--‘ ~~- ‘ 'fl'w‘ik 5%: 3 .W \l' ‘ y ' (“I . 4‘ ‘ if?" . . .. 1.11. ‘ f “-u' ‘." .1“(”I‘ V. n ‘Vr J. .;' .Afivfi.’- fi°fikv “ ‘ 1'" $57." l‘v- i . '1" :‘l '1 pvt/‘4‘!” \ :41: §§ § EV. fix '- u. 5’ fly," 3" U ‘ 04'1". .‘ ”91'8“ A. \ 'JJ k‘1 ‘0 ‘ r iu‘f‘.n;o>1 L - ' - I V ; . l". b 71‘ p l ‘ n ‘.» -.- :9. a f l“ . ». l ‘ .; I W. ., «.n‘ }%E* “WW I " ’ *6; "J 2. “r . _ n H '%£' 5-» - . .. ~ "-m u - ‘I\v§£~ . . . I. I.» , v ) M av # \ Lg: 39.»... 7" ‘ 'fnj-f ‘-‘.‘. . , - 4 V .\ ‘ . . 7. . . ) -‘. u |"|"'1 a. lu'.\“'l.‘\t l.‘.“'."";""‘ v..:‘.:' . . , "1 I 1‘34“ ...“,,‘l‘::" “J": " '75-'32 L" .‘) . ‘ ‘ W .m- Hw. ‘1.” n‘ u m. , .L.,_..r-,I'(';: “ - ‘._\ .I‘.‘ Jon. I .l.‘ «I _ s .(I‘. r .H ' h ’ ' . .;-"cr.!eu~m1w:. '.--‘.:’...“o'5-.’.5 W . . . . ‘ . . M . « w . M» ‘; ' l_-" ..‘ W" MTHJI U H" ’.|”:"|y’.”| . " ' m ‘ ‘ L“ r ,. ' ‘ ,, “u , ~ ‘. .’ . a. ‘I " i' ,9 flan")? '.\\",I' v‘ ‘ I “'1 1w,‘: h¢“' ‘ ”‘fifl4$"ufl!~i ,.$'(d§¢mfmT;W%ku£?W' h I l, ‘ " .I — ‘I ‘ ' .H: ‘u .‘H ‘ I‘ K ‘. . I ‘1; "14:. ., '_‘ ‘ ~ ' ..‘.‘ '.'.(‘ I . ".1 \ 2-‘1' flk. '. ‘l I‘A‘DLT‘ 4")!!!“[nutty-mad. LI! R Y lVlkfl1fi§MHISCIDB University This is to certify that the thesis entitled THE EFFECTS OF ACTIVITY PATTERNS 0N LONGEVITY 0F ATHLETES AND NON-ATHLETES presented by Kenneth Ellis Stephens has been accepted towards fulfillment of the requirements for Master of Arts degree in Department of Health, Physical Education, and Recreation , fflflQ/i/ / 4711 MM. / Major professor Date 10-30-78 0-7 639 THE EFFECTS OF ACTIVITY PATTERNS 0N LONGEVITY 0F ATHLETES AND NON—ATHLETES By Kenneth Ellis Stephens A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Health, Physical Education, and Recreation 1978 ABSTRACT THE EFFECTS OF ACTIVITY PATTERNS ON LONGEVITY OF ATHLETES AND NON-ATHLETES By Kenneth Ellis Stephens The purpose of this investigation was to examine the relationship between longevity and activity patterns, as reported in 1960, of athletes and non-athletes who had attended Michigan State University prior to 1938. Groups selected for comparisonincluded, by birth decade, living athletes, deceased athletes, living non-athletes and deceased non-athletes. Vocational activity, avocational activity and combined activity ratings were analyzed. A loglinear model was used to examine the relationship between athletes and non-athletes, while Chi-square tests were used in subsequent analyses. No difference between athletes and non-athletes was noted in terms of percent decreased, birth decade and total activity level. Likewise, no effect on total activity of impending death was noted in the popula- tion examined between 1960 and 1976. It was noted that both avocation- ally and vocationally, those most active in 1960 appeared to have a smaller percentage deceased between 1960 and 1976. DEDICATION To Ellis, Margaret and my family--for their continuous encouragement. ii ACKNOWLEDGEMENTS Sincere appreciation is expressed to Dr. w. D. VanHuss for his continuous encouragement, excellent guidance, and unending patience both in the preparation of this thesis and throughout my graduate program. Recognition is given to Dr. H. w. Olson for his assistance in the preparation of the thesis and to Dr. H. Sprague for both his edi- torial advice and his handling of all statistical analyses. Gratitude is extended to Jo Ann Janes for the initial typing, to David Anderson for the preparation of the graphs included in the thesis, and to Dr. C. D. Tweedle for his friendship and support. Special thanks is extended to my parents and family; for without their love, patience and understanding, this thesis would not have been completed. TABLE OF CONTENTS Chapter Page I. THE PROBLEM .............................................. 1 Statement of the Problem ............................. 2 Significance of the Study ............................ 3 Limitations .......................................... 3 II. REVIEW OF RELATED LITERATURE ............................. 5 Introduction ......................................... 5 Non-cohort Athlete Studies ........................... 6 Cohort Athlete Studies ............................... 18 Habitual Studies ..................................... 30 Summary .............................................. 37 III. METHODS OF PROCEDURE ..................................... 38 Data Base - Background ............... ~ ................ 38 Current Study Design ................................. 41 Sample Selection ................................ 41 Activity Ratings ................................ 41 Statistical Analysis ............................ 43 IV. RESULTS AND DISCUSSION ................................... 46 Athlete-Non-athlete Comparisons ...................... 46 Deaths Between 1960 and 1976 Related to Total Activity ........................................ 51 Avocational Activity Comparisons ..................... 51 Vocational Activity Comparisons ...................... 55 Total Activity Comparisons ........................... 58 V. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS ................ 61 Summary .............................................. 61 Conclusions .......................................... 62 Recommendations ...................................... 62 REFERENCES ...................................................... 64 APPENDICES A. QUESTIONNAIRES (1952, 1960, 1968, 1976) .................. 68 B. STATISTICAL ANALYSES ..................................... 83 iv Table 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 2.10 2.11 2.12 LIST OF TABLES Page Comparison of Actual to Expected Deaths Among Yale University Athletes (From Anderson, 1916) ................. 7 Percent Actual of Expected Deaths According to American Men Table (From Dublin, 1928) ............................. 8 Mortality by Birth Decade of Athletes of Ten Colleges Compared with Expected Deaths by American Men Table (From Dublin, 1928) ....................................... 8 Mortality of Athletes of Ten Colleges, Compared with the Expected Mortality According to the American Men Ultimate Table. Specified Class Groups Experience on Classes 1905 and Prior (From Dublin, 1928) ............................. 9 Actual versus Expected Deaths by Selected Age Groups (From Hartley and Llewellyn, 1939) ........................ 11 Longevity of Athletes and Non-athletes (From Schmid, 1952) 12 Mortality of Major League Baseball Players Compared with White Males in United States Po ulation (From Metropolitan Life Statistical Bulletin, 1975 .......................... 14 Summary of Non-cohort Athlete Longevity Studies ........... 16 Deaths Among Yale Graduates as Compared with Expected Deaths of American Men Ultimate Table, 1905-1923 (From Greenway and Hiscock, 1926) ............................... 19 A Comparison of Expectation of Life in Years (From Dublin, 1932) ..................................................... 19 Survival Rates of Sportsmen and Controls Excluding Deaths Due to War and Accidents (From Rock, 1954) ................ 21 Harvard Athletes, Alive or Dead of Natural Causes; Percentage of Men Reaching Ages 70 and 75, by Birth Decade (From Polednak and Damon, 1970) ........................... 23 Table 2.13 2.16 N .18 Harvard Athletes: Mean Age at Death for Men Dead of Natural Causes (From Polednak and Damon, 1970) ........... Mean Age at Death of Athletes and Non-athletes (From Olson, 1972) ............................................. Comparison of Average Life Span of Crew Members and Controls at Harvard and Yale (From Prout, 1972) .......... Summary of Cohort Athlete Longevity Studies .............. Morta1ity Rates for Men Aged 45-64 Years from All Causes (From Morris and Heady, 1953) ............................ Summary of Habitual Studies .............................. Configuration for Numbers in the Michigan State Longevity Study 1952-1976 (From Olson et a1., 1978) ................ Summary of Loglinear Model Analyses ...................... vi Page 24 25 26 28 31 36 40 SO LIST OF FIGURES Figure 2.1 Vocational and Avocational Physical Activity by Age (From Rose and Cohen, 1976) ............................... Percent Athletes and Non-athletes in Varied Avocational Activity Levels ........................................... Percent Athletes and Non-athletes in Varied Vocational Activity Levels ........................................... Percent Athletes and Non-athletes in Varied Total Activity Levels .................................................... Percent Athletes and Non-athletes Deceased ................ Percent Athletes and Non-athletes in Varied Birth Decades. Total Activity Level Reported in 1960 and Deaths Between 1960 and 1976 ............................................. Avocational Activity Level in 1960 and Percent Deceased Between 1960 and 1976 by Birth Decade ..................... Vocational Activity Level in 1960 and Percent Deceased Between 1960 and 1976 by Birth Decade ..................... Total Activity Level in 1960 and Percent Deceased Between 1960 and 1976 by Birth Decade ............................. vii Page 35 47 47 48 48 49 52 53 56 59 CHAPTER I THE PROBLEM Investigations reporting the effects of activity on longevity have been diverse in their findings. Early experimentation led to the conclusion that college athletes, when compared to the general popula- tion, were provided with a slight advantage in terms of lifespan (1,5, 6,9,10,14,18,22,41,42,46,50,51). Further studies compared university athletes with a more comparable group, other college students enrolled at the same time in the same school. Results indicated that all college students were longer-lived than the normal populace (21). In addition, it was found that other college students, particularly honor students, seem to have lived longer than cohorts who were athletes (l9,22,28,29, 37,38,44). A final group of papers attempted to assess later life (post-college) activity patterns and evaluate their relationship with mortality. These studies focused primarily on occupational activity levels and concluded that activity did indeed provide some measure of protection against death (2,15,16,34,39,45,49). Research pertaining to avocational and vocational participation patterns throughout both college and in later life was scant and non- conclusive. The effects of habitual activity patterns on mortality and longevity were also non-conclusive. Statement of the Problem The purpose of this investigation was to examine the effects of activity on the longevity of former athletes and non-athletes who had attended Michigan State University prior to 1938. Specifically, subject activity was measured using responses given to the first follow-up questionnaire of the 1952 study entitled Longevity and Morbidity of College Athletes. Three activity ratings were utilized which included vocational activity, avocational activity, and combined activity. Groups selected for comparison under these parameters were living athletes, deceased athletes, living non-athletes, and deceased non- athletes. The current study sought to provide insight into the following questions: 1. Who was more active in 1960, 22 years after the youngest subject's graduation, athletes or non-athletes? 2. Did activity level as reported in l960 decrease significantly prior to death (excluding war and catastrophic deaths) and if so, how long prior to death was a decrease in one's activity apparent? 3. How was mortality affected by various types (vocational, avoca— tional or combined) and intensities (sedentary, light, medium, and heavy) of activity? Significance of the Study Information concerning the relationship of physical activity to length of life should be provided by the present study. Although one animal study (43) has demonstrated increased longevity as a result of regular physical activity, this has not been observed conclusively in man. Resulting from this study will be information concerning the relative physical activity of former athletes and non-athletes in their later years and this relationship to mortality. Limitations l. The study was limited by use of a mailed series of question- naires (1952, 1960, 1968, and 1976). The numbers of non- respondents, especially large in the original 1952 study, was a source of possibly biased infbrmation. 2. Use of a population of male students in attendance at Michigan State University prior to 1938 and largely of a rural origin may have influenced the data and its applicability to predic- tions for current longevity. 3. The study considered only those respondents alive in 1960 who either were proven deceased or responded to the 1976 question- naire. 4. The study was additionally limited to the examination of selected, subjectively-rated variables. 5. The study examined activity as reported in the 1960 question- naire response only. 6. Any deaths attributed to war or catastrophic causes were omitted. 7. A case for causation cannot be made on the basis of the data at hand (a non-causal relationship currently exists). CHAPTER II REVIEW OF RELATED LITERATURE Introduction -Many diverse commentaries have been presented on the values of physical activity as a means for preserving or enhancing health, and consequently life. Galen was of the opinion in the second century A.D. that "while athletes are exercising their profession, their body remains in a dangerous condition, but when they give up their profession they fall into a condition more parlous still; as a fact, some die shortly afterwards; others live for some little time, but do not arrive at old age. . . ." (9). Hippocrates concurred stating that "the truth is, however, that no one is in a more risky state than they. . . ."(48). While such pronouncements bore great weight, ancient converse statements such as that by Timaeus in the Dialogues of Plato which indicated his belief that the body "by moderate exercise reduces to order according to their affinities the particles and affections which are wandering about . . ." (7) have clouded the issue throughout history, and indeed, until much more recent times. More current investigations have sought to resolve this issue and in some instances have expanded the study beyond the athlete-non- athlete comparisons to examine the role of activity in terms of health and life promotion. Consequently, the current review of literature 5 focused on three main types of studies conducted within the last one hundred and twenty-five years. Non-cohort Athlete Studies The initial studies were conducted along actuarial lines by com- paring the general population with select groups of athletes. The first such investigation was initiated in 1873 on 251 men who had rowed in the Oxford-Cambridge Boat Races between 1829 and 1859 (9). In this study Morgan concluded that, taking all the oarsmen together, each individual was "likely to survive the race on an average of some forty-two instead of forty years" as predicted by Dr. Farr's English Life Tables. A confirming study by Meylan conducted in 1904 (18) on 152 Harvard University crew members found athletes favored by 2.88 years (5.09 years 'with allowance for Civil War deaths) over mortality table values. A subsequent study by Gaines and Hunter in 1906 (12) found that Yale athletes graduated prior to 1905 experienced deaths only 49% of those predicted by insurance tables. Additionally, it was shown other graduates had a 70% mortality ratio when compared with an insured population. Anderson in 1916 (1) compared 808 Yale track, football, baseball, and crew athletes with two insurance populations. As indicated in Table 2.1, the athletes were again favored with an apparent increased longevity. The 1927 study by Hill (10) on British cricket players drawn from Wisden's Cricketers Almanac again illustrated a superior longevity for athletes at every age. Table 2.1. Comparison of Actual to Expected Deaths Among Yale Univer- sity Athletes (From Anderson, 1916) Per Cent of Per Cent of Number Actual to Actual to of Number Expected Deaths Expected Deaths Year of Letter of Actuarial American Earliest Winners Deaths Society Table Table Data Track 276 13 62 52 1868 Football 213 16 58 52 1872 Baseball 148 ll 47 ' 42 1865 Crew 171 18 45 41 1855 Total 803 58 52 46 ---- Dublin's initial research in 1928 (6) was confined to pre-l905 graduates of ten Eastern American universities and numbered 4,976 letter winners in six sports or sport categories. Comparisons made with the Medico-Actuarial Table and the American Men Table of Mortality found that taking all athletes together the actual deaths represented, respectively, 93.2% and 91.5% of the table values. Dublin was careful to point out the limitations involved in use of actuarial tables, and did express some concern that while his athletes presented a favorable mortality ratio it might have been still better, considering that the college men selected represented "the cream of the cream of American man- hood". Dublin also reported mortality ratios for the data when broken down into sports, age, and colleges (see Tables 2.2, 2.3, and 2.4). Table 2.2. Percent Actual of Expected Deaths According to American Men Table (From Dublin, 1928) All Classes Classes Prior Classes Classes Classes Sports Combined to 1880 1880-1889 1890-1899 1900-1905 All sports combined 91.5 94.1 94.6 93.6 72.6 Baseball 98.0 98.6 97.4 103.1 81 4 Crew 94.1 92.2 72.1 124.0 113.4 Football 88.3 87.6 94.7 91.6 63.4 Track 91.8 .... 121.7 86.6 73.0 Two or more sports 78.3 85.8 80.8 81.0 62 9 Table 2.3. Mortality by Birth Decade of Athletes of Ten Colleges Compared with Expected Deaths by American Men Table (From Dublin, 1928) Expected Per Cent Actual Deaths by Actual of Class Group Deaths American Men Table Expected Deaths All classes 1,202 1,314 91.5 Prior to 1880 454 483 94.1 1880-1889 322 340 94.6 1890-1899 310 331 93.6 1900-1905 116 160 72.6 Table 2.4. Mortality of Athletes of Ten Colleges, Compared with the Expected Mortality According to the American Men Ultimate Table. Specified Class Groups Experience on Classes 1905 and Prior (From Dublin, 1928) All Classes Classes Prior to Classes Classes Classes College Combined 1880 1880-1889 1890-1899 1900-1905 All colleges combined 91.5 94.1 94.6 93.6 72.6 Amherst 74.7 91.1 62.9 73.6 56.2 Brown 74.1 80.4 99.5 59.8 51.7 Cornell 94.6 99.5 104.1 97.9 70.0 Dartmouth 113.4 112.4 113.8 127.1 80.9 Harvard 100.1 107.5 99.2 94 9 85.0 Massachusetts Agricultural 84.4 98.6 94.0 * Tulane 94.2 .... .... 75.7 120.3 Wesleyan 85.3 93.4 84.2 73.6 61.2 Williams 79.5 .... .... 61.9 101.1 Yale 89.2 82.0 93.7 108.6 71.1 *Rate not significant 10 Another investigation by Reed and Love (42) contrasted Army officers with the actuarial populations, and also West Point athletes with other West Point officers. His 1931 investigation showed favorable longevity for officers over the general populace and also for athletes over other officers. Two 1930's investigations focused on oarsmen again. Cooper et a1. (5) in 1937 found that of the 100 Ormand College rowers 24 had died as compared to an expected 31.8. His comparison was the average Australian insurance holder. The second oarsmen investigation, a 1939 study by 'Hartley and Llewellyn (9), used Oxford-Cambridge boat race veterans of the 1829 to 1928 era. As with Cooper's work, however, the numbers were relatively small (767). Hartley and Llewellyn compared the study groups, after they had divided it into four groups based on age, with standard mortality tables. As can be seen in Table 2.5, when compared with assured lives of their own generation, college oarsmen seemed favored in terms of longevity. They also pointed out two factors of significance; specifically, that this superiority tended to diminish in late years and also that when compared to the more current 1924-29 standard table the comparison proved less favorable than in up to age 50 comparisons in the first three periods. In a study which deviated from the preceding investigations in that high school athletes were compared with the United States Census Bureau standard mortality ratios rather than college athletes, Wakefield (50) found that there were 123 actual deaths as opposed to an expected 181.1. Additionally, this 1944 examination found longevity was more favorable in young players (14, 15, and 16 year olds' ratio was 54.2% 11 Table 2.5. Actual versus Expected Deaths by Selected Age Groups (From Hartley and Llewellyn, 1939) Expected Deaths by Age Group Actual Deaths Standard Table Actual as Percentage of Expected Period 1.--1829-62. Standard Mortality Table HM Up to 50 .. .. 22 25.2 87.3 51 to 70 .. .. 1 1.0 100.0 Over 70 .. .. -- -- -- A11 ages .. 23 26.2 87.8 Period 2.—-1863-93. Standard Mbrtality Table 0M Up to 50 .. .. 53 62.1 85.3 51 to 70 .. .. 39 59.8 65.2 Over 70 .. .. 18 21.6 83.3 A11 ages .. 110 143.5 76.7 Period 3.--1894-1923. Standard Mortality Table Mean of'OM and A 1924-9 Up to 50 .. .. 32* 40.7 78.6 51 to 70 .. .. 85 98.0 86.7 Over 70 .. .. 108 125.8 85.9 All ages .. 225 264.5 85.1 Period 4.--J924-8. Standard Mortality Table A 1924—9 Up to 50 .. .. 3 3.4 88.2 51 to 70 .. .. 13 13.2 98.5 Over 70 .. .. 20 21.9 91.3 All ages .. 36 38.5 93.5 *Excluding 37 Great War deaths. 12 versus l7, l8, and 19 year olds' ratio of 78.9%) and that the ratio for boys playing in three games in one day was 69.4%. It is also noteworthy here that the group ratio was 67.9%. Montoye et a1. (22) have cited a report on 400 deceased athletes in Czechoslovakia with at least ten years of sports competition con- ducted by Schmid in 1952. With war deaths omitted, the mean age for various sports was reported along with a comparison of athletes and non-athletes from the general population (see Table 2.6). Again the athletes seemed to have been favored. Table 2.6. Longevity of Athletes and Non-athletes (From Schmid, 1952) Mean Age of Mean Age of Non-athletes Athletes Year of Birth (Years) (Years) 1861-1870 58.24 66.90 1871-1880 60.28 61.72 Pomeroy and White in 1958 (39) attempted to contrast the longevity of Harvard football players with that of the general Massachusetts popu- lation and other Harvard graduates who had not played football but found "adequate data fOr any such statistical comparison were lacking". They did examine lifelong habits of exercise in a coronary group, both decreased and living, and a group living without coronary disease. It was concluded that "men in the coronary heart disease group engaged in less vigorous exercise than did those without heart disease". 13 While Karvonen's 1959 paper (14) focused on training of the cardio- vascular system, he did examine the length of life of sportsmen involved with their sport over a prolonged period. It was concluded in his study of 388 pre-1930 champion skiers, many of whom still skied, that "training does not shorten life but may even prolong it". The compari- sons with the general population were certainly in the athletes' favor, while a similar comparison with a select insured population showed little in the way of difference. In a continuation of Karvonen's emphasis on cardiovascular studies in former endurance athletes, Pyorala et a1. (41) found 93 men aged 40 or more of which 40 had been long distance runners and 53 skiers. A control group selected from a similar social class was compared over numerous variables, one of which was physical activity. It was found that "irrespective of age and absence or presence of cardiovascular disease, the average degree of physical activity was higher in the athletes than in control subjects". Age at death or mortality ratios were not presented, consequently no athlete-population comparisons were made. Conversely, a Danish study by Schnohr in 1971 (47) on 297 male athlete champions born between 1880 and 1910 sought to examine differ- ences in mortality when compared with the general male population. Up to age 50 a mortality ratio of 61% in favor of the athletes was signifi- cant. For the age periods 50 to 64 years and 65 to 80 years the ratios were 108% and 102% respectively. Schnohr, in a closing comment, stated neither former athletic champions nor less successful athletes studied continued vigorous exercise after years of competition. 14 The final non-cohort athlete study was conducted in 1975 by the Metropolitan Life Insurance Company (51) on major league baseball players and used the white male population of the United States as a comparison base. Mortality ratios for all players indicated that those playing in the 1876-1900 period had a mortality ratio of 103%--whereas the ratio was 71% in the 1901-1973 period. Additional information was provided concerning mortality by position and batting average (see Table 2.7). Table 2.7. Mortality of Major League Baseball Players Compared with White Males in United States Po ulation (From Metropolitan Life Statistical Bulletin, 1975) Players Who Had Careers Beginning 1876-1900 1901-1973 Mortality Mortality Deaths Ratio* Deaths Ratio* Batting Average .300 or more ........ 77 88% 102 71% .250-.299 ........... 356 99 547 74 .200-.249 ........... 364 107 558 65 less than .200 ...... 263 109 588 75 All ................. 1,060 103 1,795 71 Position Played Pitcher ............. 242 105% 594 79% Catcher ............. 118 108 184 69 First Base .......... 72 100 96 79 Second Base ......... 65 105 104 68 Third Base .......... 69 103 88 55 Shortstop ........... 58 90 94 67 Outfield ............ 221 95 342 72 Source of basic data: The Baseball Encyclopedia, 2d ed. New York, Macmillan Publishing Co., Inc., 1974. 15 Table 2.8 is a summarization of the non-cohort athlete studies to date. In general it can be seen that athletes had apparently secured some benefits from their participation as they lived at least as long or longer than the general population. Prior to accepting this conclu- sion one should be forewarned of the problems inherent in these types of investigations. As previously pointed out, many of these studies had either too small an initial number or too few of their numbers were deceased to make significant predictions. In addition, many of the mortality tables selected for a comparison provided the investigators with an inherent source of bias. Actuarial tables provided a contrast group of select individuals, those who generally were capable, both physically and financially, of securing life insurance. General popula- tion tables reflect both the healthy and the unhealthy in the population, and thus may have tended to accentuate the differences between groups. Use of either table may have provided additional bias in that the com- parison is made between mortality ratio on a single year's table and actual athletes, whose life spans many years, consequently many tables. The very use of college athletes, especially those selected in the early portion of the century and before, could be considered an influ- ence on the results. Dublin perhaps summarized it best when he stated that "the college man was a picked man; his home represented an economic standard far above average; he was usually of American parentage, a race stock with an excellent rate of longevity; his occupation after leaving college was usually one in which he was not subjected to the hazards involved in so many pursuits". It was his belief that as a result it 16 Ah~—~_=w.c .uoaccu ua.n~ ouuex su_—uuuox couc>cm mouo_;.< occcuaucm c-.~quuo:< cc. sna— 22... 83:3 Foes: choc—uuo n z >v=um auwuuuux undo; any: 55.... 3...: a... 5:53.. 1.25 3.82. 353308. 2.: apnea m~.~um~. an wouo>em wou~_;u< .0 u—aeh 30: coouuol< nouuuuonuca ena~ I Aou>uoo :«v doo— uno— o>cg use new: Axhxuu couo_;a< sues» uauuoauo<16onvot o~o.c oeuuugu< “cauiooc— a~o~ Acv cu~;:= n .02 o-seh auua sou—new acouqu-suco -< we. chums—a uOJUuuu .oom< _~< an ucluuaucuum vouo>em uo~o~z~< c .02 o—ach 0“; gnu—new e~¢.n snuuusn coaunoooq “wo— Aouv ~_«= Ahnm couo~su< o—A-h suouuom un¢unauo< coo invade: o—or «co—imnc— cuo— adv conuoe=< nouomgu< Amuv noucaz no. cue-x uuuuouuo: vouoacm nauonsu< ovuach oucnuaocu coda—vogue: suu0u9>uea o—a» moo— was cao— vac nos—cu cologne huuuuo>nca 28» 2: 8 332:. a3££< 22: 2:32: 335.5 «2 Eat... 2.2.2.: 3.: 3: 51o: calcuxo ou—auh suuauo>u== ouuvuuAqu 33» ad a. 3.3.... 8.3.3 :3 .3295 3:... :5 :~ a... 38.5 32.32 2.: .3 59:: oucollou unnuvcuu co—uc—zscm cauvuqalou aegis: couu¢_=acn voculcwu u¢o> haucwuuco>cu 333m Afi>mmco4 302.2 228.52 “B ALE-=3 .m.~ 2am... l7 u- Oman: uuuuauuox anouuqoou noucum vogue: 0:» .0 Ana—e: «nuns—m dungeon: A_nv ouua "no" cuuam hauucuuo: coo—nasc— echoes; couougu< onusav coma-naaom ~cuococ nn~.o oaucoa Lena: amouncsn— msom cauuuoaouuoz accurac— on on. uoo~.ouuu¢ Auu—uuuoz wouoasu< sods-osu undo ow ou< cu cuuqu sud—wane: vouobuu noun—:u< coda-games nun: unuococ sow sauce: o~o~ioocu usou Anew uaocsum ova: yo: candy-aloe roman: su«-uuot .>u«>uuuc couuuusnom chausw van ounces: A~ev no oouuov nozuus a o>c= amoeba; nouo~gu< so«c«m woodcuom s—lovcuu no ocean-«n nee; anucum soon .un uo u-uuo>m coma-«anon oucnuaucu saw: noose cauua—aaom oucauao¢~ uuouuuv uauouuucuuuicoz .nno— nmouaoeou we: couuaua u—no— uo>o caucquuuun Com—olm rues sowed» cunt unaccou cucuxm 90—5 .ocouu—nou uw>o nuns» sic an vanes-L nous—:u< nnmgi—mou can coo—Inna— can ilisu saunas ono~ 0»; anon Aauv coco>uqx .onuuuowo ~c:~«aas we. oaoaouuu> good :« vouocco oeuqavnuu annoys: sauna uuoeouou .o~a«oooa no: ousuc vcn gang-gage; coluouuoa uuuauoou yuan Aonv ouugn condensing coda-uaaomuuuo~;o< vovuuouaocz nuuoosgo-non: unaccoo Conn owe layout: vuo>uuu anoqiooo- anon vac scum-om nouo~gu< caucuzu< once» cv.~noo.u no vouo>os uo~o~g~< icon caducuaaoa ~auocoo coo nous->o—ocsuouo onc—u~cc— «nod Auuv unison coda-h and; oaocou anus-us nuuauoxuon uooaun no.~o cue-u Ana—nauo: cones-u noun—=u< mo auouan nob-um anode: a~o.~ gnu: ac-«vcu ana.u_—o~ ccou Acme e_o.uox-: Na.na ouun: mud—nuuot a sauna; u~.no nay-x auguuuuo: n vogue; «5.05 cuuan suuuuuuot N weapon A< .<+1om..lo .12. cologne uuucuosuca outdo; he» cu-osuqa un.~u o-o¢ uuuuuuuct u vauuos couobau ucuuusu< noun-h uuu~uuuot vu-ecoum a has luau we. cucuwo cumuiaucu onu— v:- souuucz oucoeloo nugget—u couuuuaaom goody-nice avg-:2 codes—anon Queuing: your meannuuuo>=~ uaaewacou--m.~ «page 18 was to be "expected that the athletes would show a lower mortality than the usual mortality tables" (6). Efforts to eleviate some of this bias and present a clearer insight into the role that athletic participation plays in one's longev- ity produced a more complete and accurate compilation of data in a new series of studies. Cohort Athlete Studies The succeeding studies used non-athletic cohorts as a control to compare peer groups. In this manner, athletes and non-athletes drawn from the same population at the same time could be compared across a wide variety of variables of which age at death or some such measure of mortality was one. This format was used by numerous investigators including Greenway and Hiscock in 1926 (22). Their study of Yale graduates and lettermen is represented in Table 2.9. While consideration must be given to the small numbers involved, the data indicated, as pointed out by Montoye et a1. (22), "that the superior longevity of college athletes may very well be due to the fact that these men were members of a select group, namely college graduates. . . ." Dublin in 1932 (22,28) attempted to resolve the criticism levelled at his earlier work by comparing the longevity of the same 4,976 letter winners with that of 38,269 graduates of eight eastern American colleges. ‘Of these graduates, 6,500 were honor students and two-thirds were selected from Harvard, Yale and Cornell. The data (see Table 2.10) 19 Table 2.9. Deaths Among Yale Graduates as Compared with Expected Deaths of American Men Ultimate Table, 1905-1923 (From Greenway and Hiscock, 1926) Expected Deaths Ratio of Actual American Men to Expected Deaths Groups Deaths* Ultimate Table (Per Cent) Non "Y" Men 317 381 83 "Y" Men . 27 29 93 All Graduates 344 410 84 *Excluding deaths due to war injuries. Table 2.10. A Comparison of Expectation of Life in Years (From Dublin, 1932) General College College College American Men Graduates Athletes Honor Men Table Age (38,269) (4,976) (6,500) (1900-1915) 22 45.71 45.56 47 73 44.29 27 41.68 41.41 43 61 40.18 32 37.59 37.25 39 48 36.03 37 33.51 33.09 35 30 31.83 42 29.44 28.92 31 07 27.66 47 25.37 24.80 26 85 23.62 52 21.43 20.85 22 79 19.79 57 17.78 17.34 19 O3 16 25 62 14.48 14.09 15 56 13.06 67 11.47 11.06 12 36 10.28 72 8.81 8.41 9.50 7.91 77 6.52 6.15 7.06 5.96 82 4.56 4.24 4.98 4.41 87 3.01 2.77 3.30 3.21 92 1.92 1.75 2.11 2.30 20 revealed that honors graduates seemed to have a slightly greater longev- ity at every age (approximately two years). Also it was shown athletes had a life expectancy which closely paralleled that of other college graduates. Another group of honors graduates, athletes and a random sample of students from the 1860-1900 Cambridge University classes was the subject of Rook's 1954 examination (44). Survival rates, presented in Table 2.11, along with average age-at-death figures indicated that the 374 honors students generally were longer lived by a period averaging 1.5 years. This difference was not significantly different from those values registered by the 379 randomly selected students or the 772 athletes. Rook's data also indicated the more heavily built man had somewhat of a disadvantage when considering the prospects of longevity. In the first of a series of studies conducted by Dr. Henry Montoye (22) on Michigan State University athletes and non-athletes, he found only 122 deaths. In this initial study age-at-death examination pro- duced no significant differences between groups. Montoye and colleagues also examined vocational and avocational activities in an effort to determine if regular exercise throughout life had an effect on longevity. It was found that beyond the age of 45 non-athletes were significantly more active in sports. This situation was reversed prior to age 45. Participation in non-sports activities yielded additional significant differences, with non-athletes indicating a greater participation at practically every age period. These results may be somewhat clouded owing to the poor initial returns from questionnaire mailings. 21 Table 2.11. Survival Rates of Sportsmen and Controls Excluding Deaths Due to War and Accidents (From Rook, 1954) No. of Survisors at Each Age Age Sportsmen Intellectuals Random Group 20 1,000 1,000 1,000 25 996 987 985 30 982 962 964 35 963 948 949 40 934 937 913 45 898 910 889 50 870 888 867 55 825 841 837 60 763 791 754 65 673 738 634 70 523 597 532 75 377 434 364 80 186 231 193 85 82 92 82 90 23 26 23 No. in group .. 723 362 325 Average age at death 67.97 69.41 67.43 Variance of estimate of average age at death .. .. 0.29 0.61 0.74 22 Seven year follow-up data presented in 1962 (21) verified the original findings on longevity. Montoye summarized "that there were no appreciable differences between athletes and non-athletes in longevity or cause of death, excluding violent deaths. Both athletes and their controls from the university sample lived longer than the general insurance risks. . . ." Montoye in 1967 (19) presented an updated paper on the status of his groups. In it he studied the original and follow-up groups of deceased subjects. He reported the mean age at death for athletes was 62 years and that for non-athletes it was 64 years (a nonsignificant difference). In discussing the effects of exercise continuance in the seven year period between studies, it was noted that "the amount of physical activity was significantly greater among subjects who survived seven years". Paffenbarger et a1. (31) reported on factors which may have been precursors of coronary heart disease in a study which utilized over 45,000 college students from the University of Pennsylvania and Harvard University between 1921 and 1950. One factor considered was varsity athletics. It was reported that an inverse relationship existed between varsity athletics and coronary heart disease, both in terms of subjects with specified coronary heart disease factors and in terms of mortality from coronary heart disease. The estimated mortality ratio at all ages was 60%. His second report discussed characteristics predisposing to stroke (32). It contrasted 171 stroke victims with 684 of their classmates and 23 revealed seven precursors of fatal stroke, one of which was non- participation in sports. The estimated mortality ratio of 0.4 reflects this indirect relationship between death from stroke and participation in varsity athletics. One of the more recent studies was completed by Polednak and Damon in 1970 (38). The longevity of 2,090 men, alive and dead, from the classes between 1880 and 1916 at Harvard was examined in the same manner as Rook. Data were presented (see Table 2.12) which indicated that minor athletes emerged as the longest-lived group, while major athletes and non-athletes failed to differ. In considering age at death (see Table 2.13) the only consistent finding was that major athletes were the shortest lived in each birth decade. An attempt was made by the authors to explain this trend based on a discussion of somatotype. Table 2.12. Harvard Athletes, Alive or Dead of Natural Causes; Per- centage of Men Reaching Ages 70 and 75, by Birth Decade (From Polednak and Damon, 1970) Reachin A e 70 Reachin A e 75 Birth Decade 1860-69 1870-79 1880:89‘ TBEUZB§"Tggfirgg“ngU:gg N %* N %* N %* N %* N %* N %* Major athletes 26 46.4 52 59.7 14 60.9 21 37.5 42 48.3 12 52.2 Minor athletes 29 58.0 86 63.2 55 76.4 21 42.0 72 52.9 48 66.7 Non-athletes 239 57.2 444 60.7 255 62.9 186 44.5 333 45.5 201 49.6 For significance of differences, see text. *Per cent of men in each athletic category in each birth decade who reached the specified age. Thus, of 56 major athletes alive or dead of natural causes in the birth decade 1860-69, 26, or 46.4%, reached age 7 . 24 Table 2.13. Harvard Athletes: Mean Age at Death for Men Dead of Natural Causes (From Polednak and Damon, 1970) Age at death 1860-69 1870-79 1880-89 (yrs.) N Mean 5.0. N Mean S.D. N Mean S.D. Major athletes 55 67.3 16.8 84 69.8 15.9 16 66.2 14.2 Minor athletes 49 67.9 18.0 119 70.8 14.7 43 67.2 16.8 Non-athletes 416 69.3 17.4 699 70.1 15.0 308 67.7 12.5 Note: Percentage of deaths among original cohort in each birth decade: 1860-69, 99.3%; 1870-79, 94.6%; 1880-89, 75.7%. Differences among athletic categories within each birth decade are not significant. Polednak expanded his study to include a total of 6,303 Harvard students born between the years 1860 and 1889 (35,37). His data were again divided into birth decades to minimize time trends and as in his earlier paper the most significant finding was the reduced longevity of major athletes relative to their classmates. The differences were small, usually between one and three years in mean age at death, with major athletes the shortest lived in each birth decade. It was anticipated that "since slightly larger percentages of minor athletes are still alive in the two most recent birth decades . . . eventually the minor athletes will improve their longevity in relation to the other two groups". Ratings for both avocational and vocational activity throughout life were derived from the 1960 follow-up of Montoye's study and pre- sented by Olson (28). In general, those alive in 1968 were found to have been more active in 1960 than those deceased between 1960 and 1968. 25 This was irrespective of the athlete-non-athlete categorization. College athletes and non-athletes failed to vary in terms of vocational and combined activity ratings. The avocationalratings indicated the former athletes were significantly more active than the non-athletes. Age-at-death comparisons favored non-athletes (see Table 2.14), however differences here were not significant. Table 2.14. Mean Age at Death of Athletes and Non-athletes (From Olson, 1972) 1952-1960 1960-1968 Total 1952-1968 Mean Mean Mean Subjects N Age at Death N Age at Death N Age at Death Athletes 46 67.32 62 72.29 108 70.17 Non-athletes 30 69.81 57 72.46 87 71.57 Both Harvard and Yale crews, numbering 172, were compared with randomly selected, matched classmates from the years 1882 to 1902. Prout (40) found a highly significant age-at-death advantage in favor of the athletes, the difference being 6.35 years at Yale and 6.24 years at Harvard (see Table 2.15). In view of the preceding studies these results were somewhat of a surprise and unexplained apart from the small nunbers involved . 26 Table 2.15. Comparison of Average Life Span of Crew Members and Con- trols at Harvard and Yale (From Prout, 1972) No. of Average Group Subjects Life Span t-test P Harvard crew 90 67.79 2.41 <.05 Harvard controls 90 61.54 Yale crew 82 67.91 2.37 <.05 Yale controls 82 61.56 Combined crew 172 67.85 3.39 <.01 Combined controls 172 61.55 A final paper in the cohort athlete series was presented by Olson et al. in 1978 (29). The average age at death for the 275 deceased athletes was 68.13 years, as opposed to 70.17 years for the 227 deceased non-athletes. At the time of the study roughly 40% of all subjects had died. While cohort athlete studies provided a unique and superior approach to the study of the effects of activity on longevity, several points should be considered before conclusive statements are made. As in the non-cohort investigations, many of these peer group studies used either relatively small numbers in formulating conclusions (22,28,30,32, 38,40) or, in some cases, the number of actual dead proved to be few (19,21,22,28,29,31,32). An additional criticism might be levelled at the peer group selection procedures. As Polednak points out in his study (37,38), the subjects still formulate a select group. His group were chosen from those who had applied for gymnasium lockers at Harvard 27 and hence were interested, to some degree at least, in athletics. Likewise, other studies did not investigate the activity patterns experienced by their cohorts while in school (22,28,31,32,40,44). In either case, college students still provide a select group with apparent and distinct advantages over the general populace in terms of longevity (22,28). Despite these criticisms, several interesting conclusions were indicated. In almost all the papers, summarized in Table 2.16, the distinct advantage registered by the athletes in the non-cohort studies was not present [the exceptions being Prout's study (40) and also that of Paffenbarger et al. (31,32)]. It appeared that minor athletes or honors students had some advantage, though not significant, over major athletes. Second, the hypothesis that college groups were longer lived, as indicated by mortality tables, was borne out thus inhibiting the extrapolation of findings to cover trends in the general population. Examination of data from both sets of studies indicated that while athletics in college may have had no significant positive effect on longevity, they certainly had no detrimental effect, as no investigation showed a significantly greater longevity for the non-athlete controls used. It is important to note, however, that these data apply only to athletics while in college. Apart from the studies based on Montoye's original group, no investigator undertook to measure later activity, either vocational or avocational. Hence, the effects of later life or Lhabitual activity may be the determining factor in the analysis of the longevity-activity relationship. 28 aucovsum huqouo>uca mono—gu< huuuua> unwouwhucz auo>unz Abacuum ucunh any ebnhhzz vs. luccb—auccoa we. Cases—accent A~nv .~Q we a. u cue-u sued-uuo: count-m caucusu< man no humouo>uca amour—~au o_— uo huuouosuca Ona_i-o— noo— wean-acouuqm nououzu< saunas: Acsucoa nucovaum sou-unbuca uuunuo>uca cuo>ucz gnaw: huIcOuou any vuu>un= van luco>~hoccom v:- wucn>~uoccom Ann. .uu go a. I cuuud huquuuo: taupe-h ou».~s»< can no uuuouoouca ona~u-o~ no ‘0 huqouv>uc= Ono~un~o— moan announcement van-noon oo~ oucovaum humouosuca coluouuua huuouosuca Aouv .~I 3.83253-..on 88p N 3 v2.3..— aouzfifcoz 3“ 33m 536:. on: 8.. 2» 33m 535:. on: 9:. $2 an 9.8.3: Asa-on an ou< :«v oytuvaum humouobuca coluOana Ana-09uca A-~ .~¢ voonouon mca oucououuun or non au-um sou«:u«x snug out one ou-um clause“: onau out «emu uo cucuco: Asuco: us uu< cuv oueovaum suuouo>uca coluuuuoa uuuouoauca A-v .uo voouoooa «Nu oocouomuua a: new Ounum savaged! ono— can mac ouuuw couu:u«t ono~ out «mau uo uncucox Anacouuucnuw1602v aucovaum nouoaecuc auguo ve- nouousu< sued ouuo>¢h Aloecquv can loos-u we. cacao: saw. nouo~£u< uuuuuoauca awoc I anon» n.— 5; unusuocou A-uouucouv so: cacao: Aouocosv csn node: vacuuAIno coanioeca ~55 unequAqu coo—raven ~aa~ Away soon Auccuuuucwumucozv oucovaum coluouuua nozuo vs. wouougu< sue: nouns-u ago: Amy Ono—~00 can cuIHUuuoa owouuoo coo Aaw hope u luau» N up sad-nocou Ao~0uucoov :0: noose: oo~.on nuuol< chug-cu mas-Ichou c~o.« nuuol< cacao-u moounOscu anau .-v sauna: "no so: a? an» 203:8 530331.52 53033 33 :3qu ARV nguoon vauoonxu cu unauu< souoacn onouucoo -c.o saucuoauca o—o» n~uqimoa~ oae huuuuoouca oncr «can goon o~a~ ecu unaccouo nasal-bu oncuvcuh nonlaz acqucuaoou Buddhas-09 ponies ecuuuuanem sue—Ila» your heunuuuoo>cu 3.83% 333:3 33;: ”22.8 mo fag-Sm .mpd 2am... 29 culhouuoa >uuouo»«== Aucouuuucuua uncovSum hu—auoausa icozv annoy eo.~ he vouo>uu noun—:uu== Omar .3050 huuuuo>uca o—ur nucuumv nude» mn.¢i¢~.o no coupe-L uo»o~:»< -~ vc- auuaunz woouiaonu «nu ecu venous: Noo~i~cm~ «so— Aoav u50ua oucovaum humouo>uc= coluuuuoa sauncoouca oboe» c.~ an coupe-L noun—gucicoz new vuoum canu£uuz uno— mus Ono euuum ccuu:9u: ono~ can -a~ Aanv cacao Anouu~:uu¢= cucuioec~ coo invade: vuctunx o—a—Iocoq -o~ «account Acouoziicoz A9303: Quad; Acct. anon we. nocuxv nucovaum yuan beauty coluouuoa seq. Anny cola: uaouuogm nouo—&u< heat: vouo>¢h coco—=u< need: abacus. nun Invade: vuubuaz o~o~n¢00~ ss— iuo>~== uuu>u¢= o—o~u¢ao— ouon van xccvo~Oh 02.3.8 05.2.5...— uogz cod action .53qu out: :5 a: :50.— voculaww he“; use: uugcu cascwpcoo--op.~ a_nae 3O Habitual Studies In an effort to better appreciate the role which activity plays in longevity and mortality investigators focused on examination of habitual activity patterns. These studies concentrated primarily on vocational activity and drew their samples from the ranks of the general public in most cases. Pearl (34) initiated the study of habitual activity when he examined some 132 occupations over the course of three years. His data, summarized in Current Opinion, showed it was "very difficult to kill a man by physical hard work before he is 40, occupational and industrial hazards excluded. But after the age of forty is passed our results tell an entirely different story. From 35 to 44 inclusive, the death rate in heavy occupations is 3.9% greater than that for the light occupations . from 45 through 54 it rises to 12.8% greater . . . from 55 through 64 to 18.6%". The main pratfall in Pearl's work may have been in his groups of 'heavy' and 'light' occupations. Examination of his heavy grouping showed many occupations currently found to be abnormally associ- ated with environmentally caused disease states, such as iron and steel workers or miners. Pearl's 1924 investigation was followed almost 30 years later by the first of a series of studies by Morris and colleagues (23,24,25,26). His paper entitled “Morbidity in Relation to the Physical Activity of Work" examined 1930 to 1932 data with special emphasis on deaths in middle age (26). While mortality from all causes combined showed little difference between light and heavy workers (see Table 2.17), seven 31 Table 2.17. Mortality Rates for Men Aged 45-64 Years from All Causes (From Morris and Heady, 1953) Intermediate Age Heavy Workers and Doubtful Light Workers 45-54 10,208 12,561 11,150 55-64 21,042 25,782 23,757 conditions were isolated in which there was greater mortality among middle-aged men engaged in light jobs. These included coronary heart disease, lung cancer, appendicitis, prostate disorders, duodenal ulcers, diabetes, and liver cirrhosis, and were contrasted with the single sig- nificant condition, accidents, in which heavy workers surpassed light workers in terms of fatalities. Focusing on coronary heart disease in London bus drivers and con- ductors, Morris (25) found that despite similar economic and social backgrounds, the more active group of conductors had less coronary heart disease, it appeared later in life, was less severe, and immediate death occurred less frequently. Conductors did have a higher incidence of angina pectoris, the more benign form of the disease. In a follow-up study (27) an attempt was made to classify drivers and conductors in terms of physique at the onset of work as evidence existed which related body build to coronary heart disease. Using uniform size as a guide, drivers were found to have had greater girths when they joined the service. These findings suggested the possibility that if coronary heart disease and somatotype were related, a self-selection factor 32 accounted for the earlier reported discrepancy between drivers and conductors rather than activity levels. In both cases, however, numbers were small and further study was warranted. Morris also reported on postal workers drawn from a 1949-50 data collection (25). By dividing the 35-59 year old workers into three activity categories, it was reported that the experience of the men resembled that of the transportation workers; that is, the active group had less coronary heart disease, it was less severe, and the incidence of angina pectoris was greater. Again the numbers were relatively small. Subsequent studies by Morris (23,24) have produced similar results and have led him to conclude in causal terms "that physical activity of work is protection against coronary (ischaemic) heart disease". Other coronary related studies appeared to confirm the work of Morris. Brunner and Manelis (4) found in Israeli kibbutzim living, those in sedentary occupations had three times the mortality and three times the incidence of myocardial infarctions. Likewise, Kahn (13) found Washington, D. C. postal carriers at an advantage over clerks in his 1963,5tudies. Frank's work (8) on "physical inactivity as a lethal factor in myocardial infarction" also lends support to this contention. Finally, Fox and Haskell (7), in reviewing the relationship between physical activity and prevention of coronary heart disease, sumarized that those presumed more active, both occupationally and non-occupation- ally, had less coronary heart disease. In addition, they state "that for exercise to be of benefit it must be continued throughout one's life". 33 Taylor et a1. (49) confirmed the previous finding regarding coro- nary heart disease and activity, but at the same time found age-adjusted rates for all deaths on the order of 11.83 per 1,000 per year for clerks, 10.29 for switchmen, and 7.62 for section men. His study was based on 1954—56 deaths in white males with a minimumof ten years experience in the railway industry. Clerks were rated as most sedentary, switchmen were intermediate in terms of occupational activity and section men were rated most active. The apparent benefits attributed to occupational activity were based on few deaths in a relatively short period of time, and no control was evident for discrepancies in social level, residency (urban-rural), or the extremely large portion of violent deaths. In 1969, Palmore (33) was investigating factors involved in pre- dicting longevity. His subjects, numbering 268, were volunteers between 60 and 94 years from North Carolina. Evaluation of varied activities, including leisure and total ratings, showed little independent associa- tion with longevity, as measured by number of years lived after testing or an estimate of the number of years the subject will have lived. This study was hampered by small original numbers, a small number deceased, and questionable methods for data collection. A study using 6,928 Californian men and women followed mortality for 5.5 years after the original data collection. Belloc (2) found 371 death certificates and had an 86% return on her questionnaire. The results for men who had engaged in active sports were found to indicate that they had the lowest mortality, one-half of that found in men who only sometimes gardened or exercised. Similar results were reported for women . 34 Karvonen et al. (15,16) reported an investigation concerning 396 elite.skiers in Finland born from 1845 to 1910 in which he followed their mortality up to 1967. He found 325 had died and 14 had disap- peared. His findings indicated the elite skiers survived on average 4.3 to 2.8 years longer than the corresponding male population in Finland. These results parallel Karvonen's initial report (14). However, additional information was obtained on avocational activity by sending out a questionnaire in 1957. Of 90 respondents, 70 surviving skiers still actively skied in the winters of 1955-56 and 1956-57. The most current study was presented in 1977 by Rose and Cohen (45). It sought to examine a large number of longevity predictors obtained from survivors of 500 deceased male Bostonians who had died at the age of 50 and over. In their study they measured degree of exertion both on the job and off (see Figure 2.1). The most significant finding was that avocational activity was a predictor of longevity, whereas vocational activity was not. To this end, those low in vocational activ- ity and high in avocational lived 3.8 years longer than those low in both. In their summary it was pointed out that only four factors were more important longevity predictors than off-job activity. Table 2.18 summarizes the habitual studies presented. With the exception of the work by Frank et a1. and that of Rose and Cohen, all investigations focused on occupational or vocational activity. Few of these studies were channeled towards an investigation of all deaths; many chose only to examine coronary heart disease and the effect activity has on it. Despite this it was apparent the investigators all favored activity as a beneficial factor in terms of improving longevity. 35 FIGURE 2.1 - VOCATIONAL AND AVOCATIONAL PHYSICAL ACTIVITY BY AGE (From Rose and Cohen, 1976) 2.6 I" On Job Activity 2.4 h- '3 I: 2.2 - o o :3 2.0 - --- moderate ———————————— l: 2 1.8 r- q I- \ 2 | 5 h ‘\ ‘ 2’ 1°\ /Off Job Activity 2 1.4 - \ a) ‘0‘ e a. l.2 ’- u\‘ 2 \\O-— E, m — ———Iigm —_____-__.._.:-.o_._ 2 JL 0'0": l l r I I l < 20 20-29 30-39 40-49 50-59 60-69 AGE DECADES 36 acu9u90~s uuu>oucoa I act nuuooueoa anccuaooo> .uOUoucous Away :o:ou huupoucoa - hauouuuc auto-uuuce< sense .0 00-900 uIcOquuosc .auccuucuoa .<.m.= sno- 9:: etc: :2: $12. eouosou suu>uuo< sauna no nous-u _c=ouueuo>< 1c:_=_n .cso— cocoon-u usuva noon vouosah n.->_uc< s>cc= canon no coca-u .oco_u¢ocb< .uuu< sue-a no nausea eocugucu .<.m.= 000— Ann. meet—cs e.e.e .e .eeeeu see As.. ._e vvuoscu Auu>uuu< cocoa“: anew: unoccueo _cco«uccc> .<.n.= ~00. no ue—uch .3 2.5: eouoocu baubuuo< aqua-u: «you: Abuse-co unecuucco> .<.a.= coon we. ac_u0»n guy -oaoo= cones-u nuu>~uo< manned: unto: hu¢36500 nausea .<.m.= oco— 1:. non voc.aleo ~¢ any ._: voucecu uu.>_uo< ocuo¢.c auto: unoccuou Iccuueoc9< .pccouuatc> .<.m.= oca. .0 Jason etecpcn uu_»¢uc< veranda yucca succoaoo —¢:c_ucoo> .<.m.= nea_ gnu. cscx cede Ion-ucu .cuvuauou: as. an. o...::: fiancee; sauo¢uo< once-u: «you: unaccuoo .ocouunoep ~c¢un~ cco— we: uoc:~n ona— Ae~.n~v coupons “aubuuo< mono-«a auto: succouou dune—acoo> equuuun .ano— guano: eaa— Ac~.m~v taboo-u uuubuuo< cocoa—c upon: hua¢9900 «Recuuroca :_QU¢un .nno~ cannot cohabcu suuo¢uo< occocua acne: unaccuou uncouucucv tuneup: anau Acwy :_uue: his. .25 3.9:... 32.2.2 2.»... 3:58.. 1.828..» 33.... e2: 33 1e... cucuweuu vouvaam cs-¢l¢u0~ alga; wank suu>uuu< soc—lean shu==ou auto hosua< 3 3.63m 332m: .3 Ems-Sm ) .w~.~ mpnmh 37 Summar A general review of previous studies seemed to indicate that activity in early life (while in college) had no apparent negative effect on mortality. Studies on later life activity, primarily occupa- tional, implied a favorable relationship with longevity. Few studies have been conducted which explore both vocational and avocational activity patterns throughout life and the relationship with morality. CHAPTER III METHODS OF PROCEDURE The purpose of this investigation was to determine the difference in the activity patterns, as revealed in the 1960 questionnaire responses, of athletes and non-athletes. Second, an examination of activity patterns and their relation to mortality of athletes and non- athletes was carried out. Data Base — Background . The current investigation was drawn from a questionnaire study of living and deceased athletes and non-athletes who were enrolled at Michigan State University prior to 1938. Athletes, defined as one who had earned a major sports letter, were listed in the Athletic Director's office at M.S.U. along with data concerning awards (letters), years of participation, and class (senior, junior, etc.). Non-athletes or con- trols, those not receiving a major sports letter, were selected accord- ing to a stratified random sampling technique from the student director- ies of the school. Specifically, for each athlete who competed in a given year a control was randomly selected from that same year and at the same class level from the student directory. In total 1,129 athletes were matched with 1,129 non-athletes. 38 39 In 1952 a questionnaire (see Appendix A) was forwarded with a return, stamped envelope to each of the 2,258 subjects. Returns from the 1952 questionnaire revealed 628 athletes and 563 non-athletes had responded. 0f the respondees, 122 were deceased. Return percentages on the initial study were 55.6 and 49.9 respectively for athletes and non-athletes. A follow—up study was carried out in 1960 with 1060 questionnaires (see Appendix A) forwarded to 558 athletes and 502 non-athletes. 0f the 92.1% of the athletes responding 52 were deceased whereas 32 deceased non-athletes appeared among the 91.4% of the returned non-athlete questionnaires. Subsequent follow-up studies were carried out in 1968 and 1976 (see Appendix A). The second fellow-up was sent to 942 subjects of which 490 were athletes and 452 were non-athletes. Percentage returns for the study were 96.1% for athletes and 97.3% for non-athletes. Of these 66 athletes and 62 non-athletes were deceased. The most recent questionnaire was mailed to 392 athletes of which 93.8% responded and to 359 non-athletes of which 92.8% responded. Death had occurred in 89 athletes and 78 non-athletes. Figure 3.1 is a summarization of numbers of subjects involved in the initial and subsequent 3 surveys. All information on the return questionnaires was coded and tabu- lated. It was then punched and verified on 80-column IBM cards, and subsequently stored on tape. 40 Table 3.1. Configuration for Numbers in the Michigan State Longevity Study 1952-1976 (From Olson et a1., 1978) Athletes Non-athletes Total 1952 Surveyed 1129 1129 2258 Returned 625 557 1182 % Returned 55.4 49.3 52.4 Deceased 67 55 122 Living 558 502 1060 1960 Surveyed 558 502 1060 Returned 514 458 972 % Returned 92.1 91.4 91.7 Deceased 52 32 84 Living 462 426 888 1968 . Surveyed 490 452 942 Returned 471 440 911 % Returned 96.1 97.3 96.7 Deceased 66 62 128 Living 405 378 783 1976 Surveyed 392 359 751 Returned 368 333 701 % Returned 93.8 92.8 93.3 Deceased 89 78 167 Living 279 255 534 41 Current Stugy_Design Sample Selection The current study considered only those respondents from the 1960 follow-up from whom a questionnaire was returned in 1976 or for whom a death certificate was obtained. Selection of this group provided access to both the vocational and avocational data on the 1960 questionnaire. In addition, during the 16 year interval between the 1960 and 1976 studies, 155 athletes and 140 non-athletes had died thus providing data for analysis of mortality and age at death versus activity pattern com- parisons. Activity Ratings Two activity types were drawn from the data contained in the 1960 questionnaire. A third variable was created from the preceding two. The variables of concern included: 1. 2. Vocational activity pattern. The United States Employment Service Dictionary of Occupational Titles was used for classi- fication of occupations. They were scaled on the basis of retired, sedentary, light activity, medium activity, heavy activity, and insufficient data. Where necessary, for statis- tical purposes, several categories were collapsed into one. Such instances were noted. Avocational activity patterns. In terms of energy output, the scale used for avocational activity patterns was structured to parallel that for vocational activity patterns. Avocational activity patterns were based on an evaluation of several 42 variables on the 1960 questionnaire which included: yard and house maintenance, calisthenics, hobbies, and recreational participation. Blind subjective analysis by Dr. W. Van Huss and Kenneth Stephens was used to determine a rating in each case. The following scale was implemented in rating each subject: Sedentary: A sedentary individual was one who either did minimal yard and house maintenance, participated in a low energy hobby, or walked less than a mile per day. In gen- eral no other or regular activity pattern existed. Light: one was considered having a light avocational activity pattern if they participated in regular house and yard maintenance or regular calisthenics. Persons who were hobby oriented, participated in seasonal (summer only, etc.) activities, or walked more than a mile on a continu- ing basis were considered as 'light'. Medium: regular house and yard maintenance, a hobby orientation, regular calisthenics which included rhythmi- cal endurance exercises, and regular participation in a low energy output sport such as golf were part of a medium .activity pattern. Heayy; a subject was considered to have a heavy activity pattern if all four above activities were participated in regularly or if a high energy output sport such as paddle- ball was part of his regular pattern. 43 3. Combination activity pattern. For each subject a combined activity pattern rating was created. This was used as an indication of overall activity by 1960 respondents. Statistical Analysis Numerical values were assigned each of the previously categorized activity levels, specifically: 1. Vocational activity patterns. Sedentary and retired individ- uals, who obviously had no vocational activity, were assigned to a common category represented numerically "1". Vocational activity category "2" included those individuals rated as hav- ing a light activity Pattern while on the job. The final category represented "3" numerically, was composed of those who were either involved in medium or heavy work occupationally. The combination of these two groups was necessary for statis- tical purposes, as there were too few in each group to enable meaningful analysis. 2. Avocational activity patterns. Those individuals found to be sedentary avocationally were grouped and represented as "1's". The second group, numerically 2, were those whose avocational patterns revealed some light activity. Owing to the relatively few responses coded avocationally as "medium" or "heavy", a “medium plus" category, 3, was created by collapsing the two previous categories into a single one thus enabling a sufficient number for statistical analyses. 44 3. Combination activity pattern. The combined activity pattern, an indication of the overall activity of a subject, was a combination of vocational and avocational categories. The initial category, 1, was composed of those subjects who were sedentary avocationally and either retired or sedentary voca- tionally. Category 2 was a relative measure of those who were not sedentary, that is, they had more activity than the sedentary subjects. Those individuals considered included in this category were those avocationally sedentary and vocation- ally light or those vocationally sedentary or retired and avocationally light. The final combined activity category included all the remaining subjects and was numerically repre- sented 3. This category also was a relative measure of total activity and indicated those individuals within it had partici- pated in significantly more activity than either sedentary or light activity subjects. While not an absolute measure of activity, these categorizations enable comparisons of subjects with progressive increases in activity levels. A "loglinear model" was derived to analyze the relationships among the four following categories: athlete (yes, no), dead (yes, no), decade of birth (1880-1889, 1890-1899, 1900-1909, 1910-1919), and total activity (1, 2, 3). This model functioned by selectively and progres- sively dropping variables, or combinations of variables, from an established model (equation). In this manner, the relative value of the categorical variables to the model could be evaluated. 45 A univariate, one-way analysis of variance was used to examine the relationship of activity level, as recorded from the 1960 question- naire, to year of death between 1960 and 1976. All other comparisons were made using a two-way cross-tabulations. An alpha level of .05 was required to denote statistical signifi- cance in loglinear analyses, and .1 for any subsequent analysis was indicative of significance. CHAPTER IV RESULTS AND DISCUSSION The results and discussion of the athlete-non-athlete comparisons will be presented in conjunction with a brief explanation of the log- linear model. Subsequent sections on deaths between 1960 and 1976 related to activity level in 1960, avocational activity patterns, voca- tional activity patterns, and total activity patterns will follow. .Athlete-Non-athlete Comparisons A loglinear model was derived to analyze the relationships among the four categorical variables, with special emphasis focused on the athlete-non-athlete comparison. Figures 4.1 through 4.5 illustrate these relationships by use of two-way tables. The results of the log- linear model indicated that there was no significant difference between athletes and non-athletes when any single or combination of categorical variables were considered (see Table 4.1). In summarizing Table 4.1, there was apparently no difference be- tween athletes and non-athletes in terms of percentages deceased, percentages in birth decades, or percentages in specific activity cate- gories. Consequently, all subsequent analyses disregarded the athlete- non-athlete variable and focused instead on the relationship of the whole group with activity type and level. 46 47 FIGURE 4.l - PERCENT ATHLETES AND NON-ATHLETES IN VARIED AVOCATIONAL ACTIVITY LEVELS 80)— 7O 60 50 % 4O 3O 2 O I O O VOCATIONAL ACTIVITY LEVELS 90 70 60 5O ‘79 4O 30 20 I O O — N: ACTIVITY LEVEL = N 3 ACTIVITY LEVEL = 75.8 | 247 285 SED 32.2 132 69 LIGHT - ATHLETE [:1 NON-ATHLETE 31 22 MED PLS not significant, :1: .05 FIGURE 4.2- PERCENT ATHLETES AND NON-ATHLETES IN VARIED 47.3 47.6 I92 176 SED RET 39.2 159 124 LIGHT 33.5 ‘ - ATHLETE [:1 NON-ATHLETE 18.9 I 3.5 55 70 MED PLS not significant, a = .05 48 FIGURE 4.3- PERCENT ATHLETES AND NON-ATHLETES IN VARIED TOTAL ACTIVITY LEVELS 80 7O 6O 5O °/o 4O 3O 20 I O O I. I— L I— N = ACTIVITY LEVEL = 36.9 120 I36 - ATHLETE |:| NON-ATHLETE 35.7 34.7 31.2 32.0 145 115 I41 118 2 ‘ 3 not significant, at = .05 FIGURE 4.4- PERCENT ATHLETES AND NON-ATHLETES DECEASED °/a 80 70 6O 5O 4O 3O 20 10 b L— - ATHLETE [:1 NON-ATHLETE I56 140 not aignificant, a = .05 49 0.0.10.0. 000.1000. 000.1000. 000.1000. u madowo 1.55.0 8. B. 8. an. 2. mm or. 3 "z o o. om , . . I on own man an» .6» I oo oxo J on I oo mpmaxEéozfl I 2. 33:: I L om mmo<0wo IP30 amid) Z. mwthIPATZOZ 024 mwkwqxhd bzwommn. and magi 50 Table 4.1. Summary of Logiinear ModeI Analyses Made] Variables Significance 1. a1] 3 2. 234/1 N.S. 3. 23/24/34/1 N.S. 4. 23/24/1 5. 23/34/1 6. 24/34/1 . 7. 23/24/34/12 N.S. 8. 23/24/34/13 N.S. 9. 23/24/34/14 N.S. S. = significant at P< .05 level N.S. = non-significant I. = athlete-nan-athiete 2. = deceased (yes, no) 3. = total activity level (sedentary/retired, Iight, medium plus) 4. = birth decade (1880-1889, 1890-1899, 1900-1909, 1910-1919) SI Deaths Between 1960 and 1976 Related to Total Activity No significant differences were observed in total activity level as reported in 1960 by year of death between l960 and l976 (Figure 4.6). A general trend was apparent which indicated that those deceased towards 1976 had a somewhat higher total activity level in 1960 than those who died shortly after the 1960 questionnaire was completed. Those deceased in the later years of the study were significantly (P<:.l) younger in 1960 than those dying in the earlier years. Generally no sharp effect on activity by impending death was seen. In other words, death occurring in the next year apparently did not inhibit activity in the population examined. Since there was no sharp effect on activity of impending death noted, all deaths recorded between 1960 and l976 were included in the subsequent analysis of activity patterns. Avocational Activity Comparisons Figure 4.7 illustrates graphically the avocational activity com- parisons by birth decade. In the oldest sample group, birth decade 1880-1889, a significant difference was noted between those who were sedentary and those rated most active in terms of percent deceased. Those rated sedentary avocationally in 1960 had a higher percentage deceased than either light or medium plus groups. The finding that the most active appeared to survive the longest might reflect either the small numbers involved in the medium plus activity group or may, in fact 52 (X- -X) HlVBO 1V 39V oh 4.. no... 81 001 Oh! 2r 0.00 060 0.00 41.40 04.0 0.00 0.00 N00 0.00 0.N0 ..¢0 0.N0 0.00 0.00 0.041 ..00 0.0x- nIbdwo P4 004 2402 N 0. 0. 0. h. N. N. hN h. .N 0. n. 0. ON 0. 0. .. "0002.32 0h0. @410. N50. Oh0. 000. 000. V00. N00. 000. "15.400 “.0 C40) b p b — — — F b — b p L P — p — n . I 00 O I 1 O l J and w V x. II I. O x/ /x x .1. xi x/ A x \ I. m. /X 3 I 00.. .1 l OO.N L 00.N 050. 024 000. 2003.00 011F400 024 000. 2. ongamm .._m>wn. >..._.>....04 10.4 mmDQu. 53 0..A 0..V 0..A 0..V no. 29106. 82.89 89108. 82-89 H moqomo :55 non. .mmm .85 .93. "2° 11%“! I. o. .m n... ad. on 3h on ov % O 3 on m V a. oo o 2. on and 2282 747/, Egan. on 352.3% I m mm l 00. 004000 1...”..0 >0 0N0. 024 000. 2003.00 00040000 ...200m.0n. 024 000. 2. 1.0>0... >....>....04 1.420....400>4 141.4 0000.“. 54 indicate the positive effect of high avocational activity levels. The 1890-1899 birth decade group revealed the same general trend, that is, those sedentary had the highest percentage deceased, followed by those rated light avocationally, and then the medium plus avocational group which had the fewest, percentage-wise, deceased. The differences were non-significant and again numbers in the higher activity levels were small. A significantly larger percentage of the 1900-1909 sedentary individuals were deceased than either those rated as light or medium plus. Consideration must be given to the relatively few deceased in each group. Only 27.8% of the sedentary group, 14.8% of the light group, and 7.1% of those medium plus avocationally had died as of 1976. Similarly, approximately 10% of those in each avocational activity group of birth decade 1910-1919 had died by 1976. The relatively small portion of those deceased and the fact that those in this birth decade were quite young between 1960 and 1976 may have accounted for the change in trend. Larger numbers in light and medium plus groups may have accounted for the lack of difference between percentages deceased in each avocational activity group. In general, an apparent trend indicated that increased avocational activity improved one's chances for survival (as indicated by percent deceased). Those most active, especially in older age groups, appeared to have the best survival rates. In addition, Figure 4.7 indicated that the oldest group had the highest percentages deceased as well as having more rated sedentary and fewer rated medium plus in avocational activity. 55 Progressively fewer and smaller percentages of the subjects in other birth decade groups were deceased between 1960 and 1976, and a trend towards increased activity avocationally with decreasing age was apparent. Vocational Activity Comparisons In the 1880-1889 birth decade group a significantly higher per- centage of those rated vocationally as sedentary-retired were deceased than either of the other two groups (Figure 4.8). The medium plus vocational group had the lowest percentage deceased between 1960 and 1976. While an apparent positive vocational activity effect existed, it should be noted that at this advanced age (subjects ranged from 71 to 80 years) few of those studied were still working, consequently, a possible source of bias existed in the relatively small numbers in the light and medium plus activity groups. While there was no significance, a similar trend existed in the 1890-1899 birth decade group. Individuals examined here ranged in age from 61 to 70 years. There was a possibility that a larger number would be more active and grouped in either the light or medium plus vocational category. This may, in part, be reflected in the smaller percentage deceased in the sedentary-retired group and the higher percentages deceased in the light or medium plus groups relative to the sedentary- retired group. For those born between 1900 and 1909, the trend was not consistent with that of the preceding two age categories. In this situation a 56 0..A 0..A 0..A 0..V an. 6.6.10.2 82.68. 82.68. $0.108. u moqomo :56 «So. 98 t mam? 6 mg "20 l . . J y/ /// //u .w 1 o. w... m... / of 1 cu V mo~ 3h 7 1 on 0.6» I 06 % O 1 on .2. . .4. can tum w. l 3 O0 0 1 2. 1 on 8.... 22.8.2 y////, n8 :5... D 1 am 1 Ba 8.55m 5.523%. 1 co. 004000 1.0.0 >0 0.10. 024 000. 2003.00 00040000 .200000 024 000. 2. 1.0)01. >...>...04 1.429.400) 10.? 0000.... 57 smaller percentage, 17%, of the sedentary-retired group was deceased; whereas, 27.5% of the light vocational group and 26.8% of the medium plus activity group were deceased. The reason for this shift may have been related to the fact that the subjects ranged from 51 to 60 years of age in 1960, prior to retirement, when they reported vocational activity patterns. Thus, in the sixteen year period covered by this investigation, more of the subjects were vocationally active and fewer were in retirement. The relatively small percentage deceased in the whole birth decade (less than 30%) may also have been a contributing cause in the deviation from the established trend. The youngest age category considered dealt with subjects of 41 to 50 years of age and, consequently, less than 12% in each activity group were deceased. These small percentages deceased were probably the reason for the lack of significance between vocational activity groups in this birth decade. The general findings suggest a vocational activity effect in the oldest two age groups, though only significant in the 1880-1889 group. In the youngest two age groupings, while no significant differences were observed, the trend appeared to indicate that a larger percentage of deceased were in the light vocational category. As with the avoca- tional activity comparisons, the older groups had a larger percentage deceased, as well as progressively larger percentages of vocationally sedentary-retired individuals. 58 Total Activity Comparisons Examination of the figure for the 1880-1889 birth decade (Figure 4.9) revealed a similar total activity pattern as observed earlier in both avocational and vocational activity comparisons. The relatively small numbers in the two most active groups and advanced age were thought to contribute to the significant difference which existed between the sedentary and the more active groups. In the 1890-1899 birth decade the second most active group had the largest percentage deceased, 52.6%, while the sedentary group was in close proximity with 51.7% deceased. The most active group, medium plus, had roughly 11% fewer deceased subjects than the other groups. All total activity differences in this age category were non-significant. The lower percentage of deceased in the most active group may be related to the combination of reported work in 1960 and also increased avoca- tional activity due to retirement in these 61 to 70 year old subjects. Small percentages deceased in all categories produced non-signifi- cant differences in total activity in the 1900 to 1910 birth decade. Those in the second most active group were found again to have the largest percentage deceased while those in the remaining two groups were virtually indistinguishable in terms of percentages deceased. It was felt that the relative youth of this group, 51 to 60 years old in 1960, was a major factor in the low numbers of "sedentaries" considered. The final birth decade category considered, 1910-1919, once again indicated the second most active group in terms of total activity had 59 0..V 0..A 02A 0..V an. 29.6.2 82.08. 89.08. 68.182 "moaomo :55 we. a .m «m m. 980.». mean "20 a... 3. 0. mo. om on ow % 8 m 3 V 8 a. O 2. oo 8.... .2232 7/42 So... D om $3233 I v.8 1 00. 004000 1.5..0 >0 0.0. 024 000. 2003.00 00040000 #200000 024 000. 2. 40>01_ >.:>_..0< 44.0... 10..» 0000.“. 60 the largest percentage deceased, l6.9%, while the most active group had 7.8% deceased and the relatively sedentary group had only 7.4% deceased. These significant differences were accounted for, in part, by the small percentages deceased. The fact that the more active groups appeared to have larger proportions deceased may well have been a function of the formation of the category or may have reflected the relative youth of this birth category in that few individuals 41 to 50 years of age are sedentary both avocationally and vocationally. The general trends in total activity patterns were much the same as for the avocational and vocational activity patterns. A larger pro- portion of the oldest subjects were deceased, as was expected, and a larger number of the older subjects were more sedentary. Conversely, the younger subjects were more active and fewer were deceased. The only unexplained general trend was that the second most active group in total activity appeared to have larger percentages deceased in all birth decades except the 1880-1889 period. CHAPTER V SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS Summar The purpose of the present investigation was to examine the effects of activity on the longevity of former athletes and non-athletes who had attended Michigan State University prior to 1938. Specifically, subject activity was measured using responses to the 1960 Longevity and Morbidity of Male Graduates of Michigan State Univer- sity questionnaire. Three activity ratings were utilized which included avocational activity, vocational activity, and total activity. Groups selected for comparison under these parameters included, by birth decade, living athletes, deceased athletes, living non-athletes, and deceased non-athletes. A loglinear model was used to examine the relationship between the four categories: athlete (yes, no), deceased (yes, no), activity rating (sedentary)/retired, light, medium plus), birth decade (1880-l889, l890- 1899, 1900-l909, 1910-l919). These results were portrayed by two-way tables. One-way analysis of variance and Chi square were used in subse- quent analyses. Appropriate graphs were constructed to represent per- cent deceased in each activity category. 61 62 Conclusions The results of the investigation have led to the following con- clusions: 1. There was no difference between athletes and non-athletes in any variable considered, i.e., percent deceased, birth decade, and total activity level. 2. No sharp effect on total activity of impending death was noted in the population examined between l960 and 1976. 3. Avocationally, a trend existed which indicated a higher per- centage of those who were most sedentary in 1960 were deceased, while smaller percentages of those active were deceased between l960 and 1976. In short, avocationally, an activity effect may.have been shown. 4. In vocational activity patterns, those most active in l960 appeared to have a smaller percentage deceased between 1960 and 1976. Recommendations l. Emphasis should be placed on collecting data which reflect activity patterns (avocational, vocational, and total) in subsequent years, and comparisons should be made in and based on such life-long patterns. 2. Continued examination of the relationship between longevity and activity is warranted. 3. Follow-up studies should be continued until all subjects are deceased. 63 4. The current study should be expanded to include athletes and non-athletes of a more current era enrolled at Michigan State University. REFERENCES 6. 10. ll. 12. 13. REFERENCES . Anderson, w. G. Further studies in the longevity of Yale athletes. Medical Times 44:75, 1916. . Belloc, N. B. Relationship of health practices and mortality. Preventive Medicine 2:67, 1973. . Breslow, L. and P. Buell. Mortality for coronary heart disease and physical activity of work in California. J. Chron. Dis. 11:421, 1960. . Brunner, D. and G. Manelis. Myocardial infarction among members of communal settlements in Israel. Lancet 2:1049, 1960. . Cooper, E. L., J. O'Sullivan and E. Hughes. Athletes and the heart: an electrocardiographic and radiologic study of the responses of the healthy and diseased heart to exercise. Med. J. Australia 1:569, 1937. Dublin, L. I. Longevity of college athletes. Harper's Monthly Magazine 157:229, 1928. . Fox, S. M. and w. L. Haskell. Physical activity and the prevention of coronary heart disease. Bull. N. Y. Acad. Med. 44:950, 1968. . Frank, C. w., E. Ninblatt, S. Shapiro and R. V. Sager. Physical activity as a lethal factor in myocardial infarction among men. Circulation 34:1022, 1966. . Hartley, P. H. S. and G. F. Llewellyn. The longevity of oarsmen. Br. Med. J. 1:657, 1939. Hill, B. Cricket and its relation to the duration of life. Lancet 2:949, 1927. Hrachovec, J. P. Health maintenance in older adults. J. Amer. Geriatr. Soc. 17:433. 1969. Jokl, E. Longevity of athletes. Physical Educator, 1944. Kahn, H. A. The relationship of reported coronary heart disease mortality to physical activity of work. Am. J. Public Health 53:1058, 1963. 64 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 65 Karvonen, M. J. Problems of training the cardiovascular system. Ergonomics 2:207, 1959. Karvonen, M. J. Sports and longevity. Adv. Cardiol. 18:243, 1976. Karvonen, M. J., H. Klemola, J. Virkajarvi and A. Kekkonen. Longevity of endurance skiers. Medicine and Science in Sports 6:49, 1974. Largey, G. Athletic activity and longevity. Lancet 2:286, 1972. Meylan, G. L. Harvard University oarsmen. Physical Education Rev. 9:552, 1904. Montoye, H. J. Participation in athletics. Can. Med. Assoc. J. 96:813, 1967. Montoye, H. J. Health and longevity of former athletes. In: Johnson, M. R. and E. R. Buskirk (Eds.), Science and Medicine of Exercise in Sport (Second Edition). New York?* Harper and Rowe, p. 366, 1974. Montoye, H. J., u. D. Van Huss and J. u. Nevai. Longevity and morbidity of college athletes: a seven year follow-up study. J. Sports Med. and Phys. Fitness 2:133, 1962. Montoye, H. J., u. D. Van Huss, N. R. Pierson, H. w. Olson and A. L. Hudec. Longevity and Morbiditygof College Athletes. Indianapolis: Phi Epsilon Kappa, 1957. Morris, J. N. Occupation and coronary heart disease. A. M. A. Arch. Int. Med. 104:903, 1959. Morris, J. N. and M. Crawford. Coronary heart disease and the physical activity of work. Brit. Med. J. 2:1485, 1958. Morris, J. N., J. A. Heady, P. A. B. Raffle, C. G. Roberts and J. u. Parks. Coronary heart disease and the physical activity of work. Lancet 2:1053, 1953. Morris, J. N. and J. A. Heady. Mortality in relation to the physical activity of work. Brit. J. Industr. Med. 10:245, 1953. Morris, J. N., J. A. Heady and P. A. B. Raffle. Physique of London busmen. Lancet 2:569, 1956. Olson, H. N. A comparison of longevity and morbidity of athletes and non~athletes. Unpublished Ph.D. Thesis, Health, Physical Educa- tion and Recreation Department, Michigan State University, East Lansing, Michigan, 1972. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 66 Olson, H. N., H. J. Montoye, H. Sprague, K. E. Stephens and N. D. Van Huss. The longevity and morbidity of college athletes. The Physician and Sportsmedicine 6:62, 1978. Olson, H. N., H. Teitelbaum, u. D. Van Huss and H. J. Montoye. ‘Years of sports participation and mortality in college athletes. J. Sports Med. and Phys. Fitness 17:321, 1977. Paffenbarger, R. S. Jr., J. Natkin, D. E. Krueger et a1. Chronic disease in former college students. II. Methods of study and observations on mortality from coronary heart disease. Am. J. Public Health 56:962, 1966. Paffenbarger, R. S. Jr. and A. L. Wing. Characteristics in youth predisposing to fatal stroke in later years. Lancet 1:753, 1967. Palmore, E. B. Physical, mental and social factors in predicting longevity. Geront. 9:103, 1969. Pearl, R. Hard labour shortens life. Current Opinion 77:633, 1924. Polednak, A. P. Previous health and longevity of male athletes. Lancet 2:711, 1972. Polednak, A. P. Longevity and cardiovascular mortality among former college athletes. Circulation 46:649, 1972. Polednak, A. P. Longevity and cause of death among Harvard college athletes and their classmates. Geriatrics 27:53, 1972. Polednak, A. P. and A. Damon. College athletics, longevity and cause of death. Human Biol. 42:28, 1970. Pomeroy, N. C. and P. D. White. Coronary heart disease in former football players, J. A. M. A. 167:711, 1958. Prout, C. Life expectancy of college oarsmen. J. A. M. A. 220:1709, 1972. Pyorala, K., M. J. Karvonen, P. Taskinen, J. Takkunen, H. Kyronseppa and P. Peltokallio. Cardiovascular studies of former endurance skiers. Am. J. Cardiol. 20:191, 1967. Reed, L. J. and A. G. Love. Longevity of Army officers in relation to physical fitness. The Military Surgeon 69:379, 1931. Retzlaff, E., J. Fontaine and w. Furuta. Effect of daily exercise on life-span of albino rats. Geriatrics 21:3:171, 1966. 44. 45. 46. 47. 48. 49. 50. 51. 67 Rook, A. An investigation into the longevity of Cambridge sports- men. Brit. Med. J. 1:773, 1954. Rose, C. L. and M. L. Cohen. Relative importance of physical activity for longevity. An. N. Y. Acad. Sci. 301:671, 1977. Schmid, Von L. Contributions to the study of the causes of death of sportsmen. Sportarzt und Sportmedizin 10:411, 1967. As cited by Olson. Schnohr, 13. Longevity and cause of death in male athletic champ- ions. Lancet 2:l364, 1971. Schnohr, P. Athletic activity and longevity. Lancet 2:605, 1972. Taylor, H. L., E. Klepetar, A. Keys, w. Parlin, H. Blackburn and T. Puchner. Death rates among physically active and sedentary employees of the railroad industry. Am. J. Public Health 52:1697 1962. Wakefield, M. C. A study of mortality among the men who have played in the Indiana high school state final basketball tournaments. Res. Quart. A. A. H. P. E. R. 15:2, 1944. Metropolitan Life. Characteristics of major league baseball players. Stat. Bull. Metropol. Life 56:6, 1975. APPENDICES APPENDIX A QUESTIONNAIRES (1952, 1960, 1968, 1976) 68 NATIONAL STUDY OF LONGEVITY AND MORBIDITY OF ATHLETES IN COLLEGES AN D UNIVERSITIES form A. This Porn is tor graduates who earned a college letter in one or more sports. (Please fill in this Form as Completely and Accurately as Possiue) Date Name at Athlete (please print) Year of Birth Weight at Graduation from Course ll" annm lS LIVING IF ATHLETE lS DREASED Presentaddnaa Age at death ..___ yrs. Cause 0! death stated on death certificate: Primary .. Fluent wekht lhs. Secondary Present general condition ot health it answer is unknown, state the generally accepted cause of (Check one): death Good Fair Was death sudden or lingering Poor Was he married .__..___.... or single .Ilarried w. Person entering lnlormation on this form: (Check one) Name Address Relatinnshig . w.-- -.........__ ...._.-.._ -... Athletic and General Sports History of Athlete Name at Sport High College Amateur Protes- Age School Non-School sional yrs. to yrs. of age yrs. to yrs. of age yrs. to yrs. of age yrs. to yrs. of age yrs. to yrs. of age yrs. to yrs. of age Activity During Adult Life, Excluding Playing Participation In Sports Include vocational and avocational activities Nunherolhoursotphysieaiactivltyuailyoralnostdeily) _ M Vigorous “odessta Mild am. to m hrs. tu-s. hrs. yrs. to yrs. hrs. hrs. 5”: vrs. to yrs. hrs. hrs. h“- Vfl- ‘9 Y“- hrs. hrs. hrs. Military Service Branch 0‘ m u. m b y”. Phyliul activity involved (check): Vigorous Moderate Mild It more than one branch of the Service. name the others and indicate the amount of physical activity Involved Economic Status of Home From Early Childhood Upward Before and Darin! College Ears After College Years Comments (check one) (check one) Saddam __.__ Satisfactory Unsatisfactory ___... Unsatishctory .__._._ 69 Medical History AILM EN T l. infectious aid Contagious Diseases (State age oi occurrence). 2. Childhood rheumatism Growing pains ..._. Chores (State, it possible, age Rheumatic lever _..-_._..__ ot occurrence oi any 1st attack uaniiestations in this 2nd attack group). 3rd attack ‘ Tonsilitis__ Tensile removed— Heart delects (give as complete a diagnosis as podble, such as murmurs. enlargement. irregu- larity. heart tailuse. etc). .— 3. Hypertension (Mention complications such as stroke. coro- nary thrombosia. heart failure. uremia. etc. along with age oi occurrence) 4. Arterio Sclerosis 5. Andria Pectorls Coronary Thrombosis Indicate frequency of clocks Diabetes Peripheral Vascular Disease .._...__. 8. Other Diseases (mention organ or body systesn enacted. and age oi occurrence): Smoking and Drinking Habits Use alcoholic drinks: never moderately excessively Use tobacco: What torn: Bow tnuch Hereditary History Relationship M if Deceased Age Ailment, ii any fie at Death Cause of Death Paternal pandiather Paternal gnndnsother Maternal grandiathes Maternal anndsnother Father “other Brothers (l! Hypertension. Coronary monbosls or Diabetes present in family, please indicate) Doyoudtinltthat participationinathleticsisbeneficial. haun- Morhasnodact‘l Pleasecmt;iiuitioalotprogran.givereasons__ Other comments which will provide additional information on yourparticipationorlaeltoiparticipetlonhssporta. Someonmples: ‘Didnotpertidpateincollegebeoausal wasnolongeranmteurathlete."Didnotparticlpateon adviceoiaphyslcian.”'Dldnotpattidpatebacauaelhad towerkmyvraythroughcollege.”'lwasn'tgoodenou¢ totnaketheteatnf' Etc 70 NATIONAL STUDY OF LONGEVITY AND MORBIDITY OP MALE GRADUATE OF COLLEGES AND UNIVERSITIES Form!.‘ihis!'orlnistorrnenvhodidnoteernaletterheperb (PleaselilliathieforuasCoupldelyondAmmtelyeePo-ifle) Date Nesne ot Alumnus (please print) Year of Birth Weight at Graduation irorn College 11‘ ALUMNUS 18 LIVING 11' W08 13 DECEASED Present address Age at death ......— yrs. Cause oi death stated on death certificate: . Primary Present weight .___... lhe. Secondary - mngmemmfiumde, lianswerisurtknomstatethegenerallyacceptedcauseol (Check one): death Good __.__. Pair ......— Was death sudden _ or lingering Poor _._—_ Was he married ...—.... or single arm-ted __ w. Person entering information on this term: (Check one) Narne Address Relationséal....- ....._-_._____. .-.—.... -.- Athletic and General Sports History of Alumnus Narne oi Sport Hill: College Amateur Proles Age School Non-School sional m to sn- ot nee yrs- to m o! en su- to m at es- m to am» o! Is- m to yrs. at age vrs. to In. of age Activity During Adult Life, Excluding Playing Participation In spot-u . Include vocational ad avocational activities Nunheroihoursoiphyslcalactifltv (Woralmoetdaib) Ase Vigorous Moderate Ilild yrs. to yrs. hrs. hrs. hrs. yrs. to yrs. hrs. hrs. 5"- m to am- hrs. hrs. Im- am» to m hrs. hrs. bre- Military Service 3M 0‘ m N. ’1']. u ______ ”3. Physical activity involved (check): “serous __ Hoderete lild limorethanonebrancholtheService. narnetheothersandindicatetheamountotphyalcal activity involved Economic Status of Home From Early Childhood Upward Fri"? and Durin! College 2a an After Celine Years Comments (check one) (check one) Satirisclory Setislactory Unsatisfactory ———... Umuthetoty _..._-._. 71 Medical mum AILMENT l. lnteeuous and Contagious Diseases (State are d acorn-renal. 1 sebum rheumatism Growimpahas-._.Chorea_._ (State. It possible. age Rheumatic {ever ... ol occurrence 0! any 1st attach Mutations In lhh Dad attack IN”)- Srd attack Tonsilltls... Tonalls removed— Heartdeteets (fiveascompletea diagnosis as podhle. ruch as murmurs. enlargement. irregu- larlty.heart lailrrre. etc).— 15ypertenslon(llant|onoompllcatlonssuehasstrohas.ooro- narythromhoalahearthllure.uremla.etc.alon¢vlth- doocurrenoe) eAreerlo Sclerosis lAndnaPectorls mam—...... 1““!!de Diabetes Peripheral Vascular Disease __... £0therDlseases(mentlonornnorhodysystemafleeted.snd asedoocurrenoe): Smoking and Drinking Habits Use alcoholic drinks: never .._..._-... moderately _...___- excessively .— Use tobacco: What term ..-... How much Hereditary History Relationship ll Living It Deceased 9!. Ailment, ll .33, .3e at Death Cause ot Death Paternal pendlather Paternal grandmother Maternal grandfather Maternal grandmother Father Mother Brothers Sisters (ll Hypertension. Coronary Thrombosis or Diabetes present in tarnlly. please indicate) Do you think that participation in athletics ls beneficlal. henn- lul. or has no edect? Please comment; if critical o! program. give reasons --.- 1 Other comments which will provide additiornl lnl'ormatlon on your partlclpatlon or lack or participation in sports. Someexamples: ’lplayed basketball torhlxhachooldur- lnxalternooneandloracluhlntheevenlmslnlmf'Dtd notplay toothall duringimloroolleayearonaooountol tr-actureoroperallon”"2te " 72 MICHIGAN STATE UNIVERSITY Department of Health, Physical Education and Recreation FOLLOW-UP STUDY OF LONGEVITY AND MORBIDITY OF MALE GRADUATES OF MICHIGAN STATE UNIVERSITY NAME OF ALUMNUS (Please print) PRESENT ADDRESS Date MARITAL STATUS (Check one) PRESENT WEIGHT lbs. RACE _Wh1te _Negro _Other _Marr1ed _S:lngle __W1dowed _D1vorced If your weight has changed more than 15 lbs. within the last seven years, please explain PRESENT OCCUPATION ANY PREVIOUS FULL TIME OCCUPATIONS: From 13__ to 19 1. From l9__ to l9__ 2. From 13___to 13__ 3. From 19__ to 19_ 4. From lQ__ to 12__ SMOKING HABITS: (Please check only those which apply) __Smoke __Do not smoke (If you do not smoke, please disre- gard the remaining questions in this section) Cigarettes: 1. Less than 1/2 pack per day;__ 2. 1/2 to 1 pack per day__ 3. Over 1 pack per day;__ Cigars: 1. Less than 3 per day;__ 2. 3 to 6 per day 3. Over 6 per day: Pipe: 1. Less than 4 bowls per day;__ 4 to 10 bowls per dayé__ Over 10 bowls per day__. UN Chew: COMP 1/4 to 3/4 pack per day__ Over 3/4 pack per day;__ DRINKING HABITS (Please check only those which apply) _Dr:lnk _Do not drink (If you do not drink, please disre- gard the remaining questions in this section) Beer: 1. Occasional bottle__ 2. l to 3 bottles per day;__ 3. Over 3 bottles per day___ 1. Occasional glass other than for religious use__ 2. Daily but less than 1/2 bottle__ 3. Over 1/2 bottle per day__ Whiskey (gin, etc.): Less than 1/4 pack per day___ 1. 2. 3. 4. Occasional glass__ 1 to 3 shots per day__ 4 to 6 shots per day__ Over 6 shots per day__ 73 LONGEVITY OF BROTHERS AND SISTERS: (If any of your brothers and sisters have died in the past seven years, please furnish information requested) Relationship Cause of Death Age at Death Brothers Sisters MEDICAL HISTORY: What ailments have you had in the last seven years? (Examples: Coronary Thrombosis, High Blood Pressure, Cancer, Diabetes, TB, etc.) Age at Occurrence l. 2. 3. 4O FAMILY: Do you have any children? __Tes __No (If your answer is es, please furnish information requested) Sons: Number living Number deceased Age and cause of death Daughters: Number living Number deceased Age and cause of death NON—VOCATIONAL ACTIVITY RECORD FOR THE PAST YEAR: 1. Do you __MOw your own lawn? ‘__Do other yard or house maintenance? (Please describe) 2. Do you __Have a garden? What do you do in connection with this? 3. Do you ‘_;Do any sitting up exercises in the winter? __In the summer? How long does each session last? When was the last time? The time before that? 4. Do you walk.or bike to work? How far? How often? . 5. Do you have any hobbies or engage in other non-vocational work or recreation regularly? EXCLUDING SPORTS (Please list below) Hobby or Activity How Often Do You Participate? a. b. c. d. 74 6. What sports did you engage in regularly during the past summer months? (Please use the list below as a guide) §pprt How Often? When Was the Last Time? The Time Before? a. b. c. d. e. 7. What sports did you engage in regularly during the past winter months? (Please use the list below as a guide) ‘ §port How Often? When Was the Last Time? The Time Before? a. b. c. d. e. f. Angling (fishing) Archery Badminton Baseball Basketball Bicycling Birling Bob-Sledding Bowling Boxing Canoeing Codeball Cricket Cross Country Curling Fencing Field Ball Football Golf Gymnastics Handball Hiking Hockey (field) Hockey (ice) Horseback Riding Horseshoe Pitching Hunting Hurling LIST OF SPORTS ACTIVITIES Ice Boating Jai Alai JuJitsu Lawn Bowling Mountain Climbing Paddle Tennis Polo (horse) Polo (water) Rowing and Sculling Sailing Shuffleboard Skating (ice) Skating (roller) Skeet and/or Trap Shooting Skiing Snow Shoeing Squash Rackets Swimming Table Tennis Tennis Track and Field Trapping Volley Ball Walking Competitive Weight Lifting Wrestling 75 Serial No. SECOND FOLLOW-UP OF THE LONGEVITY AND MORBIDI'IY OF MALE GRADUATES OF MICHIGAN STATE UNIVERSITY Name of Alumnus Date Street City State PERSONAL INFORMATION 1. Have there been any changes in your marital status since IND (our previous follow up)? Yes D No D (If yes to question I. answer A; if no. move on to question 2) A. Please Explain 2. Present weight ...____ lbs. A. Have you lost 15 lbs. or more since 1960? Yes E] NOD (If yes to question A, answer I and 2; if no. move on to question 3) 1. How many times did you lose this much weight? 1-2 times 0 3 or more times 0 2. Any specific reason for these weight fluctuations? 3. Height (in inches) 4. Which of these body type classification do you feel is closest to your body build? Stocky D Medium 0 Slender C] OCCUPATIONAL INFORMATION 5. Are you presently working (job or self employed)? Yes D No C] (If no. answer A; if yes, move on to question 6) A. Have you had a job or been self employed at any time since 1960? Yes D No D (If no. skip to question 7; if yes, move on to question 6) , 6. Answer the following questions about your present occupation or the last job you have had since 1960. A. What kind of work (for example. engineer, teacher, doctor) B. About how much time on the job is spent sitting? Practically all 0 More than half C] About half E] Almost none [J . About how much time on the job is spent walking? Practically all C] More than half 0 About half D Almost none D . About how much walking getting to and from your job? Blocks Miles . What type of transportation do you use to and from your job (check all that apply) Subway C] Bus 0 Car E] Bicycle 0 Others (Please describeL How often do you have to lift heavy weights or carry heavy things on the job? Frequently D Sometimes D Very infiequently (or never) D . How many hours a week do you work on your job? (Hours per week) . How much tension in your job? Great Deal E] Some E] Very Little E] None C] Any responsibility for supervising other workers on the job? Yes D No D (If yes. answer 1; if no, move on to J) [ l. About how many on the average do you supervise? __ ] When did you start on this Job? Year K. Just before this job were you doing the same type of work? Yes. did the same type of work C] . I was on that job years. No. this was my first job D. No. did different type of work C] . If you check this item. please answer the following questions, I, 2. 3, and 4: I. How long did you do this different type of work? _. years. 2. What kind of work was it? 3. On this job did you spend more or less time sitting than your present job? More C] Less 0 Same D 4. Was there more or less walking on this earlier job than on your present (or last) job? More 0 Less D SameC] MO 0 3" 2‘20 t—a 76 LEISURE TIME ACTIVITIES 7. How often do you do the following? (For each activity listed, please check whether you do it frequently, sometimes. or very infrequently.) Frequently Sometimes Very Infrequently (Or Never) A. Take walk in good weather D E] D B. Work around the house or apartment 0 D 0 (painting, repairing, etc.) C. Gardening in spring or summer E] E] C] D. Take part in sports during season U C] D E. If you take part in sports, please indicate what kind of sports and frequency either by the week or year. Frequency Frequency SPORT Per Wk. or Per Yr. SPORT Per Wk. or Per Yr. D Angling (fishing) 0 Judo C] Archery [3 Lawn Bowling __ _' D Badminton D Mountain Climbing ____. ___.. D Baseball [3 Paddle Tennis __ ___— [3 Basketball D Polo (horse) ._..__. _— Cl Bicycling D Polo (water) __ ___.— C] BobSledding C] Rowing 6r Sculling _. _— [3 Bowling (exclude lawn bowlinghere) C] Shuffleboard __ _— D Boxing D Skating (ice) ___.. ___...— D Canoeing [3 Skating (roller) —_ ___— D Codeball D Skiing ...__ ___... C] Cricket 0 Snow Shoeing .___. __ 0 Cross Country 0 Squash Rackets __ ___..... C] Curling 0 Swimming _ ___.. D Fencing B Table Tennis __ ___.. C] Football C] Tennis ___.. ____ C] Golf D Track at Field ___.. __— 0 Gymnastics C] Trapping __ _. D Handball [3 Volleyball ___... __— D Hiking D Weight Lifting ___..—_ ...—___...— 0 Hockey ifield) 0 Wrestling ___... _— D Hockey rice) D Horseback Riding Others: D Horseshoe Pitching C] _— CJ Hunting U —-- 0 Ice Boating D __ [3 lat Alai D __ F. Have you been using an exercise plan at any time during or since 1960? Yes D No C] (If yes to question I“, answer I and 2; if no, answer question C) 2. Give a brief explanation of the exercises and amounts of time spent. 1. Please check how often you used this plan. Frequently 0 Sometimes D Very infrequently C] G. Up till the time you graduated from high school did you live mostly on the farm?[:l How many years? DIET RECALL Or did you live in the city? 0 How many years? 8. List the things you ate and drank yesterday (this should preferably be a week day). When possible. give the specific name of the item, e.g., Fresca or Coca Cola, rather than soft drink; McDonald's hamburger; whole milk, skim milk, half and half. rather than just milk. indicate the amount you ate or drank in terms of cups (200 ml), tablespoons, teaspoons, ounces, numbers and approximate size, e.g., small, large. medium for fruits, vegetables, etc. You may list meats either in ounces or size of pieces: one hamburger patty (3" diameter x 1" thick) weighs 3 01.: an average serving of steak (3" x 3" x V3") weighs 3 02. Be sure to include everything you ate or drank yesterday - candy, liquor. coffee (list sugar and cream, if used), popcorn, potato chips, etc., as well as your regular meals. To help you estimate sizes, a rule is marked off on the edge of this page. 77 Breakfast Morning Snacks Amount or Amount or 1.3mm Wk: Amount or Amount or _ltsL Size.— _Iism L" Dinner Evening SnacLs Amount or Amount or .4321 52L...” __Lsmt AL A. Check date of diet record: Sun.[] Mon.[:] Tues.D Wed.[:] Thurs.[] Frilj Sat.D 8. Did yesterday's meals include any special or unusual event, e.g., party, birthday, anniversary, picnic. etc.? Yes C] No D I. If yes, what was it? C. Does the above represent your usual day's food intake? Yes E] No C] I. If no, how did itdifi'er from your usual intake? D. Check the column which indicates the approximate frequency with which you consume each food. w ' l etc. 9. Doyou drink coffee? Yes D NOD (If yes, answer question a; if no, go on to question 10) A. What is the average number of cups per day? 143 C] {-6 C] 7-9 D more C] SMOKING HABITS lo. Doyou smoke at the present time? Yes D No D (If yes to question IO, answer A and B) A. About how old were you when you first began to smoke? Yrs. old. B. What is the average number of cigarettes cigars pipefuls you smoke per day. (continue on to question 11) (If no to question 10, answer C) r C. Did you ever smoke regularly? Yes D No C] ] (If yes to C, answer I, 2, and 3; if no, move on to question ll) 78 l. About how old were you when you started smoking? Yrs. old. 2. About how old were you when you stopped smoking? Yrs. old. 3. When you were smoking, what was the average number of cigarettes cigars pipefuls that you smoked per day? DRINKING HABITS 11. Do you drink at the present time? Yes D No D (If yes to question 11, answer A) A. Please check the amounts you usually drink. Beer Wine Whiskey (gin, etc.) C] Occasional bottle D Occasional glass other than for religious me D Occasional glass D l to 3 bottles per day D Daily. but less than V: bottle D 3 to 6 shots per day C] over 3 bottles per day C] Over Va bottle per day C] over 6 shots per day (continue on to question 12) (Ifno to question 11, answer B) LB. Did you ever drink regularly? Yes C] No D (If yes to question B, answer I and 2; if no. go on to question 12) 1. Please give the number of years that you drank regularly before you quit Yrs., and why you quit 2. Please check the amounts you usually drank. Beer Wine Whiskey (gin, etc.) D Occasional bottle D Occasional glass other than for religious use D Occasional glass D l to 3 bottles per day I] Daily, but less than 'A bottle 0 3 to 6 shots per day D over 3 bottles per day D Over Vi bottle per day [3 over 6 shots per day HEREDITARY HISTORY 12. If there are any changes in this history since 1960, will you please bring this information up to date, and make any additions or corrections in the data listed below. I A. Father's occupation MEDICAL HISTORY 13. If you have had any of these diseases since 1960, will you please bring this information up to date. Make any correction or addition in the data we listed below. Age at Are you still Are you taking Ailment Onset troubled with medication or this condition? treatment for it? Yes No Yes No High Blood Pressure Angina Pectoris Stroke (Cerebral Thrombosis) Heart Attack (Coronary Thrombosis) Rheumatic Heart Diseaue Cancer Diabetes Tuberculosis Ulcer Liver Ailrncnt Arthritis Cont Other Hlllllllllll 79 Serial No. THIRD FOLLOW-UP OF THE LONGEVITY AND MORBIDITY OF MALE GRADUATES OF MICHIGAN STATE UNIVERSITY Name of Alumnus Date Street City State Social Security Number PERSONAL INFORMATION I. Have there been any changes in your marital status since 1968 (our previous follow-up)? Yes D No D (If yes to question I, answer A; if no, move on to question 2) A. Please Explain 2. Present weight—_lba. A. Have you lost 15 lbs. or more since 1968? Yes C) No Cl OCCUPATIONAL INFORMATION 3. Are you presently working (job or self employed)? Yes D No D (If no, answer A; if yes. move on to question 4) A. Have you had a job or been self employed at any time since 1968? Yes D No C! (If no, skip to question 5; if yes. move on to question 4) 4. Is this the same job you reported on the 1968 questionnaire? Yes D No D (If yes, move on to question 5; if no, answer the following questions A through J. A. What kind of work (for example. engineer, teacher, doctor) B. About how much time on the job is spent sitting? Practically all Cl More than half 0 About half 0 Almost none 0 C. About how much time on the job is spent walking? Practically all Cl More than half El About half Cl Almost none 0 D. Doyouever walktoor from work? YeaD NoCl If yes. how far do you walk? Blocks Miles How many times a year Do you ever bicycle to and from work? Yea D No El If yes, how far do you cycle (both ways)? Blocks Miles Number of times per year E. What type of transportation do you use to and from your job (check all that apply)? Subway 0 Bus El Car 0 Bicycle El Walking 0 Others (Please describe) I". How often do you have to lift heavy weights or carry heavy things on the job? Frequently 0 Sometimes 0 Very infrequently (or never) Cl G. How many hours a week do you work on your job? (Hours per week) B. How much tension in your job? Great deal Cl Some El Very little 0 None E) I. Any responsibility for supervising other workers on the job? Yea Ci No D (If yes. answer 1; if no, move on to .l) l. About how many on the average do you supervise? J. When did you start on this job? Year LEISURE TIME ACTIVITIES 5. How many hours a month do you do the following activities and which months? (List number of hours involved in each activity under the month(a) you participate. Leave blank where not involved.) ACTIVITY 80 Jan. Feb. April June July Aug. Sept. Oct. Dec. Fishing - bank. boat. ice Fishing - wading Archery. target Badminton Baseball . hard. solt Basketball Bicycling - pleasure Tobagganing. sledding Bowling. including lawn Canoeing or rowing JOOOlflO Curling FenCing Gardening Lawn mowing - riding Lawn mowing - power mower Lawn mowing - hand mower Snow shoveling Goll - walking Goll - power can Handball. including paddleball. racket and squash Walking - back packing Walking - cross country Walking - mountain climbing Walking - pleasure Home workshop (carpentry) Horseback riding Horseshoe pitching Hunting - bow and gun Sailing - ice and water Judo. including karate Paddle tennis Rowing. skulllng Shultleboard (not hand) Skating - ice. roller Skiing - downhill Skiing - cross country Skiing - water Snowshoeing Dancing - ballroom Dancing - square Swimming . pleasure Swimming - exerctse Table tennis Tennis - singles Tennis - daubles Volleyball Weight lilting Calisthenics - home Calisthenics - Health Club Others: 81 6. If you have been routinely exercising under a home exercise plan or Health Club plan (commercial. YMCA. Athletic Club. etc.) answer the following questions: A. Number of hours per month .which months (circle): Jan., Feb.. Mar.. Apr.. May. June. July. Aug.. Sept" Oct.. Nov.. Dec. B. What type of exercises? DIET RECALL 7- U“ the thin“ you I“ and Wk yesterday (this should preferably be a week day). When possible. give the specific name 0‘ the item. $8.. Fresca or Coca Cola. rather than soft drink; McDonald’s hamburger: whole milk. skim milk. half and half. rather than just milk. Indicate the amount you ate or drank in terms of cups (200 ml). tablespoons. teaspoons. ounces. numbers and approximate size. e.g., small. large. medium for fruits. vegetables. etc. You may list meats either in ounces or size of pieces: one hamburger patty (3" diameter x 1" thick) weighs 3 02.; an average serving of steak (3" x 3” x is") weighs 3 or. Be sure to include everything you ate or drank yesterday-candy. liquor. cofiee (list sugarand cream. if used). popcorn. potato chips. etc.. as well as your regular meals. To help you estimate sizes. a rule is marked off on the edge of this page. Breakfast Morning Snacks Amount or Amount or m Am :53:— .Lunch Attamoon Snacks Amount or Amount or L!" §_lz- 4m 52:— Dinner Evening Snacks . Amount or Amount or I ltem Size Item Size l l l A. Check datsot‘ diet record: Sun. 0 Mon. D Tues. 0 Wed. 0 Thurs. Cl Fri. 0 Sat. 0 B. Did yesterday’s meals include any special or unusual event. e.g., party. birthday. anniversary. picnic. etc.? Yes D No D 1. Eyes. what was it? C. Does the above represent your usual day’s food intake? Yes D No D 1. If no. how did it differ from your usual intake? D. Check the column which indicates the approximate frequency with which you consume each food. Ice cream ' Cream or custard . in collee. tea. etc. . on . on sweet rolls. etc. low or non-calorie preserves. marmalade 82 E. Do you drink coffee? Yes D No D (If yes. answer questionA; if no. go on to question 8) A. What is the average number of cups per day? 143 D 4-6 [3 7-9 0 more D SMOKING HABITS 8. Do you smoke at the present time? Yes D No D (If yes to question 8 answer A and B; if no. answer C) A. What is the average number of cigarettes_ . cigars_ . and/or pipefulls_ you smoke per day? B. Have you stopped at any time between 1968 and now? Yes D No D If yes. how long did you stop ? C. Did you smoke regularly any time between 1968 and now? YesCJ NoD If no. go on to question 9. If yes. how long? How many cigarettes_. cigars_. pipefulls__did you smoke per day? Dmitmimrrs 9. Do y-zu drink alcoholic beverages at the present time? Yes El No D (If yes to question 9. answer A and B: if no. answer C) A. Please check the amounts you usually drink. leer Wkie Liquor D Occasional bottle 0 Occasional glass other than tor religious use 0 Occasional glass Utto3bottlesperday DDaily.butlessthan'/ibottle Datobshotsperday D over 3 bottles per day 0 Over vi bottle per day 0 over 8 shots per day B. Had you stopped drinking at any time between 1968 and now? Yes 0 N00 If no. go on to question 10. If yes. for how long a period did you stop? C. Did you drink regularly at any time between 1968 and now? Yes D No El 4 Ifno.goontoquestion 10. If yes. for how long a period did you drink? How much? (Please check the amounts.) Beer Wine Liquor U Occasional bottle 0 Occasional glass other than for religious use 0 Occasional glass 0 1 to 3 bottles per day 0 Daily. but less than V: bottle 0 3 to 6 shots per day 0 over 3 bottles per day CI Over ‘A bottle per day D over 5 shots per day HEREDITARY HISTORY 10. As of the inditn'duals listed were still alive. Will this information RELATIONSHIP A. Father’s occupation (when working) MEDICAL HISTORY 11. In 1968 you indicated you had the following conditions. Will you please bring this information up-to-date. Make any correction or addition in the data we listed below. Are you still Are at taking Aliment Age at troubled with med tion or Onset this condition? treatrnent tor it? Yes No Yes High Blood Pressure Angina Pectoris Stroke (Cerebral Thrombosis) Heart Attack (Coronary Thrombosis) Rheumatic Heart Disease Cancer Diabetes Tuberculosis 0000000000000 0000000000000 0000000000000 0000000000000} APPENDIX B STATISTICAL ANALYSES 83 UNIUARIATE 1-UAY ANOVA CASESIYEARDX360-76IKENDATA2tYES ANALYSIS OF VARIANCE OF 89.TOTACT N3 273 OUT OF 298 SOURCE BETWEEN WITHIN TOTAL ETA. 02355 YERRDX N (60) 11 (61) 16 (62) -19 (63) 20 (64) 16 (65) 13 (66) 18 (67) 21 (68) 17 (69) 27 (70) 12 (71) 12 (72) 17 (73) 19 (74) 15 (75) 18 (76) 2 BRAND 273 COHHAND DF SUH OF SORS HEAN 30R F-STATISTIC SIGNIF 16 256 272 ETA-SGR= .0555 HEAN 1.6364 1.6250 1.7368 1.7500 1.6875 1.6154 1.3333 1.7143 1.9924 1.7037 1.7500 2.0000 1.3235 1.7368 1.9333 2.1111 2.5000 1.7546 9.0167 153.54 162.56 (UAR COHP- -.22701 -2 VARIANCE .65455 .51667 .64912 .61842 .76250 .58974 .23529 .61429 .4852? .52422 .93182 .54545 .52941 .64912 .63810 .81046 .50000 .59764 ?CORRELATE V389r10 C88 S=NONE .56355 .59977 (RANDOH EFFECTS STATISTICS) STD DEV .80904 .71880 .80568 .78640 .87321 .76795 .48507 .78376 .69663 .72403 .96531 .73855 .72761 .80568 .79881 .90025 .70711 .77307 .93961 .5243 ZUAR AMONGa -O.) CORRELATION HATRIX CASES=YEARDX360-763KENDATA23YES NI 273 DF- 271 CORRELATION BETUEEN B9.TOTACT AND 10.YEARDX 8 .0100- .1557 .1431 84 UNIVARIATE 1-UAY ANOUA CASES=KENDATA23YES ANALYSIS OF VARIANCE OF 99.ADJAGE N= 298 OUT OF 298 SOURCE BF SUM OF SGRS NEAN 50R F-STATISTIC SIGNIF BETHEEN 16 3344.5 209.03 1.7417 .0391 UITHIN 281 33725. 120.02 TOTAL 297 37069. (RANDOH EFFECTS STATISTICS) ETA= .3004 ETA-80R: .0902 (UAR BUMP: 5.1109 ZUHH AhUNHs 4.003 YEARDX N MEAN UARIANLE STD UEU (60) 11 70.818 162.76 12.758 (61) 19 69.158 100.92 10.046 (62) 21 69.095 186.29 13.649 (63) 22 66.909 84.753 9.2062 (64) 16 73.313 97.563 9.8774 (65) 14 69.286 120.37 10.971 (66) 18 68.611 54.605 7.3895 (67) 22 67.409 142.82 11.951 (68) 21 63.857 138.83 11.783 (69) 29 64.655 123.66 11.120 (70) 5 62.800 136.74 11.694 (71) 13 64.462 94.936 9.7435 (72) 19 64.053 113.50 10.654 (73) 20 64.250 150.20 12.256 (74) 16 62.563 92.662 9.6261 (75) 20 59.800 111.69 10.710 (7’6) 2 59 . 500 .51) . 500 7. 77:32 when“ 2““ 66.07: .34.81 11.172 (DmmHNU TEDRFILATE U=9V~10 S=NONE C=UVS§1$U10£60-76 CORRELATION nHiHlX CA5E5=AENUAIA23lES$YEARuxtaor7o N= 293 DF= 296 RE .05U0= .1136 R9 .0100= .1490 85 TUOUAY CROSS-TABULATION CASES=KENDATA23YES 44. 1.ATHYN AVOCACT (1) (2) N8 786 TOTAL= 827 410 376 COLZ MISS 41 22 19 COLX SED 532 247 285 COLZ 67.7 60.2 75.8 LIGHT 201 132 69 COLZ 25.6 2.2 18.4 HEDPLS 53 31 22 COLZ 6.7 7.6 5.9 TESTS OF INDEPENDENCE STATISTIC SIGNIF DF= 2 N= 786 HAXIHUH LIKELIHOOD 22.864 .0000 CRAHER’S PHI= .1694 CHI-SQUARE 22.560 .0000 CONTINGENCY COEFF= .1670 TUONAY CROSS-TABULATION CASES=KENDATA2£YES 45. 1.ATHYN OCCACT (1) (2) N8 776 TOTAL= 827 406 370 COLZ MISS 51 26 25 COLZ SEDRET 368 192 176 COLZ 47.4 47.3 47.6 LIGHT 283 159 124 COLZ 36.5 39.2 33.5 HEDPLS 125 55 70 COLZ 16.1 13.5 18.9 TESTS OF INDEPENDENCE STATISTIC SIGNIF DF= 2 N= 776 HAXIHUH LIKELIHOOD 5.1692 .0754 CRAHER'S PHI” .0816 CHI‘SQUARE 5.1653 .0756 CONTINGENCY COEFF= .0813 86 TUONAY CROSS-TABULATION CASES=KENDATA21YES 89. 1.ATHYN TOTACT (1) (2) N- 775 TOTAL. 827 406 369 COLZ HISS 52 26 26 COLZ (1) 256 120 136 COLX 33.0 29.6 36.9 (2) 260 145 115 COLX 33.5 35.7 31.2 (3) 259 141 118 GDLX 33.4 34.7 32.0 TESTS OF INDEPENDENCE STATISTIC SIGNIF DF- 2 N= 775 HAXIHUH LIKELIHOOD 4.7480 CHI-SQUARE 4.7484 .0931 .0931 CRAHER’S PHIa CONTINGENCY COEFFB THOUAY CROSS-TABULATION CASES=KENDATA23YES 9. 1.ATHYN DEATHYN (1) (2) N3 827 TOTAL= 827 432 395 COLX YES 296 156 140 COLZ 35.8 36.1- 35,4 N0 531 276 255 CDLZ 64.2 63.9. 64.6 TESTS OF INDEPENDENCE STATISTIC HAXIHUH LIKELIHOOD .40078 -1 CHI-SQUARE .40073 -1 BINOHIAL TEST OF SYHHETRY SIGNIF .8413 .8413 .0000 DP8 1 N= 827 CRAHER‘S PHI= CONTINGENCY COEFF: FISHER EXACT PROBI .0783 .0780 .0070 .0070 .4494 87 TUOUAY CROSS-TADULATION CASES=KENDATA23YES 96. AGEGRPS Nae TOTAL= COLZ MISS COLZ F6T7 COLZ F8T9 COLZ F10T11 COLZ F12T13 COLX TESTS OF INDEPENDENCE 794 827 33 91 11.5 165 20.8 278 35.0 260 32.7 HAXIHUH LIKELIHOOD CHI-SQUARE 1.ATHYN (1) (2) 416 378 16 17 41 50 9.9 13.2 88 77 21.2 20.4 150 128 36.1 33.9 137 123 32.9 32.5 STATISTIC SIGNIF DF= 3 N= 794 2.3031 .5119 CRAHER’S PHI= .0539 2.3049 .5116 CONTINGENCY COEFF= .0538 88 TUOUAY CROSS-TADULATION <1} AGEGRP53F6T7 9. 44.AUOCACT DEATHYN HISS SED LIGHT HEDPLS N8 87 TOTAL= 91 4 69 16 2 COLZ YES 78 4 64 13 1 COLZ 89.7 92.8 81.3 50.0 NO 9 O 5 3 1 COLZ 10.3 7.2 18.8 50.0 TESTS OF INDEPENDENCE STATISTIC SIGNIF NAXIHUH LIKELIHOOD 3.7810 .1510 CHI-SQUARE 5.3240 .0698 THOUAY CROSS-TABULATION 42> AGEGRPS3F8T9 9. 44.AUOCACT DEATHYN HISS SED LIGHT HEDPLS NI 160 TOTAL. 165 5 112 44 4 CDLZ YES 78 4 57 20 1 COLZ 48.8 50.9 45.5 25.0 N0 82 1 55 24 3 COLZ 51.3 49.1 54.5 75.0 TESTS OF INDEPENDENCE STATISTIC SIGNIF HAXIHUH LIKELIHOOD 1.3463 .5101 CHI-SQUARE 1.3002 .5220 CASESIKENDATA23YES DF- 2 NII 87 CRAHER’S PHI= CONTINGENCY COEFFI CASES=KENDATA23YES DF- 2 N' 160 CRANER’S PHIa CONTINGENCY COEFFI .2474 .2401 .0901 .0898 89 THOHAY CROSS-TABULATION <3} AGEGRPSIF10T11 CASESIKENDATAZiYES 9. 44.AUOCACT DEATHYN HISS SED LIGHT HEDPLS NI 262 TOTAL. 278 16 187 61 14 COLX YES 62 8 52 9 1 COLZ 23.7 27.8 14.8 7.1 NO 200 8 135 52 13 COLZ 76.3 72.2 85.2 92.9 TESTS OF INDEPENDENCE STATISTIC SIGNIF DF- 2 NII 262 HAXIHUH LIKELIHOOD 7.3877 .0249 CRAHER’S PHI- .1584 CHI-SQUARE 6.5734 .0374 CONTINGENCY COEFF= .1564 TUOUAY CROSS-TABULATION <4} AGEGRPS3F12T13 CASES‘KENDATA28YES 9. 44.AUOCACT DEATHYN HISS SED LIGHT NEDPLS NI 245 TOTAL- 260 15 141 71 33 COLZ YES 26 4 15 8 3 COLZ 10.6 10.6 11.3 9.1 NO 219 11 126 63 3O COLZ 89.4 89.4 88.7 90.9 TESTS OF INDEPENDENCE STATISTIC SIGNIF DF3 2 N- 245 HAXIHUH LIKELIHOOD .11586 .9437 CRAHER’S PHIa .0215 CHI-SQUARE .11276 .9452 CONTINGENCY COEFFB .0214 90 TUOUAY CROSS-TABULATION <1} AGEGRPS3F6T7 CASESIKENDAT923YES 9. 45.0CCACT DEATHYN HISS SEDRET LIGHT HEDPLS NI 87 TOTALI 91 4 72 6 9 COLX YES 78 4 67 5 6 COLZ 89.7 93.1 83.3 66.7 NO 9 0 5 1 3 COLZ 10.3 6.9 16.7 33.3 TESTS OF INDEPENDENCE STATISTIC SIGNIF DFI 2 N= 87 HAXIHUN LIKELIHOOD 4.6907 .0958 CRAHER’S PHII .2688 CHI-SQUARE 6.2844 .0432 CONTINGENCY COEFFI .2596 TUOUAY CROSS-TABULATION <2} AGEGRP83F8T9 CASESIKENDATA23YES 9. 45.0CCACT DEATHYN MISS SEDRET LIGHT HEDPLS NI 160 TOTALI 165 5 82 52 26 COLZ YES 78 4 43 26 9 COLX 48.8 52.4 50.0 34.6 NO 82 1 39 26 17 COLZ 51.3 47.6 50.0 65.4 TESTS OF INDEPENDENCE STATISTIC SIGNIF DFI 2 N= 160 HAXIHUH LIKELIHOOD 2.5971 .2729 CRAHER’S PHII .1264 CHI-SQUARE 2.5583 ..2783 CONTINGENCY COEFFI .1254 91 TUOHAY CROSS-TADULATION <3> AGEGRPS3F10T11 CASESIKENDATA23YES 9. 45.0CCACT DEATHYN HISS SEDRET LIGHT HEDPLS NI 258 TOTALI 278 20 100 102 56 COLX YES 60 10 17 28 15 COLZ 23.3 17.0 27.5 26.8 NO 198 10 83 74 41 COLX 76.7 83.0 72.5 73.2 TESTS OF INDEPENDENCE STATISTIC SIGNIF DFI 2 NI 258 HAXIHUH LIKELIHOOD 3.7005 .1572 CRAHER’S PHII .1180 CHI-SQUARE 3.5895 .1662 CONTINGENCY COEFFI .1171 -TUOHAY CROSS-TABULATION <4> AGEBRP83F12713 casasnxeunnrgzgyss 9. 45.0CCACT DEATHYN MISS SEDRET LIGHT HEDPLS NI 241 TOThLI 260 19 87 121 33 COLZ YES 26 4 10 14 2 COLX 10.8 11.5 11.6 6.1 NO 215 15 77 107 31 COLZ 89.2 88.5 88.4 93.9 TESTS OF INDEPENDENCE STATISTIC SIGNIF DFI 2 NI 241 HAXIHUH LIKELIHOOD 1.0144 .6022 CRAHER’S PHII .0607 CHI-SQUARE .88828 .6414 CONTINGENCY COEFFI .0606 THOHAY CROSS-TADULATION <1) AGEGRP83F6T7 CASESIKENDATAZ3YES 9. 89.TOTACT DEATHYN HISS (1) NI 87 TOTALI 91 4 56 COLZ YES 78 4 54 COL! 89.7 96.4 NO 9 0 2 COL: 10.3 3.6 TESTS OF INDEPENDENCE HAXIHUN LIKELIHOOD 10.734 .0047 CRAHER’S PHII CHI-SQUARE 13.862 .0010 CONTINGENCY COEFFI THOHAY CROSS-TABULATION <2> AGEGRP83F8T9 CASESIKENDATA23YES 9. 89.TOTACT DEATHYN NISS (1) (2) £3) NI 160 TOTALI 165 5 58 57 45 COLZ YES 78 4 30 30 18 COL! 48.8 51.7 52.6 40.0 NO 82 1 28 27 27 COL: 51.3 48.3 47.4 60.0 TESTS OF INDEPENDENCE STATISTIC SIGNIF DFI 2 NI 160 HAXIHUH LIKELIHOOD 1.9391 .3792 CRAHER’S PHII CHI-SOUARE 1.9281 .3814 CONTINGENCY COEFFI 92 (2) 18 16 88.9 2 11.1 (3) 13 8 61.5 5 38.5 STATISTIC SIGNIF DFI 2 NI 87 .3992 .3707 .1098 .1091 93 TUOUAY CROSS-TADULATION <3) AGEGRPS3F10T11 CASESIKENDAT423YES 9. 89.TOTACT DEATHYN MISS (1) (2) (3) NI 258 TOTALI 278 20 69 92 97 COLZ YES 60 10 15 24 21 COLZ 23.3 21.7 26.1 21.6 NO 198 10 54 68 76 COLZ 76.7 78.3 73.9 78.4 TESTS OF INDEPENDENCE STATISTIC SIGNIF DFI 2 NI 258 MAXIMUM LIKELIHOOD .63515 .7279 CRAMER’S PHII .0499 CHI-SQUARE .64235 .7253 CONTINGENCY COEFFI .0498 THOHAY CROSS-TADULATION <4) AGEGRP88F12T13 CASESIKENDATA23YES 9. 39.131431 DEATHYN MISS <1) <2) (3) n- 240 TOTAL- 260 20 54 33 103 30L: 133 26 4 4 14 3 COLz 10.3 7.4 16.9 7.3 no 214 16 so 69 9s 30L; 39.2 92.6 33.1 92.2 13313 or INDEPENDENCE 314113113 SIGNIF nr- 2 u- 240 unx13un LIKELIHOOD 4.5566 .1025 33433313 931- .1412 CHI-SQUARE 4.7872 .0913 CONTINGENCY COEFFI .1398 ER "'TITJ'I‘UQITJLEMIMWfifijilmfiilflfljiflffliifim‘s