren 248 87 81 58 58 Waahtgnaw 1,481 587 418 324 801 Wayne 48,491 80,672 10,960 10,914 8,610 Wexrord 386 112 106 93 61 State 95,286 39,811 26,116 23,338 17,846 1. International List Number, 1939. p 2. Residence according to the 1930 population. Source: .Annual Reports, MHchigan Department of Health, 1935- * .ICIAA Mnk'ln 1R- 367 Table 2]. TOTAL NUMBER OF INFANT DEATHS IN COUNTIES OF MICHIGAN, 1935-1944 (Exclusive of stillbirths. By place or residence) State and : Number : State and : Number County :of Deaths: County :0: Deaths Alcona 38 Lake 40 Alger 122 Lapeer 858 Allegan 320 Leelanau 90 Alpena 194 Lenawee 501 Antrim. 118 Livingston 154 Arenae 81 Luce 102 Baraga 96 Mackinac 108 Barry 140 Macomb 1,019 Bay 765 manistee 155 Benzie 93 marquette 486 Berrien 688 mason 133 Branch 804 mecosta 158 Calhoun 896 menominee 286 Cass 179 Midland 300 Charlevoix 148 Missaukee 106 Cheboygan 151 Monroe 589 Chippewa 367 Mbntcalm 890 Clare 100 Montmorency 48 Clinton 207 Muskegon 1,163 Crawford 52 Newaygo 176 Delta 411 0ak1and 2,255 Dickinson 833 Oceana. 188 Eaton 246 Ogemaw 97 Emmet 167 Ontonagon 180 Genesee 8,151 Osceola 132 Gladwin 99 Oscoda 24 Gogebic 274 Otsego 64 Grand Traverse 803 Ottawa 371 Gratiot 353 Presque Isle 117 Hillsdale 810 Roscommon 33 Houghton 396 Saginaw 1,885 Huron 347 St. Clair 727 Ingham. 1,158 St. Joseph 206 Ionia 276 Sanilac 863 Iosco 52 Schoolcraft 112 Iron 130 Shiawassee 386 Isabella 504 Tuscola 330 Jackson 757 Van Buren 232 Kalamazoo 655 Washtenaw 622 Kalkaska 43 Wayne 14,499 Kent 1,604 Wexford 200 Keweenaw 35 State 43,818 Source: Annu Reports or the Michigan Department or Health, 9 19“ , Table 8 e I. 19. £0. 81. 88. 83. 84. 85. 86. 87.6 83c 8’s 368 APTENDIIIIII INTERRAIIOEAL LIST OF‘CAUSES OF DEATH; (rho fifth revision of the detailed International List) Infectious and Parasitic Diseases Iyphoid fever Paratyphoid fever Plague Cholera Uhdulant fever (brusellosis) Cerebros inal (meningococeue) aeningitis ,.Anthrax infection.by bacillus anthraeis) Scarlet fever UheOpinc'eeugh Diphtheria (infection by C.diphtheriae) Er sipelas 2e anus tuberculosis (all ferns) Tuberculosis of the respiratory systen(inc1udin¢ the bronchial and aediastinal 1ynph.nodes) tuberculosis of‘the meninges and central nervous syste- fuberculcsis cf the intestines and peritoneu- Iuberculcsis of the vertebral column tuberculosis of the bones and Joints(except vertebral colunn . Tuberculosis of the stin.and subcutaneous cellular tissue tuberculosis of the 1yaphatie systen(exeept bronchial, aegiaetinal, nesentcric, and retreperiteneal lynph no es tuberculosis of the (shite-urinary systen Tuberculosis of other organs Disseainated tuberculosis Leprosy Septicenia and purulent infection (nonpuerperal) Conccoccus infection Other diseases due to bacteria (except dysentery) Dysentery Malaria Other diseases due to parasitic protosea (except spirochetes) 1. this list, revised in.1938, is currently in use. Since the list Ias adopted for-ally in.1893, it has had five revisions as follows; 1900, 1909, 1980, 1989, and 1938. ‘0“ v-1.~‘ -- . .. _ . _. , A, - ‘ i 7 . . . - e I U ‘I ' N ‘ w ‘ v 0— ~ .- - -. _ - r, 7 u. D . .- i O . ._ . - _ v - . ‘ e . . O ' - d . A u§ . .‘ . . . ; - _ . g . _ e . , '~ - \- . , . - r . .-. \- . . . C ‘ > e ' ~ . _ _ . . . A \9 .. w .o- - -. r ‘ ‘- . .~ I . - - . . ' . . l . 7 , , V ,n I , . . ‘ . . ‘ .7 -~ No V , . . . e I , ‘ . I . p v . *4 .. .- a- _ ‘. ‘ ‘ ,2 U . . . . _. 6' . _ . s .... ' A w .- - p- ..r - .. —' . _ . , , . . _ A. . ' ‘ . .\ . r . . - - . v .7 .. -- , . ' . ~ .. : V : I. _. ‘ 4 . a ., . . .‘ ‘ .1 -- . .-. , . w- ‘ . ~ . . l v . . .- ' . ' ' . . a ‘ x *, _ , s ... , 4 O s T J A - c - - .. ~ .I. ' v _. ‘ - J. - .' ‘ ‘ . ... - . ... H . .. . ' , . . - - .1 r e b - a u . ~ ‘ '. '7 . - - .4- ._ _ . O r . , 1. . ~ ~\. . ‘~ . ' . ‘ . 6 ‘ a . § . .- u __ A ' . ‘ « - I O .' _ - ,. , A ... . . O . ’ O I . . I l ' a . . ' ., . - A _ . O n A . 6 . .. n. .- ‘o-.--‘I “.Ac- -..-._‘- a-b -p‘.‘ “.34-- - . - - y, ‘ . _ . . .... . -. . .- ~ _A » — ' ... A 7 " V ~ .v '6 a . A e . k . . on. v .. ' ‘ ‘ '~ -.. v . ~ 569 I. Infectious and Parasitic Diseases-con. 30. 30a. 30b. 30d. 30c,e-g. 32. 333a 33b. 34. 35. 36. 37. 33. 39. 40. 41. 42. 43c 44. II. ‘5'55e 45. 458s 45b. 450e 45d. 450e 45!. 45. 461e 46b. 46°e 46d. 460. 46:. 468s 46h. 46m. 47. 47‘s 47d... 47b,0,fe Syphilis (all forms) Locomotor ataxia (tabes dorsalis) General paralysis of the insane Aneurysm.of the aorta Other forms of syphilis Relapsing fever Other diseases due to spirochetes Influenza, respiratory complications specified Influenza, respiratory complications not specified Smallpox Measles Acute poliomyelitis and acute olioencephalitis Acute infectious encephalitis lethargic) Other diseases due to filtrable viruses Typhus fever, typhus-like diseases (due to Rickettsia) Ankylostomiasis Hydatid disease Other diseases caused'by helminths Mycoses Other infectious and parasitic diseases Cancer and other tumors Cancer and other malignant tumors Cancer of the buccal cavity and pharynx Cancer of lip Cancer of tongue Cancer of mouth Cancer of Jaw bone Cancer of unspecified parts of the buccal cavity Cancer of pharynx Cancer of the digestive organs and peritoneum Cancer of seephagus Cancer of stomach Cancer of duodenmm Cancer of rectum.and anus Cancer of intestines (except duodenum.and rectum) Cancer of liver and biliary passages Cancer of pancreas Cancer of mesentery and peritoneum Cancer of other and unspecified sites Cancer of the respiratory system Cancer of larynx Cancer of lung and pleura Cancer of other respiratory organs 370 II.-Canccr and other tuners cont'd. CB. 49. ‘9‘, be ‘90-.c 50. 51c 51‘. 61a,c-c. . 58‘s .3bo 530c 53. 5‘s 5‘. 55§-.o 561. C‘Co 57fio0o 57‘s 5'0o 58c, f.0thcr 59. 50c 61. 5'. 53‘. C38. m-Oo C‘e C‘o C‘c ‘7c ‘3. 59. 70c 71c of the uterus of other female genital organs of ovary,fallopian tube, and paranotriua of vagina,vulva,cthor and unspecified sites of the breast of the sale genital organs of prostate of other male genital organs of the urinary organs (nalc and fcnalc) of kidney of bladder of other and unspecified sites Cancer Cancer Cancer Cancer Cancer Cancer Cancer Cancer Cancer Cancer Cancer Cancer Cancer of the skin (except vulva and sorotun) Cancer f the brain and other arts of ocntr nervous system including “gliomawxccp when specific as benign). Cancer of other unspecified organs lonmalignant tuners of female genital organs nonmalignant tuners of the brain and other parts of the central nervous system llonnalignant tuners of other and unspecified organs rumors of female genital organa(naturc unspecified) rumors of the brain and other parts of central nervous syetcn(naturc unspecified) unspecified organs (nature unspeci- fanors of other fied) ti diseases of nutrition and of the endocrine g s, c or ‘encr ceases, an av 39sec Acute rheumatic fever Acute rheumatic pericarditis Acute rheumatic endocarditis Acute rheumatic avocarditis Other acute rheumatic heart diseases ferns of acute rhcunatie fever, including rheu- natisn(unqualificd) (Chronic rheumatisa and other rheumatic diseases u Diabetes mellitus Diseases of the pituitary gland Simple goiter hophthalnie goiter Other diseases of thyroid and parathyroid glands Diseases of the thymus gland Diseases of adrenal glands(not specified as tuberculous) Other general diseases Scurvy Beriberi Pellagra (except alcoholic) Riekets Other avitaninosos - ...” 371 IV. gigeases of the blood and bleod-forning_3_r5_a_ng 78o 73‘e 73b-d. 748. 748. 75. 76. Honorrhagic eonditi one Pernieious anenia Other anenias(exeept splenic) Leuhenias dleulcenias Diseases of the spleen Other diseases of the blood and blood-forming organs 7. Cgonic poisonigg and intoxication 77§e VI'OCe 73c 79. Alcoholic pellagra Other and unspecified alcoholisn,inol.aeute and chronic Lead poisoning Chronic poisoning by other mineral or organic substances '1. Eiseasec of tag Queue gates: and sense orggg 80. 81. 92. 83. 83a. 88b, 0 0 83d. 84. 35c K 85c 87. 88. 89. Encephalitis (nonepidcmic) leningitis(not due to ncni eceeeus Diseases of the spinal cord except disseminated sclerosis) Intracranial lesions of vascular origin Cerebral hemorrhage or effusion(cxcluding birth intaries) Cerebral embolism,thrombosis,and softening Heniplegia and other paralysis of unspecified origin Mental diseases and deficiency(except general paralysis of the insane) Epilepsy Convulsions (under 5 years of ago) Other diseases of the nervous system Diseases of the organs of vision Diseases of the ear and mastoid process econotor ataxia and VII. Diseases of the circulatog gates: 90-95. 90‘s 90].. ’1‘. b 0 .IOe 98.. 98b. 92c. 78d. Diseases of the heart (all ferns) Chronic rheumatic pericarditis Other pericarditis Bacterial, and other acute or subacute endocarditis lndocarditidnot specified as acute,ohronic, or rheunatie, under 46 years of age) Diseases of the aortic valve (without nenticn of diseases of the. nitral valve or rheumatic fever) Diseases of the nitral valve (whether or not specified as rheumatic) Diseases of other and unspecified valves and chronic endocarditis,specified as rheumatic Diseases of other and unspecified valves and chronic endocarditis,not specified as rheumatic ...- 0‘..- .- -.-~>- --O« . -- a - ...a— . .- .. -.a. a a .-e .- . l ' . V ..-~o ‘ n . C * ~ 9 - 4.... -o—-- 1 ,-‘ . u .- ...-.-..\ O i A *> . :Qnr-wq-tu J . , a. l -7 v - ' D .. .x-..-.a— 8 .... _ A a s - .- u-wn. Q“- 372 VIL-Discascs cf the circulatory system -cont'd. Diseases of the heart(a11 found-Continued 93‘s 93b. ’SCe ’3‘. ’3. e 94‘. 94b. 952. 95b. 85.. 96. 97. 98. ’9. 100. 101e 162. 103. Endocarditis(not specified as aeute,ehronic,er rheumatic, {45 years of age and over). Acute myocarditis (except rheumatic) Myocarditis (not specified as acute,chronie,or rheumatic, ‘under 46 years of age) Chronic myocarditis and.myocardial degeneration,speeified as rheumatic Chronic ayoccrditis and.nyecardia1 degeneration, not specified as rheumatic Other nyocarditic(net specified as acute,chronic,cr rheumatic.) Diseases of the coronary arteries Angina pectoris runetional diseases of the heart(without mention.of organic lesion) Other diseases of the heart,specified as rheumatic Other diseases of the heart,not specified as rheumatic Aneurysm (except of heart and aorta) Arteriosclcrosis(cxecpt coronary or renal sclerosis) Gangrene Other diseases of the arteries Diseases of the veins Diseases of the lymphatic system High blood pressure (idiopathic) Other diseases of the circulatory system VIII. Diseases of the regigtory system 10‘. 10.. 106. 10‘to 10‘be 1060c 1019 108. 109. 110. 111. 11.. 113e Diseases of the nasal fossae and accessory sinuses Diseases of the larynx Bronchitis (all forms) Acute bronchitis . Chronic bronchitis Bronchitis (unspecified) Dronohopneumenia(including capillary bronchitis) Lobar pneumonia . Pneumonia (unspecified) Pleurisy (not specified as tuberculous) Hemorrhagic infarction, thrombosis,edema,and chronic congestion of the lungs Asthma emphysema Pulmonary . 114a,b.Silicosis,other and unspecified forms of pneumoconioses lldcoe.Other and unspecified diseases of respiratory cystma, including gangrene and abscess-of lung(cxc.tubercu1esis) . .A ¢ _ ~. .— .. A . . e v . v I. . ... -,.-- .- -‘ a o v A a, O . a . . \ .--. q . >- a p. , . t . , n . . ‘- --.... s ~ -. a x . . . l, e-‘ .o I .4 u. l e u c O .. 115. 116. 117. 11.. 11'. 180. 181. 133‘. 1833. 133. 13‘. 185. 135. 187. 138. 189. ceases of the di stive 373 Diseases of buccal cavity,pharynx,tonsils,and adnexa Diseases of the esophagus Ulcer of stomach or duodenum Other diseases of the stomach(exccpt cancer) Diarrhea, cnteritis,and ulceration of the intestines (under 2 years of age) Diarrhea,enteritis,and ulceration of the intestines (2 years of age and over) , Appendicitis Hernia Intestinal obstruction Other diseases of the intestines Cirrhosis of the liver Other diseases of the liver biliary calculi Other diseases of the gallbladder and biliary duets Diseases of he ancreas cxc t abctcs l Peritenitis can e not a at.“ di ”1 itus) I. Diseggcs of the finite-w system 130. 1‘1.. 1315. 133. 133. 13‘. 135 . 136 . 1‘7 . 138 . 139.. 1395. 13.6. Acute nephritis Arteriosclerotie kidney Other chronic nephritis Nephritis unspecificd(10 years of age and over) Other diseases of kidneys and ureterc(cxcept diseases associated with pregnancy,childbirth, or puerperium) Calouli of the urinary passages Diseases of the urinary bladder ' Diseases of the urethra (except calculus) Diseases of the prostate . Diseases of other ale genital organs(not specified as venereal) Diseases of the ovaries,fal1epian tubes,and parametria Diseases of the uterus Diseases of other and unspecified female genital organs II.-Diseases of prewachildbir , and the pucnerium 1‘0. I‘ll... 1‘1‘. d’t O 1482. 148b. Abortion with mention of infection Abortion without mention of infection( spontaneous, therapeutic, or unspecified),with mention of toxemia, with or without hemorrhagc,trauma or shoot Other abortions without mention of infection letOpic gestation,with mention of infection Ectopic gestation,without mention of infection 163. 1“. 145. 1‘6. 1‘7. Hemorrhage of pregnancy (death before delivery) foxemias of pregnancy(death before delivery) Other diseases and accidents cf pregnancy(death before delivery) - Hemorrhage of childbirth and the puerperium Infection during childbirth and the puerperium , . \ . . ‘ - A a , _. m .J . . . s — . u- .. a \ x . 2,, l r—-s . q -. -. .. o a e . I . r ,j . . .- a v - A . . ' ‘ \ . . . - , z u . ‘ an- A... s . I a | t u. v. A/ -..-e- ... A ~ . a ,. ; r a i _ e .. gush. -« o-- v -.- - - -—~ ‘ ,- v‘ - .t n---‘ ... u e - . . a . O . v. \a ' O x . . . . . _ .. - ! — ~ ‘l I Q n . . - \ ‘. e '_ . .c » t « d . _. . , ~ A a e V' - \v _, ' - . ‘ ‘ . h u o a . - . a. . . 1 , . - I . . ~ . . .. - 7.. , , .-. w ,1 - - b . la ; p .4 -, , n .0 ~.. _ a . . . l n . A4. - - .4 . . n . x. _ . A ~. ... e . . ‘ ' ’ ' ; I r ‘ .’ . \- . ' ‘ . q -’. ' s * . o Is! -n— a ea 9 on --.. v‘ - .-. - m 9' av». ‘ gr~- -. -' I ' v ‘ I . i. .. . . ' - l e O .4 w. - . I . i ' ' a .. _. .. . . L A. t '_ . _ - a ‘1‘ n ‘ I . . . ‘ ‘ » u , “ . s ‘ . . / h. \. -. . a e . . e .. . L . , .. . \ '. ‘ a... . . . . c ' ' " v .4. . . . . . s ' ~‘ -. o ..4. ’ ¢ ‘ a . . A n ‘ o o m. .»— .. .. .. n ‘4 a s I-- vr- e -.-... . \ a n v v 4- -~ 0 . , . , .. V o I 'e -, . . 7 . 1 -7 ' . . ~ _ l . .. — ' . .... t ' A ‘ , . t ,. 4 ' h . . ' a i, v _. m , . . . . . - . . I ,. a; - .. ‘ .. . . . i - ‘ . 4 7.. , . . 1 . . - s ' ‘« - r . 4 . t e s ‘ I ~ . e .... an n I A u ' _ .1 . a . , . a i o _\ ‘ - _ - . -J a- ._ . . v‘ ' . 1 I. ~. ‘ 4.. Q r ‘J . I - o ...-u e e O I O I .. I .. s e e-‘-' O. .. .. e"- C e . .. e e I In 0 e s ¢ . 3'74 II.-Discases of prowlI childbirth and the p3 cgeriumfficn. lad. Puerperal toxemiastcxcludin; death before delivery) 1“. Other accidents and specified conditions of childbirth 150. Other and unspecified conditions of childbirth and the puerperium {IL-Diseases o; the skin and cellular tissue 151. Oarbuncle and music in. Daemon and acute abscess 158. Other diseases of the chin and cellular tissue Inn-Diseases of the bones and or ltd. Osteomyelitis and pcriostitis 155. Other diseases of bones(cxcept tuberculosis) 156. Diseases of Joints and other organs of movement mpom nital nlfcations lEVa-d. congenital malformations of central nervous system lflc,f. congenital malformations of cardiovascular system ll'l‘,h,m. Other congenital malformations of movement. IV.-Discases peculiar to the first year of life 158. congenital debility (cause not stated) 169. Premature birth(causc not stated) 160. Iniury at birth 161. Other diseases peculiar to first year of life 163. Inn-Senility min-Violent or accidental deaths 168,164. Suicide 163. Suicide by poisoning 11“. Suicide by other means loo-168. Homicide led. Infanticiddhomtoide of infants under 1 year of age) 166. Homicide by firearms , 167. Homicide by cutting or piercing instruments 168. Homicide by other means 169-lid. Accidental deaths 169. Railway socidcntstercmcllisions with motor vehicles) 170. Iotor-vehiclc accidents. 170a. Oollisions between automobiles and trains l‘IOb. Collisions between automobiles and street ears 170c. Automobile accidentstcxoept collisions with trains or street cars) l?Od.- Hotcreyclc accidents(cxccpt collisions with automobiles) 171a. Streetcar accidents(except collisions with trains or motor vehicles) l?lb. Other and unspecified roadotranspcrt accidents . ‘ _ _ . », . . - .A. , . ‘ . ‘ . , A , -' ‘ ~ ‘ ' . ‘ u-w . ..o n - a —- 4- ..«~ A\&~. 'n’ .._ - n. --.-.... “...—.3, e ..-..- a» g..- .e -»_‘ ... -.. I s I . A , ,. . , . . . . c . - I e ' r I . a ‘ 7 , ‘ . ’ A . , . . -. .4 -I w A. ‘ s . d' . , ' ~0 - h \ I A... 4e ‘ o I-- ...—W .n ,- v 9 W“.--. ... .C- Q—nm ...-M—r-M ... .. ~-~—.m.. Huu‘a... ...A‘ ..., ._ .._ _ .... . A . - . ‘ ‘ m J ._ . _. . . - A - O ‘ . A -- ' ' | e . . _ p p L , , . .7 . . .7 c c -' ‘ m _ ' t e . . . ‘ . a . . v - - ' " , ' V - ‘. '~ . ‘ s . ‘ I ‘ , . ' . m V, 0 no. .‘u o« .... s» s - p -a A s ..- v”- w ’7‘ , -._. ’ . — a. - . . ’ ‘ , A ' -. ‘ _ . y _ , , v. . . c '— - , A . l . ‘ ._ , . _ _ p 4. _ | ‘ -. . - - . p _A . ' f 1 x . . , a —_ ~.e . I ' - . ‘ h ' ' ' ' ' . . i . ' . _ > , . . . , _ ' _ V ' . ‘ ' .-e. A . ~ e ‘ A A ' ‘ v - __ o. ~ . A. - . . A .~ ,V e - . c _ - g m e - ..a~e A. ' | ’ . ‘ ' 1 H . . - , . ‘— - l‘-.-~-~--- -.-. . o l- g -.r- -.‘ ---.. 5 - ..~ I. .4- .; ¢. - .._. f' s . _ ‘ . V _. ~ .. O . ' a .- r .. .. . .. Q ~ ._ . x I t , V ‘ - . _ . A 7 , . _ .m, . . c . . ... g. n . A. a ‘ . ‘ - ‘ _‘ . - - .4 . . — . e l I. n - c ~ 4 —. v. . , ~’ ' ' ' .s- . - . '~ - . - ‘ . ... .. 'n. O - . -. . » . . . . .. e ' . "' . e . . - ' s ' - s. a. . . i .‘ I ' . . . u . . ~‘ _ . s .. , - _ § A . . . A . § . _ , _ , ‘ . . . . . _ t -4 . ‘ ' . ‘ ' _ a ~ 1 ' -.. A .. - . - . ~ H _ . t‘ I - . . I i . 7 .. . e - , ' c . 3 ‘ . . - \ - .. A 375 XVII.-Violent or accidental deaths-Con; 173. 173. 17‘. 1,5‘90e 1'5d..c 17‘c 177s 178e 179s 180. 181. late 183c 134s 185s la‘lc 1865c 1.7a laac lB’e 190e 191e I’Be 1’3e 1"e 195a. lfifiho 195‘s 1’60... 19‘e 1'7c 193c Accidental deaths tater-transport accidents Air-transport accidents Accidents in mines and quarries Agricultural accidents Forestry accidents Other accidents involving machinery lbod poisoning Accidental absorption of’poisonous gas Acute accidental poisoning by solids or liquids Conflagration AAccidental burns(cxcept eonflagration) Accidental mechanical suffocation .Accidental drowni Accidental injury y firearms Accidental injury by cutting or piercing instruments Accidental inaury by fall Accidental inaury by e OataclysmIall deaths attribu cd to a cataclysm,regard- less of their nature) Incury by animals(not specified as venomous or occurring in the course of agricultural and forestry operations) Hunger or thirst Execssive cold Excessive heat Lightning Accidents due to electric currents(cxcept lightning) Poisoning by venomous animals(not occurring in.the course of agricultural and forestry operations) Sequelae of preventive immunisation,ineeulation.or vaccination. Other accidents due to medical or surgical intervention. Obstruction,suffocation.or puncture by ingested objects Other and unspecified accidents Deaths of’military personnel during operations of‘war Deaths of civilians due to operations of”war Legal executions MIL-Ill-dcfincd and unknown causes 199. 800‘s ”Ob... Sudden death Ill-defined causes ‘Unhncwn or unspecified causes b ' . x . \. .a ‘ . , -L. - _ ,- .» . - . ‘ s v . ' .. m .. ' e m . J . ' I > » ' ‘0‘ s I v ‘ . ’ .- t » . .s. u . . ‘ s . " .. .... so»- 0 c u v . e _ v t e . . , . , n ' . -’ J. ' \- .s . ., . . . « . < - n H. _ v C‘ y . 1 ~ . .A m l. - . r \, . vac—valu. c e . o . n e . b . -. 1 .s, ‘J - a 1 v e ' . ~ . e .A d ~.. . v - - --=-c UV ~ \- a .‘ L .4 ‘ . e . - s e _ .. 5 ~ , . -..~ - . e . m a v .— i ’ . ~ . b e u . s . .. KI .. I m ..- ‘7 . C 4 . .- .. . -' ' x. . .- . e , . . ' .-. . . . 4 . e . r 'v . ’ . . g _ s . I K ' A. - i .. A._ . ul ‘ - . v- - ‘e- . e . a s ' . -. . -, c (I - _; \ x ‘ L I I - ‘ " _ i .L e . e ' § .— . . I , _ m . . .— ’- y—cu- .--.-.- as... en‘p- .. ‘ ‘ e “ - .. 376 APPENDIX IV (The following material is presented here because it is outside the scope of the study as outlined but suggests an approach to the study of mortality differentials which may be fruitful if systematically developed.) MORTALITY DIFFERENTIALS AND SEX ROLES: AN EXPLANATORY HYPO‘I‘HHSIS Students of mortality have been intrigued by the re- latively persistent excess of male over female deaths for a period of years. As early as 1665, John Graunt observed that 'Physioiens have two women patients to one man, and yet more men die than women."1 But it is yet to be explain- ed satisfactorily. Perhaps during the time of Graunt. the explanation would have been rooted in the concept of the evil nature or sinfulness of the male which, if given a sociological interpretation, would be closer to the hypoth- esis offered here than that of current writers who attribute sex mortality differentials to the biological make-up of the sexes. Early scholars (Pearl‘93 and Willcoxs) ascribed the differential to the biological superiority of the female, but recently emphasis is being given to social causes as an explanation of the difference (Thompson4 and Landiss). All contemporary authorities, however, have not adopted the l. Graunt John Natural and Political Observations Made Upon the Bills of Mortmty, (Edited Htfi Introduction by WaRSrT‘W. ifioox , John Hepkins Press, Baltimore, Maryland, 1939, p. xiii. 2.. Pearl, Raymond, 2p. cit., pp. 181-183. 3. Willcox, Walter F., 93. cit., p. 120. 4c Landis, Paul He, _ORe 0115c, Pe 214e 377 sociological explanation. For example, Whelpton recognizes social factors as possible causes of the variation but he stresses the statement “... that biological differences be- tween the sexes cause an important part of the differential mortalityt'l' He cites the same evidence as other adherents of the biological explanation, namely, the higher death rate for boy than for girl babies. That the infant mortality sex differential is a result of biological factors is acceptable to most students; however, the differential has increased over a.period, partly at least, because of the decrease in births per family. Social factors underlie the decisions of parents to have fewer children, and, thus, even the infant mortality sex-ratio favoring the female cannot be wholly attributed to biological influences. These are examples of the old dilemma, biological ‘ versus environmental explanations of behavior, which has pro- duced little more than a long list of untestable hypotheses.3 .Although Hholpton asserts that the maternal death rate is 'oloarly'3 a biological matter, as strong a claim can be made for the sociological explanation. .Are maternal deaths not l. thlpton P. K. Needed Po ulation Research, The Science Press Company, ianoaster, EennsyIvania, I538, p. 115. 2. Brown says emphatically that not a single question per- taining to a single trait has been answered by the biolo- logical versus environmental arguments. See Brown, 3.1., Psychology and the Social Order, mcGrawéfiill Book 00., Inc., ew Yer , 1933; (SeIectIon reprinted in Readi¥gs‘ig General ac Sociolo , Edited by Robert W. O'Brien, 0 Books- We, Palo Alto, California, 1947, pp. 117-118.) 3e Whelpton, Po K's, 22c 0113c, De 115. 3'78 considered by many authorities as largely preventable today? Besides, that kind of hypothesis pertaining to both sexes could one test employing the maternal death rate? Vance is apparently accepting the biological explanation when he writes as follows in All [13233 19321.93 "Since the wastage of male life by stillbirth and infant death is especially high, it seems probable that the really frail sex is the male, while nature protects the future child-bearer by endowing her with greater vitality and resistance to disease."1 In so far as could be ascertained the mortality sex- ratio has not been studied systematically to test any par- ticular explanatory hypothesis, although the generalization that males have higher death rates than females has been known for well over 100 years. The sociologist has done little more than to repeat the biological explanation ihich originated among the early demographers and physicians. As Whelpton suggests, research is needed to explain the dif- ferential even though society may or may not choose to reduce that portion of it which can be modified.3 In the following pages, a social causation hypothesis is presented as a possible explanation of mortality sex diff- erentials, namely, that the degree of obedience to sex roles3 1e Vance, Rupert Be, 22o Cite, pe 47s as Whelpton’ P. K., g. Cite, p. 115. 3. Sex role is defined as mutually related systems of atti- tudes and values ascribed by society to males and females. See Linton, Ralph, The Cultural Back round of Personalit , D. Appleton-Century—53., Inc., 1945; (Selectfon reprintei in Readi s ig,Social Ps cholo , Edited by Theodore M. Newcomb, keno f. HartIey, ané others, Henry Holt and Company, New York, 1947, p. 368). 379 union affect health varies inversely with the degree of en- vironmental resistance to their enforcement, other things being equal. Only certain aspects of the hypothesis, those involving data prepared for this dissertation, are discussed. The analysis is based on the following assumptions: 1. The prescribed rules which characterize the sex roles dominant in the United States force the male into greater health risk situations than the female. The sex role systems are constructed so that the male is required to expose himself to danger more than the female if he behaves as a male should. With a quantitative measure of the differential sex role ex- posure to health risk situations this assumption would be unnecessary and a more exact test of the hypothesis would be possible. 2. Sex roles defined by the general culture permeate all major segments of the society, whether or not they are en- forced in a particular class. However, isolated groups, such as certain Indians who are not exposed to the general cultural compulsions, cannot be assumed to possess the dominant sex patterns of behavior. ‘ 3. medical science has develOped in such a fashion that its benefits are equally available to males and females. It may be that some diseases peculiar to one of the sexes have received more attention fronimedical scientists than corre- sponding diseases for the other sex. It is assumed that these offset each other. In any event, deaths or illnesses ihich could be designated as primarily due to an undeveloped field 380 of medical science affecting only one of the sexes would come prise a relatively small proportion of the total deaths or illnesses in a given population for a specific period of time. 4. Properly calculated death rates are indexes of obe- dience to sex behavior rules defined by the sex roles. With this meaning a higher death rate is expected for males than females because the male role requires greater exposure to health risk situations. If death rates of males and females are equal, granted the first assumption above, environmental resistances are preventing the sex roles from operating. The sex roles are given by the culture; they are seg- ments of the social structure. Furthermore the male and female roles are constructed so that, although their rules are quite different, they are closely interlocked as to division of labor, social functions, and moral values. In our society the sex behavior system demands that the male assume the role of the protector,--- even at great personal sacrifice,--- in behalf of the female who is to be protected. This is interpreted by some as a socialized expression of the self-perpetuation drive. However, the overt expression of the male sex role is fortified by the development of a sub- Jective attitudinal system of behavior toward risks encumr bent upon the male, some aspects of which are irrational to the survival of either of the sexes. For example, the pro- tector role of the male which requires that women be removed first from a sinking ship has survival value for the female, and therefore has a rational basis, but the male attitude 381 that it is sissy to go to a doctor'has no survival value for either sex.1 Thus the very nature of the sex structure de- mands that the male asswme certain.health hazards as the protector, and certain others as a "true male." The sex role for the female, on the other hand, is so constructed as to encourage her to adapt herself protectively to en- vironmental health risks. Thus if permitted to operate unhampered as controls of behavior in a given environment, sex roles would produce a certain differential health in- dex (death rate) which would be fairly constant, other things being equal. However, within the same environment there are con- ditions (resistances) which may prevent the operation of the sex behavior controls. Poverty, for example, often forces women into risk situations from which they would be pro- tected in.a.higher economic group. Thus one could expect a greater degree of obedience to prescribed behavior of the sexes affecting health in areas or classes where environment- al resistances to sex roles are low because the general economic and cultural levels of the people are high. For example among Negroes, one would expect the degree of obe- dience to sex roles to be low because of the great environ- mental resistances to the role imposed by their economic status and education. And urban people would be able to obey the prescribed sex behavior rules more than rural people 1. It is these irrational aspects of the role which should be subjected to further research and clearly identified. 582 because of their greater control of environmental resistance to such behavior as indicated by their higher standards of living and greater variety of vocational Opportunities when pressed. Moreover the availability of health facilities makes it possible for the urban female to make the most of the advantageous position in which the sex role places her. Tie Female 3335 and Health The female role requires that women be both healthy and attractive, if not beautiful, since marriage and child- rearing are its ultimate objectives. To be beautiful is one of the basic themes in.the advertising beamed to the American woman. With beauty come health and happiness. Hence the young girl is groomed from childhood for the adult role which she is to play. These are among the most power- ful health drives for women. Traditionally her role is that of the weaker sex. Further the experiences of puberty and childbirth make contacts with the physician a normal event. Thus she establishes early the "going to the doctor" habit. Moreover the essential nature of the duties and functions she performs in a modern society do not impose great physical risks upon her. _ngngggglgggQHealth The male role is more or less the converse of that of the female. Masculinity has few aspirations to be beautiful, and there is very little, if any, relation be- tween.the concept of beauty and.health in the male system 383 of behavior. The male role provides no reason for acquiring the "going to the doctor habit." Sickly boys are labeled as ”sissies." The male is not aware of "father" and "occu- pational' roles until in the teen ages, although good health contributes greatly to the successful fulfillment of these roles. Furthermore the occupations probably do not elimi- nate him for physical handicaps or health reasons as come pletely as the traditional role of marriage eliminates the unattractive female. Even for marriage and having a family good health is not necessarily a prerequisite. Perhaps the only part of the male role that contributes to health is the athletic drive, the worship of athletic heroes, but even this adds to his health risks if his participation is too strenuous over a long period. Ada tabilit of Sex Roles 3g Environments; Changes Effectigg Health- If the assumption is accepted that death rates meas- ure the degree of obedience to or the enforcement of the rules regulating social behavior of the sexes, the degree of environmental resistance to the prescribed behavior can be estimated roughly on a "more or less than" level. Granted the assumptions, other things being equal, two propositions logically follow: (1) The greater the degree of obedience to the rules defined by sex roles the lower the death rate of females, the higher the death rate of males, and the greater their ratio; (2) The greater the mortality sex-ratio the less the resistances to the rules defined by sex roles, 384 and the lower the mortality sex-ratio the greater the re- sistances to the rules defined by the sex roles.1 With the foregoing assumptions and definitions in mind, one question to be considered is whether the degree of obedience to the male role affecting survival change as rapidly as obedience to the female role. Employing the age-adjusted death rate for each sex as a measure of obe- dience to the role, one wishes to know in which role con- formity decreased the most. The data in Table 1 shows that the female not only has a lower death rate but also that her death rate decreases more than that of the male over a period of time. The respective percentage decreases are 36.4 and 28.2 for age-adjusted death rates. Table 1 chaos AND AGE-ADJUSTEDl DEATH RATES PER éfiooo POPULATIEfib DEATH REGISTRATION STATES OF 1910, FOR T YEARS 1910 1940, BY SEX Meal Year Age Age Cpude Adjusted Crude Adjusted 1910 15.6 17.37 13.8 14.94 1940 12.3 12.48 10.0 9.50 Percent Decrease 21.2 28.2 27e5 36e4 1 Adjusted to the 1940 age distribution of the death-regis- tration States of 1910 by the direct method. sour0.3 Table 9e100 1. It occurs to the writer that one condition under which this would not hold is that of a class of the pepulation where the sex roles are extremely powerful producing a.high mor- tality sex-ratio and where the resistances, although great, are not powerful enough to level the sex behavior rules. Such a condition might be found, for example, among a people with upper class or nobility culture values but who now have low levels of living. 585 One may conclude from these data that the female role was able to take advantage of decreasing environmental resist- ances affecting health over a period to a greater degree than the male role. Table 2 MORTALITY SEIhRATIOS OF THE DEATH REGISTRATION STATES OF 1910, FOR 1910 TO 1940 (Death rates adjusted for age)1 Year Death Rate: M§1° fégzt. .x 100 United Stgtss 1940 131.4 1930 122.8 1920 109.0 1910 115.3 1 These death rates are adjusted to the 1940 age distri- bution of the death-registration States of 1900 by the direct method. Source: Computed from.adjusted death rates given in Table 9.6. The significant fact to be observed in Table 2 is that the mortality sex-ratio increased from.1910-1940, with the exception of the year 1920. One cannot conclude from this, however, that the nature of the male and female roles are becoming more divergent. But the trend is additional supporting evidence for the hypothesis that female sex roles are more adaptable to social changes which increase survival. By hmplication the trend indicates that environmental resis~ tances have decreased, except for the year 1920, permitting 386 the sex roles to produce an increasing mortality sex-differ- ential. Although there may have been changes in the rules governing the social behavior of the male and the female, in order to have produced such differentials by the changes alone, there would have been necessary a great increase in masculinity and femininity of the respective sex rol‘e‘s. Changes in the roles themselves, if they occurred, are like- ly to have been in the opposite direction, namely, decreases in masculinity and femininity. It is more likely that the network of social rules governing the female, including all of the mores which surround her, are of such nature as to permit, and in many cases to encourage, the female to make the most of improvements in medicine and in health care. ‘3‘ “bits tis-Sitfiglfi-fii‘sE-nfla Another question is whether the degree of obedience to sex role compulsions affecting survival increases more for one class than another. If so the mortality sex-ratio trend. should reveal the pattern, and the class having the greater environmental resistances would be identified. It is of interest to know whether the mortality sex differential increases faster for rural or urban people, white or colored groups, high or low income groups. Only one of these, namely, race, can be considered, because other data are unavailable (see Table 3). Mortality sex-ratios are substantially higher for the white than for the colored population. Moreover the ratios 387 of ihites show slightly greater increases from.1910 to 1940 than those of nonewhites. These increases were 12.28 and 10.01 respectively. In 1920 the death rates of males and females were approximately equal for colored people, the ratio being 101.95. For the same year the white ratio was 109.28, which was the lowest of the four decennial years. This was the aftermath of the influenza epidemic. These low differentials illustrate that where the sex structure pre- scribes protection for the female, she tends to receive it unless hardships of the environment prevent the operation of the protective rules, or uncontrollable epidemics tend to overshadow them, Thus when times are extremely difficult, for instance, during war, bombings, floods, or great epi- demics, the female is exposed to more nearly the same risks as the male and dies at about the same rats. Table 3 MORTALITY SEIFRATIOS OF THE DEATH REGISTRATION STATES OF i910, JOB 1910 TO 1940, BY RAGE (Death rates adjusted for age) [Mortality Sex-Ratio Year ' Colored lhite 1940 117.73 130.72 1930 111.12 123.01 1920 101.95 109.28 1910 106.99 116.42 (I: Adjusted rates are based on the 1940 standard million pop- ulation of the death-registration States of 1900. Source: Computed from age specific death rates given in Vital Statistics Rates in the United States 1900-1940, labi. , P. O .— 388 These data indicate that the white sex roles are able to express themselves through the control of behavior more completely than those of the colored, a difference which in turn reflects a low degree of environmental resistance among the white and high resistance for the colored group. _S_e_§ 321;; In Rural-Urban Ar__e_a_s and Health Do males and females respond to prescribed behavior rules for the sexes in rural areas more than in urban areas? The answer is "no" according to the data in Table 4. In 1940, the rural area has a lower mortality sex ratio than any of the various sized cities of the Uhited States. This means that environmental resistances to the roles were of such a nature in rural areas as to prevent the male and female roles frOm producing a high mortality sex differential. To the extent that the sex role explanation of mortal- ity differentials is tenable, the male living in small Ameri- can cities responds to the prescribed rules for males more completely than males in large cities. He has higher death rates than males in other urban places. Furthermore, the dif- ference between.his death rate and that of the female living in the small city is greater than the corresponding difference in any other sized city or in rural areas (see Table 4).1 l. Eliminating the infants, where biological differences be- tween the sexes may be an influence, and all those under 15 years there the sex roles may not be completely estab- lished as controls of behavior, the age-adjusted mortal- ity sex-ratio for white people in small American cities was 141.7 compared with 140.4 for the total population. The corresponding ratios for colored people were 128.7 and 128.2. 389 According to this analysis, therefore, the environmental re- sistances are less in the small city than in other sized cities, and, being less restricted, the sex roles operate more completely producing a greater differential mortality by sex. Table 4 AGE-ADJUSle DEATH RATES PER 16000 POPULATION AND MORTALITY SEIFRATIOS, UNITED STATES, 1 40, BY SEX AND RESIDENCE W __£L___Psl~ . lg_______. FFEEI_' R tes r l OOngopulation M21. .1: 100 Residence White Colored Male Female Male Female White Colored Total llefi 8.8 13e9 15.0 131e8 113s? Cities: . 100,000 over 12.5 9.1 19.1 15.6 137.4 123.1 10,000-100,000 12.4 9.0 20.8 16.9 137.8 123.4 2,500-10,000 13.2 9.4 22.3 17.4 140.4 128.2 Rural 10.3 8.2 15.1 13.6 125.6 111.0 I Adjusted to the age distribution of the United States, 1940, by the direct method. Source: Based on age-specific death rates given in Linder, Forrest E. and Grove, Robert D., Vital Statistics Rates gn__ the p______nitea §_______tates 1900-1940, Tame 4”."‘"" It was suggested in Chapter IV of this dissertation, while observing the Michigan mortality data, that the come bined effects of the tempo of city life and its stresses and strains on sex roles as one passes from the younger to the older ages was harder on males than females. In spite of the fact that the urban female is more and more taking up work outside of the home, according to this analysis the 390 protective aspects of her role tend to be maintained,1 al- though the urban female still has a slightly higher death rate than the rural female. That she tends to maintain her role is indicated by the fact that the difference between rates of females in rural and urban areas is less than the corresponding difference of rates for males. The tendency of urban females to accept ”outside the home" jobs therefore should not be over-emphasized in considerations of its ef- fect upon sex roles. Her duties do not usually carry with them.anough responsibility to create great worry and her failure in a job seldom offers a real threat to her status as a woman. On the other hand, the urban male is surrounded by occupational stress and strain. ,Although many urban wives work, the responsibility of economic security for the family rests firmly upon the shoulders of the husband. In the male role system, the husband is responsible for “bringing home the bacon", even if the wife in some cases does bring it most of the time. In the relatively few urban families where the female is the economic mainstay, the situation is likely to be explained vaguely as a condition resulting from.the husband’s illnesses or misfortunes. Furthermore, the man of the house, unless he is very exceptional, might be expected 1. See, for example: J. E. Walters, "women in Industry," ‘ghg Annals ggjthe American.Academ. of Political and Social cience, Septmfibzr I943, pp. 5 - ;'Eliza5etE fii—Efilver, IWomen in the Service," ibid., pp. 63-68. These articles illustrate some of the probIems created by attempts to maintain the female role in emergency situations. 391 to have some sort of inferiority complex, a condition accen- tuated, at least, by the operation of pressures associated with the sex roles. Such conditions, of course, increase the health risks of the urban male. §EEE§£I The essential facts upon which this analysis has been based are as follows: 1. The death rates of both males and females are decreas- ing, but the difference between their rates is increasing. 2. mortality sex-ratios are substantially higher for white than colored people. moreover the ratios of non-whites have increased less than those of whites. 3. The rural mortality sex-ratio is lower than for various sized cities. 4. The highest mortality sex-ratio occurs in small cities of 2,500-10,000 pepulation. Here the male has the highest death rate among all residential groups., 5. The mortality sex-ratio is highly variable both in time and among different classes of the population. Among residential groups it varies from 111 to 128 for colored people and from 125 to 140 for the white population. In terms of the approach developed, these facts seem to warrant the following tenative statements pertaining to the health of sex roles: l. The female compared with the male role is more adapt- able to social changes which increase survival. 392 2. Sex roles among colored people, compared with those among the white population, are not enforced as efficiently, reflecting greater restrictions placed upon them by the en- vironment. 3. Rules governing the social behavior of the sexes are not enforced as completely in rural as in urban areas. The male and female roles are probably as well defined among rural people but are not enforced because of environmental resistances related to lower standards of living, including lack of health facilities and ability to purchase medical care. 4. Apparently in the small American city (2,500-10,000), the expected behavior of males and females is enforced more completely, in so far as it affectssurvival, than in larger cities or in rural areas. This appears to be true for both white and colored people. 5. The variability of the mortality sex-ratio is, at least, a strong argument in favor of the social causation hypothesis of mortality differentials. Additional research is needed to identify the par- ticular attitudes inherent to the male role which contribute most to health risks. This is primarily the task of sociol- ogists and other social scientists. When accomplished it becomes the responsibility of health leaders to interpret such research findings in terms of health needs and health programs. In the words of Anderson, "... the mandate of the people seems clearly to place upon the health department a 393 responsibility for concerning itself with all forces of mor- tality, whether due to environmental, communicable, nutri- tional, physiological, or psychological factors.“ 1. Anderson, Gaylord W., "The Political Impact of Modern Science on Public Health," 'me Annalsp; the American Acadeg _o_f_ Political and Social Science, January I917, p. 2 e MICHIGAN STATE UNIVERSITY LIBRARIES 1 I III 3 129313082 8 35