1141‘ ,.. :(f '5. '.'-. ," “2“. u-1':'._'1‘-‘ - u ¢ ,. .‘B .1 u‘ .' -‘ 4* gzflmffifiis :I, .-~ 'fl - I..~.'1"~":\ WM .' ' .- 4. . . 1 ‘11:; 3 ..,:.,’erm - 111‘ n .f‘.‘ NRC-.2". .. ... .., 1 C‘ \ p. 1‘ ¥ _ .3 “13.1.. '1: 1 . .3.” . ”A n. 1 |“"‘" '1'“ L~:’;,‘ :' -lIzvll‘Wl : ' . rim“ ; 1-1: .-.'u"é'-"—'-‘n‘!"" 1.....4. . 1111 .1 1 ‘ . 5;, ..,. . ~|1¢1E.V.Iua1vlqlul" 1' "33;" "11:13 1.? ‘.1 _ JanQ- ‘Lw‘ - =11. Tu-rrmv‘h',‘ 2 w .4 .—. ._v__.11-.«.1v~1..x~w‘~""r “r":n'" w 572'; ‘ ' 2 533“;!:‘1f5?1~2 .. (‘4? ""1“ .1; ;.l ~11. M1111 5“ "" ‘1 1 "- L; 1" :. “.4: .1 "' L :41“, «.4 1. ,. , 33-33ij , 1.1 1. 1‘12!" "12.534141“ .. 1 11 J51 - V . “ ”1"" "33:“; 1 $13!“? I ”5.111155%: 1.1“," .. .~ my; - 11,4" 73’3"“ 4.1.33" 1» . 13.1 «"15, 21‘1”“ $3.17 3?“ “0113“” J” I. A; u». "11 . "1' ’ W...“ ‘J‘:‘ fl“: 4'21; -lfi'1‘._ 5."). . _ «Mt-CQ'TM . 1:: if ‘1‘“: Q 3:11.21" ‘ 51.73:}?- flg ‘ - ihjr-Qfig‘ul . ‘1; '-"s;‘:«.; "%F"r $‘1é1 5?“ 17,... ., ”13.51"“ 43.1 1 ~ warm“: 313??" 2er- ‘ "111m flak—n 7.7;“;" 1. " 1- (.4 1 V . «-:~ {#551321 .3. "m """"‘-’.?~. '3'). a ‘ _ r 11.. "'11:" 5:2? 1 1 “31'2“. 4:11-15:33; "23$ My WE'- s ’ S iééj‘: “13!? ggutgg 1 1:33.511]; . . ~ -q. a“ 1‘ y “I 31%;.7712-1 -.‘. LTW- ~ ~_‘ 1335—— :- 1. «Er-am “W x; .1 Talon-rm.“ 15L- .1. , n-v‘ SITY LIBRARI RIES WWW MICHIGAN STAT WWWWWW 3‘1293 W This is to certify that the thesis entitled INFLUENZA IMMUNIZATION COMPLIANCE AMONG A GROUP OF ELDERLY BLACK AND WHITE RESIDENTS OF LANSING, MICHIGAN presented by Karen Denise Johnson-Webb has been accepted towards fulfillment of the requirements for M. A. degree in Geography W WWI Major professor Dat? MAY L994— 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution LIBRARY Mlchlgan State University PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. DATE DUE DATE DUE DATE DUE Willie I \ \ \ \ \ \— \ \ \__‘ \ \ k [\ \ SU Is An Affirmative Action/Equal Oppo \] x w "unity Iwnuuon www.mh.‘ INFLUENZA IMMUNIZATION COMPLIANCE AMONG A GROUP OF ELDERLY BLACK AND WHITE-RESIDENTS OF LANSING, MICHIGAN BY Karen Denise Johnson-Webb A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Geography 1994 ABSTRACT INFLUENZA IMMUNIZATION COMPLIANCE AMONG A GROUP OF ELDERLY BLACK AND WHITE RESIDENTS OF LANSING, MICHIGAN BY Karen Denise Johnson-Webb Elderly compliance with influenza immunization is low in the U.S. Few studies to determine compliance levels and related factors have been population-based and geographical. This study utilized a population-based sample derived using maps of elderly population distribution to identify census tracts with high proportions of elderly black residents. Two groups of black and white subjects were interviewed concerning their compliance and factors which may relate to it. Factors which were significantly associated with compliance were: gender, physician recommendation of vaccine, history of immunization, and intent to receive an immunization in the future. iii To My Husband Tanya W. Webb whose undying devotion and support enabled me to complete this work. iv Acknowledgements I wish to thank Dr. John M. Hunter for his expertise, guidance, support and encouragement in the preparation of this thesis. I also wish to thank Dr. Gary Manson for advice. Dr. Bruce Pigozzi took me under his wing, lent me a listening ear and offered sage advice. I am grateful to Dr. Judy Olson and the Geography Department and the Center for Remote Sensing for academic and financial support. Harriet Ashbay, Sharon Ruggles, and Judy Slate were always cheerful, helpful and supportive. Harriet was with me as an undergraduate. Marilyn Bria was my champion from the start of my graduate career. I will never be able to adequately express what her encouragement has meant to me. Cathleen McAnneny shared her expertise and her friendship. I am indebted to her. The help and support that Tricia Zuwerink gave me cannot be adequately expressed in words. I could not have completed this work without the love and support of my family. I owe a great debt of gratitude to my husband, Tanya and to Danielle, my daughter. TABLE OF CONTENTS List of Tables List of Figures Chapter One: Introduction and Problem Introduction Influenza Vaccination Demonstration Project Statement of Purpose and Hypotheses Chapter Two: Review of the Literature Chapter Three: Methods Study Area Pre-Test of Survey Instrument The Survey Bias Statistical Analysis Chapter Four: Results and Discussion Descriptive Findings of the Total Sample Chi-Square Analysis Analysis of Factors among Black Subjects Analysis of Factors among White Subjects Analysis of Factors among Female Subjects Analysis of Factors among Male Subjects Analysis of Factors between Male and Female vii. 12 18 18 25 25 28 29 32 32 44 45 48 48 50 Subjects vi Analysis of Factors between Black and White Subjects Analysis of Census Tracts Hypothesis Testing Reasons for Non-Compliance Physicians of Subjects Medical Conditions of Subjects Discussion Census Tracts Chapter Five: Conclusions and Recommendations Recommendations to the Ingham County Health Department Recommendations for Future Research Appendices Appendix Appendix Appendix Appendix Appendix 'n m C1 (3 tn » Appendix Bibliography Introduction Consent Form Immunization Questionnaire Responses to Questionnaire Churches Attended Physicians of Respondents 50 53 55 84 86 95 96 99 106 113 115 117 119 119 120 121 123 126 128 130 vii LIST OF TABLES Table 1. 10. 11. Demographic Data, Ingham County and Lansing, Michigan Selected Demographic Variables Formulas Totals for the Entire Sample Proportion of Frequencies of Medical Checkups Among Subjects and Respective Compliance Distribution of Age, Education and Income Chi-Square Analysis- 2X2 Contingency Table Results Significant Results of Chi—Square Analysis of Factors Related to Compliance among Black Subjects Significant Results of Chi-Square Analysis of Factors Related to Compliance among White Subjects Significant Results of Chi—Square Analysis of Factors Related to Compliance among Female Subjects Significant Results of Chi-Square Analysis of Factors Related to Compliance among Male Page 21 26 30 33 38 43 46 47 49 49 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. viii Subjects Significant Results of Chi-Square Analysis of Factors between Male and Female Subjects Significant Results of Chi-Square Analysis of Factors between Black and White Subjects Compliance Rates of Census Tracts Significant Results of Chi-Square Analysis of Factors among Census Tracts Significant Results of Chi-Square Analysis of Factors by Census Tracts among White Subjects Significant Results of Chi—Square Analysis of Factors by Census Tracts among Black Subjects Significant Results of Chi-Square Analysis of Factors by Census Tracts among Female Subjects Significant Results of Chi—Square Analysis of Factors by Census Tracts among Male Subjects Proportion of Factor in Subjects/% Immunization Compliance in Tract 5 for Total, Race and Gender Groups Proportion of Factor in Subjects/% Immunization Compliance in Tract 13 for Total, Race and Gender Groups Proportion of Factor in Subjects/% 52 52 54 57 58 59 61 64 65 67 7O 23. 24. 25. 26. ix Immunization Compliance in Tract 15 for Total, Race and Gender Groups Pr0portion of Factor in Subjects/% Immunization Compliance in Tract 16 for Total, Race and Gender Groups Proportion of Factor in Subjects/% Immunization Compliance in Tract 36.01 for Total, Race and Gender Groups Proportion of Factor in Subjects/% Immunization Compliance in Tract 37 for Total, Race and Gender Groups Medical Conditions Reported by Subjects 73 76 79 82 98 Figure 10. 11. 12. 13. 14. 15. 16. 17. Map. Map. Map. Map. Map. Map. Map. Proportion of Immunizations by Vaccination Location LIST OF FIGURES Location of Ingham County Health Department Lansing, Michigan Total Distribution of Elderly Residents Census Tracts, Lansing, Michigan Distribution of Black Elderly Residents Study Area Immunization Compliance in the Study Area Proportion of Volunteer Work by Type Reasons for non-compliance Reasons for non-compliance: subjects with history of vaccination Reasons for non-compliance: subjects with no history of vaccination Reasons for non-compliance: black subjects Reasons for non-compliance: white subjects Reasons for non—compliance: female subjects Reasons for non-compliance: female subjects with no history of vaccination Reasons for non—compliance: female subjects with history of vaccination Page 19 2O 22 23 24 35 36 41 87 87 88 88 9O 9O 91 91 18. 19. 20. xi Reasons for non-compliance: male subjects Reasons for non-compliance: male subjects with no history of vaccination Reasons for non-compliance: male subjects with history of vaccination 93 93 95 Chapter One Introduction and Problem Introduction. Influenza is an acute, usually self-limited febrile illness, characterized by such syndromes as: common cold, pharyngitis, croup, tracheobronchitis, bronchiolitis, or pneumonia. Outbreaks of varying severity occur almost every winter and epidemics of influenza are associated with enormous morbidity and lost time from school and work. Substantial mortality due to pulmonary and other complications occurs in groups that are at high risk for complications of influenza. The viruses that cause influenza belong to the family Orthomyxoviridae. These viruses are comprised of three types - A, B, and C - that are related closely in structure, chemical composition, and biological activity. The three types have differing epidemiological behaviors, host ranges and immunological specificity of surface and internal proteins. Type C may resemble A in its ability to infect animals as well as humans. No evidence exists linking Type B to the existence of extrahuman reservoirs of infection. Influenza A is the most important type epidemiologically, and has the widest range of hosts (Kilbourne, 1975). Influenza A has been responsible for pandemics affecting one quarter of the U. S. population (eg. 1957, 1968) (Forbes, 1988). Antigens of influenza are hemagglutinin and 1 2 neuraminidase. These antigens determine virulence. Immunity to the antigens of the influenza virus, especially hemagglutinin, reduces the likelihood of infection and lessens the severity of the disease should infection occur. Antigenic drift within a subtype of influenza may be so marked that vaccination or infection with one subtype may not confer protection against another subtype, even those subtypes that may be distantly related. Major epidemics of influenza continue to occur because of these antigenic drifts. Influenza is transmitted from person to person by the respiratory route. The virus is carried in moisture droplets and influenza has a relatively short incubation period, usually 24 to 72 hours. Influenza, usually accompanied by pneumonia, is one of the ten leading causes of mortality in the U. S.(Pyle and Patterson, 1987). Eighty-five percent of these excess deaths occur in persons who are 65 years and older (Lui and Kendall, 1981). The effects of an influenza epidemic are emphasized by an increase in mortality due to influenza, pneumonia, cardiopulmonary and other chronic diseases which can be aggravated by infection with influenza (MMWR, 1992). Influenza has been associated with 10,000 to 40,000 excess deaths each year (Lui and Kendall, 1981). Hospitalizations for pneumonia and chronic medical conditions have also been shown to increase substantially during influenza epidemics 3 (Perotta et al, 1985; Barker, 1986). The impact of an influenza epidemic can be alleviated by vaccinating all high-risk individuals each year in the fall before the influenza season. Persons over the age of 65 years and those experiencing underlying health conditions such as chronic pulmonary or cardiovascular system disorders, asthma, chronic metabolic diseases, renal disfunction, hemoglobinopathies, and children on long-term aspirin therapy are at an increased risk for complications from influenza infection and are at higher risk than the general population of being hospitalized if infected. These individuals are strongly recommended to take an influenza vaccination each year (MMWR, 1992). Influenza vaccines are about 80% effective in protecting individuals from infection (Kilborne, 1975; Barker and Mullooly, 1988). Howells et al (1975) found that elderly individuals experienced significantly lower rates of bronchopneumonia and mortality when vaccinated against influenza. The antigenic make—up of these vaccines is reviewed every year in order that the most recently circulating strains of the virus may be included. The vaccines usually contain a trivalent combination of one or more subtypes of influenza A and B. Purification has rendered the reaction rate of recent vaccines very low. Nevertheless, certain individuals may experience influenza-like symptoms 8 to 12 4 hours after vaccination, although current vaccines do not cause infection with influenza. Twenty-five percent of those vaccinated may experience a mild local reaction at the site of the injection (Cecil, 1991). Individuals for whom influenza vaccination is contraindicated are those who are allergic upon exposure to eggs or to other components of the influenza vaccine. Delaying vaccination of adults with fever is advisable until the symptoms have subsided. With the exception of the swine influenza immunizations of 1976, the U. S. Immunization Survey reported that each year between 1960 and 1985, only about 20% of high risk individuals were vaccinated against influenza (Fedson, 1990). National immunization rates remain low at present. A survey conducted by the Behavioral Risk Factor Surveillance System in 32 states and the District of Columbia in 1986 reported compliance rates among subject aged 65 years and older to be approximately 32%. The National Institute on Aging reported overall compliance rates for the elderly were 20% (MMWR, 1986). The Influenza Vaccination Demonstration Project. In 1988, the Health Care Finance Administration (HCFA) embarked upon a project to demonstrate the cost- effectiveness of influenza vaccination (Fedson, 1990). The Influenza Vaccination Demonstration Project (IVDP) was developed jointly by HCFA and the Centers for Disease Control (CDC). The goal of this project was to test the 5 cost-effectiveness of vaccinating Medicare enrollees for influenza. The demonstration projects were funded for 1988 through 1989. Three additional sites were added in 1989 through 1990. Influenza vaccines, purchased by the U. S. Government, were distributed throughout the intervention areas and vaccination programs were implemented in physicians’ offices, health department clinics, hospitals, nursing homes, and Health Maintenance Organizations (HMOs). Providers were reimbursed for the costs of administering the vaccine. If this project is deemed cost-effective by an independent source, influenza vaccine will be covered as a benefit under Part B of the Medicare program. Seven counties in Michigan were chosen to take part in the study as either control areas or intervention areas. The intervention area counties included Calhoun, Ingham, Jackson, and Kalamazoo. The control area counties included Kent, Muskegon and Ottawa. Influenza vaccinations were distributed in the intervention area free of charge to Medicare B beneficiaries. In the control areas, vaccinations were distributed using methods in existence at the time of the comparison. The project also included influenza surveillance, and promotion and distribution of influenza vaccine. In Ingham County, in 1991, the trivalent vaccine Flozone, by Connaught, was provided free of charge to Medicare B beneficiaries. This vaccine protected recipients .. If” 6 against Type A Taiwan 1/86 (H1Nl), Type A Beijing/353/89 (H3N2) and Type B Panama/45/90 (ICHD, 1992). Ingham County Health Department (ICHD) established 60 additional immunization clinics throughout the county, distributed vaccine to providers and acted as intermediary in the billing process. Vaccine was also offered five days per week in the ICHD walk-in clinic in Lansing. The ICHD is located in South Lansing at 5303 S. Cedar Street in census Tract 53.04 (Figure 1). Total compliance with influenza immunization for residents of Ingham County who were 65 years and older was approximately 51.5% in 1990 and 68.3% in 1991. A total of 5317 (32%) Medicare B influenza vaccines was administered by ICHD. Other providers administered 10,498 (63%) Medicare B vaccines and nursing homes administered 844 (5%) Medicare vaccines in 1991 (ICHD, 1992). An official in the Ingham County Health Department Disease Control Office told the investigator anecdotally that very few black residents of Ingham County had received an infernza vaccination the previous year. For unknown reasons, they had also had a cancellation of an immunization clinic set up by ICHD in a black church in Lansing. Statement of Purpose and Hypotheses. This study falls within the realm of access to health care and how that access varies geographically across selected census tracts in Lansing. Its purpose is to determine the compliance rates with influenza immunization LANSING CENSUS TRACTS 33.01 32 E l 33.02 1 31.02 I4 34 3 2 5 7 8 9 NL s 4 a] a 5 13 14 65 5 10 16 15 19 12 21 17. 01 24 20 22 47.02 235 23 3602 26 29.01 g 37 27 36.01 28 29.02 ® ICHD 55.01 0 miles 1 Scale Figure 1. Map. Location of Ingham County Health Department 8 of a sample of black and white residents of Lansing, MI and to determine if these rates differ statistically. Subjects’ demographic characteristics, along with their preventive health care and social behaviors will be analyzed to determine if they are associated with compliance with influenza immunization. The sampling design of this work is geographical; a population distribution map of the black elderly population of Lansing was used to determine a major portion of the study area. The investigator was primarily concerned with the compliance behavior of black elderly residents and if that behavior varied among and within census tracts. A population-based sample centered on those census tracts which held greater proportions of black elderly residents was chosen for analysis. The following formal hypotheses (null) will be tested: 1%1 : No difference exists between the compliance rates of black and white elderly subjects. }%2 : No difference exists in compliance rates among census tracts in which high proportions of black elderly residents live and in those in which lower proportions live. }%3 : No difference exists between the compliance rates of those subjects who received a recommendation from their physician to take the vaccine and those who did not. 1%4 : No difference exists between compliance levels of those who visit their physician frequently and those who do not. INS : No difference exists between compliance levels of those who have higher levels of education 9 and those who have lower levels of education. }%6 : No difference exists between the compliance rates of those of higher income and those of lower income levels. 157 : No difference exists between compliance rates of those who are members of an HMO and those who are not. The first hypothesis is based on the findings of other studies which found that compliance among whites differed statistically from that of blacks (Rives and Mooney, 1978; Aho, 1979; MMWR, 1989; Gemson et al, 1988; Sievert et a1, 1989, Stehr-Green, 1990). The alternate hypothesis states that compliance rates between black and white subjects differ significantly, and that black subjects comply at lower levels. The second hypothesis is based on the assumption that if black elderly residents comply at lower levels, then the census tracts in which they make up a large proportion will correspondingly have lower compliance rates than do tracts with lower proportions of black elderly residents. The alternate hypothesis states that compliance rates between census tracts included in the study differ significantly. The third hypothesis is based on the finding that a doctor recommendation of the vaccination has been demonstrated to be positively associated with vaccination compliance in other studies (Ennis, 1976; Rundall and Wheeler, 1979; Aho, 1979; Sievert et al, 1989; Public Health Service, 1989). The alternate hypothesis is subjects who are _. In" 10 recommended by their physician to have an influenza immunization have statistically higher compliance levels than those who are not recommended by their physician. The basis for the fourth hypothesis is that studies have demonstrated that frequent visits to a physician are positively associated with vaccination compliance (McKinney and Barnas, 1989, Barton et al, 1990). The alternate to this hypothesis is subjects that visit the doctor more frequently have higher rates of immunization. The basis for the fifth hypothesis is that Monto et al, 1977; Rives and Mooney, 1978 and MacPherson et al, 1980, demonstrated that level of education was positively associated with immunization compliance. The alternate hypothesis is level of education of subjects is positively associated with compliance. The sixth hypothesis is based on the findings of Monto et al, 1977; and MacPherson et al, 1980, which included a positive association between level of income and vaccination compliance. The alternative hypothesis is level of income of subjects is positively associated with vaccination compliance. Preventive health behaviors have been demonstrated to be associated with compliance and thus, the basis for the seventh hypothesis (MacPherson et al, 1980, Stehr-Green, 1990). HMOs represent themselves as supporters and providers of preventive health care. A logical assumption is subjects who are HMO members have higher compliance levels. The 11 alternate hypothesis is subjects who are HMO members have significantly higher rates of compliance rates than those who are not. 1' if Chapter Two Review of the Literature A formal review of the literature reveals an extensive amount of research on and concern with the compliance behavior of elderly subjects with influenza immunization. Several population-based studies were reviewed. Pyle (1984) compared spatial similarities and differences in influenza vaccine acceptance. His study also analyzed the relationship between city size and acceptance of vaccine. Macpherson et al (1980) surveyed residents of Vermont by mail to determine swine influenza immunization compliance and demographic and other characteristics associated with compliance (total compliance 60%). Elderly rural residents had strikingly lower levels of compliance. Elderly single males and single subjects as a whole also had lower compliance rates in this study. Single subjects had lower compliance rates. Residents with higher education levels reported higher levels of compliance. Higher income residents, especially females had higher levels of compliance. A relationship was also found between preventive health care, history of vaccination for influenza and well doctor visits and higher compliance levels. A similar survey performed by the Behavioral Risk Factor Survey System in 44 states in the U. S. reported that men had slightly higher immunization rates in the 18 to 44 years age group and white 12 AI .fl" 13 elderly respondents had higher compliance rates than blacks (MMWR, 1989) (32% total compliance). Monto and Ross (1979) reported 69.3% compliance with * swine influenza immunization from their population-based telephone survey in Lenawee County, MI. Monto et a1 (1977) also found that subjects who were female, of a higher education and income level, who were married and who were older had higher compliance rates with swine influenza immunization (22% total compliance). Rives and Mooney (1978) conducted a survey by personal interview in the general population of South Delaware and found that nonwhites, those over 45 years and those with less than a high school diploma had lower swine influenza vaccination compliance levels (23% total compliance). Stehr-Green et a1 (1990) found race other than white along with lack of other preventative health behaviors to significantly decrease the likelihood of immunization among persons 65 years and older. They also found that frequent and recent medical checkups were related significantly with compliance (total compliance 32%). Rundall and Wheeler (1979) surveyed senior citizens in Tompkins County, NY by mail about their health beliefs in relation to swine influenza immunization compliance. The most important determinants of compliance were perceived susceptibility to the disease and perceived risk of receiving the immunization (total compliance 72.1%). Ennis et al (1976) surveyed subjects by telephone to determine 14 levels of compliance and factors associated with it. The most common reason for noncompliance was lack of a perceived need for it. Non-compliance was also significantly associated with lack of a physician recommendation for the vaccine (26.9% overall compliance). Several studies examined physician behavior and attitudes toward influenza immunization of their patients. McKinney and Barnas (1989) found that elderly patients who had been vaccinated the previous year and those who made multiple visits to physicians had a higher rate of immunization compliance (47% vs 39%). Rates were unrelated to age, gender or race. Physician uncertainty about the contraindications of the vaccine and its efficacy were found to be factors in the rate in which they offered vaccine to their patients. A telephone survey of physicians in New York, NY found that physicians whose caseloads consisted of a majority of minority patients recommend influenza vaccination significantly less than do physicians with majority white patient caseloads, (Gemson et al, 1988). Factors associated with this difference included: physician training and education, socio—economic level of the patients and time spent with patients. In a hospital-based study in the United Kingdom (UK), Tranter (1976) found that physicians received influenza immunization at nearly non-existent rates and may not, as a result of their own attitudes toward the vaccine, have offered it to their patients. Nicholson et al (1987) I» 4‘ "‘ 15 surveyed general practitioners in the UK and found that clinics that had a policy in place for the administration of influenza vaccine had significantly higher rate of compliance among their patients (19.5% total compliance). The Public Health Service (1988) conducted surveys in independent living senior citizen housing complexes in Fulton and DeKalb Counties, GA, and found that the most important factor associated with vaccination status was recommendation of vaccine by a health care provider (overall compliance 55.0%). Whites also had significantly higher rates of compliance than did blacks. Several other investigators studied subjects found at senior citizen centers or those who resided in senior living complexes. Aho (1979) interviewed Black and Portuguese subjects in Providence, RI. who were residents of a senior citizen complex about their immunization status and health beliefs. Efficacy and safety of the vaccine, gender and race, along with planning to be immunized in the future were significantly associated with immunization status (52-56% overall compliance). Sievert et a1 (1989) surveyed residents of senior citizen housing communities in person and by telephone about their immunization status and other demographic variables. Black respondents had lower immunization levels than did whites. Physician recommendation was also an important factor significantly associated with compliance in this study population. Compliance rates for institutionalized elderly patients 16 were determined in several studies (Setia et al, 1985; Patriarca et al, 1985; Siewert et al, 1988; Bloom et al, 1988). Rates ranged from 33-87% compliance. Factors significantly associated with compliance with influenza immunization for institutionalized subjects included: familial consent, educational promotion, physician recommendation, and the facility having an immunization program in effect. A large proportion of the studies reviewed are concerned with increasing compliance with influenza immunization among elderly patients through various means (Larson et al, 1979; Mullooly, 1987; Hutchinson, 1989; Leirer et al, 1989; McDowell et al, 1990, Barton et al, 1990; Wakefield, 1990; and Hutchinson et a1, 1991). Vaccination rates were generally increased by these efforts. Barton found a significant difference between physician visit rates among compliers and non-compliers. Wakefield found that educational promotion did not increase compliance levels significantly in his study population. Hutchinson and Shannon found no significant difference in compliance levels between those who had and had not been reminded by postcard to receive a vaccination. The generally high compliance rates seen in several of the swine influenza studies are attributed to the extensive efforts of the National Influenza Immunization Program (suspended in December, 1976). This program was a large scale campaign in the U.S. to immunize individuals against 17 the perceived threat of the swine flu in 1976. The literature points to several factors significantly associated with influenza vaccination compliance. These factors include race other than white, higher education and income levels, vaccination the previous year, physician recommendation of vaccination and multiple visits to the physician. In the studies where gender was found to be a significant factor, females had higher levels of compliance than did males. History of taking preventive health care measures were found in some cases to be significantly associated with compliance. Subjects’ fear of the risks of the vaccine versus its efficacy affected compliance levels in several studies. A great range of compliance levels were seen in the various studies. The swine influenza immunization studies showed high rates of vaccination in general (22.0 to 69%). Several of the more recent studies also found varying rates of compliance (19.5 to 56.0%). Several of the samples could not be considered representative of the entire elderly population because of their sample selection and indeed some of the studies included subjects from all adult age groups. In most cases compliance levels remained between 20% and 30%. Chapter Three Methods Study Area. Lansing (pop = 122,700) is the capital of Michigan and is situated in the northeast corner of Ingham County (Figure 2). The population of Lansing comprises approximately 45% of the population of Ingham County. The elderly population (aged 65 years or older) of Lansing is 12,443, and comprises 10% of the total city population (Figure 3). The total black population of Ingham County is 21,905 and 79% of the total black population of Ingham County resides in the City of Lansing, Ingham County. The black elderly population of Lansing is 1,165 (9% of the total black population and .1% of the total population of Lansing). Table 1 shows detailed demographic data on Ingham County and Lansing, MI. Lansing is divided into 50 census tracts (Figure 4). Census tracts were considered to be eligible for consideration in the study if 10% of the elderly population which resided in that tract were black (Figure 5). Of 15 such census tracts, seven were selected for field study (Figure 6). All census tracts in which over 25% of their elderly population were black were selected for field study (Tracts 5, 15, 16, 36.01). A well known government subsidized apartment complex for social security recipients, Riverfront Apartments, is situated in Tract 13, therefore it was selected for field study. Tract 37 was selected at 18 l9 Figure 2. Map. Lansing, Michigan. LE NOW! no . O K N. w COUSTA [@313 . E 71 . ‘0‘! 3 WWI.” 3 27 UNNISONVILLE '< cum no ’ a E as 2 ‘ Par. 27 :6. c 5 cannery, 2; ‘- R‘ND “(”0“ cumo~ co. 1? ”—- w°~ °°' ., 1 ~ - - ... Mun N a” ' . j; . Q .. i‘ ' ' 69 : d I‘Vd 4 ,5- 69 Q Caumy. at G ND .rPGE SAGINAW HWY . . u. 3... . ‘ "SIGMA" ‘ at: Leann: 27 ELIIA I 43 “ uncanny AV ' I Sal-Asian SING $1.10: HWY gen)“ ‘9‘ a“ NEIDA .' E O , . . . ~o ENTER In. none nwv ’9 " m. " p: h "Vin Cut." a I __ . . 0| OKEMO ’ 4V *. : M 1 ". .2. u '13.. x a ‘ ‘Q h t 27 -- _-‘ . ‘ . . g ‘. . U 0‘ a @b 5 L NS, G 10131 - a MERIDIAN _. .. _ I "' "In V0"; , g; $5 . Q 09 ~ .2 . a 2 If I ‘ 127 g g ombfiglte J.) . 5 f *° \ now no \HOLT ' Fons: 1.1.5 ‘ \ ‘ NAorth Nun 20 ELDERLY POPULATION DISTRIBUTION Proportions of Elderly Residents .................. 1:1 mama... .................................... [:1 1%1050/0 a 6%1010‘7. ° """ ‘ 11%toi5% 3a,. 16%1020% - 20%+ Source: U. S. Buow d Cor-Ia, 190 Figure 3. Total Distribution of Elderly Residents 21 TABLE 1 DEMOGRAPHIC DATA INGHAM COUNTY AND LANSING, MICHIGAN 1H .. Ingham County Lansing Total Population 281,912 122,700 Total Black Pop 27,837 21,905 (% of total pop) (9%) (18%) Total White Pop 223,705 91,545 (% of total pop) (79%) (75%) Total Population 65 yrs + 24,376 12,443 (% of total pop) (9%) (10%) Total Male 65+ 9,342 5,848 (%pop 65+) (38%) (47%) Total Female 65+ 15,034 6,596 (%pop 65+) (61%) (53%) Total Black Pop 65+ 1,203 1,165 (%pop 65+) (.4%) (9%) Male 476 425 (%Black 65+) (40%) (37%) Female 727 730 (%Black 65+) (60%) (63%) Total White Pop 65+ 22,779 10,780 (%pop 65+) (8%) (60%) Male 11,077 3,922 (%white 65+) (49%) (36%) Female 11,702 6,858 (%white 65+) (51%) (64%) 22 LANSING CENSUS TRACTS LH3301 32 g I 33.02 1 31.02 ‘4 34 3 2 5 7 8 9 NL .- 4 § 6 13 14 65 10 9 16 15 19 12 21 17.01 24 20 22 17.02 25 23 36.02 26 29.01 g 37 36.01 27 28 29.02 51 53.03 53.04 I 52.01 § £53.02 . . . l—_—_._l Scale Figure 4. Map. Census Tracts, Lansing, Michigan 23 BLACK ELDERLY POPULATION DISTRIBUTION Proportions of Elderly Residents Who are Black [:3 o to 9% Black 1010 25% Black 0 miles 1 26 to 50% Black ‘: 51 to 75% Black Scale m Over 76% Black Source: U. S. Buoau 01 090315on Figure 5. Proportions of Elderly Residents Who are Black 24 STUDY AREA Legend StudyTracI I Riverfront . Capital Commons l° ".11.. 'l . South Washington Park 5an Sauce:U.SI BtnauotConwsJflO Figure 6. Map. Study Area 25 random for field study. Table 2 shows selected characteristics of the census tracts included in the study area. Pre-test of the survey instrument. The survey instrument was pretested by conducting 5 interviews. Four of the interviews were conducted at a local senior citizens center. The fifth interview was conducted with the help of a Lansing resident who is not a senior citizen. Problems with the form and content of several questions became evident during the pretest. After the pretest, the investigator decided to include a question concerning whether the subject lived alone. One of the pre- test subjects was married and her spouse lived in a nursing home. Also, the questions concerning income were modified to include four sources. The Survey. Subjects were contacted through a door-to-door survey. The investigator knocked on every door on every street of a selected tract. If the resident was at home and reported being aged 65 years or older, he or she was asked if they would be willing to participate in a short interview (Appendix A). If the subject was willing, he or she was asked to read and sign the consent form (Appendix B). The interview began after the reading of the consent was completed (Appendix C). Appendix D shows the percentage of responses to the survey questions. 26 TABLE 2 SELECTED DEMOGRAPHIC VARIABLES for the Study Area Variable I Tract # 5 13 15 16 36 01 37 Total Population 2,070 1,620 2,583 1,273 4,312 6,508 % Female 52.3 51.8 52.9 53.3 51.2 51.9 % Black 43.4 15.5 64.8 77.2 43.9 14.2 %65 yrs and + 8.0 14.0 11.0 14.0 5.0 13.0 Household Income Median ($) 23,608 8,332 10,147 35,921 37,759 23,596 %Retirement Income 14.0 18.7 11.0 13.6 13.2 17.5 Mean($) 9,142 3,637 4,777 9,018 9,478 6,330 %SSI Income 22.6 43.4 22.3 33.6 15.5 25.2 Mean($) 7,338 5,413 6,912 10,712 7,036 7,614 Education < 12 yrs 27.7 37.1 43.3 27.1 20.4 25.8 12 yrs 36.6 32.9 13.3 11.4 29.9 31.2 13 -15 yrs 29.0 22.4 24.2 40.2 35.7 13.1 16 yrs and + 6.0 7.6 19.2 21.3 14.0 10.7 All Residents % 2, the X? statistic is included in the written results (not in the tabulated results). The strongest association with compliance with influenza immunization was found based on past history of influenza immunization (n=194) (Relative Risk [RR]=3.74, 95% CI=2.18, 6.43, p<.000). [Three subjects did not know if they had received a past vaccination for influenza and were eliminated from consideration in this test.] Those who had an influenza vaccination in the past were more likely to have been vaccinated in 1991. Respondents whose physician had recommended that they 45 be vaccinated were more likely to have been vaccinated in 1991 (RR=2.18, 95% CI=1.62, 2.95, p<.000) (Table 7). Intent to have an influenza vaccination in the future was significantly associated with compliance (p<.000, df=2, JG=113.55). Medical checkups were significantly associated with vaccination compliance (RR=1.52, 95% CI=1.02, 2.27, p=.03). The number of checkups a subject had in a year was associated significantly with compliance (p=.05, df=3, Xfi= 7.75). Chi-square for linear trend was performed on the number of check-ups variable and no significant trend was found. Females had a significantly lower rate of immunization compliance (51.1%) than did males (68.9%) (RR=.74, 95% CI: .59, .94, p=.03). Analysis of Factors among Black Subjects. Table 8 shows significant findings among black subjects. Among black subjects (n=79, 6 subjects did not know if they had a past immunization and were eliminated from this test), history of past immunization was significantly associated with immunization compliance (RR: 2.80, 95% CI=1.80, 4.36, p<.000). Black subjects who had been immunized in the past were more likely to have been vaccinated in 1991. Black subjects who had a recommendation of the vaccine from their physician had higher compliance levels (68.8%) than those who did not (RR=2.40, 95% CI-1.51, 3.82, p<.000). Black subjects who intend to receive an influenza immunization in the future (n=75) may have higher compliance rates (81.4%, RR=13.02, 95% CI-3.30, 50.21, 46 TABLE 7 CHI SQUARE ANALYSIS 2 X 2 Contingency Table Results ---—u-—_--——-----—---~-—u—w--—~~--—m-——-_——-—-—-—--~----—---— FEMALE 68.5 51.1* .74 .59, .95 BLACK 42.6 48.8 .80 .61, 1.04 HOUSE(N=193) 70.1 57 2 1.04 79, l 37 DR RECOMMEND. 52.8 76.0* 2.18 1.62, 2.95 PAST SHOT(N=194) 70.1 72.3* 3.74 2.18, 6.43 CHECKUPS 79.7 60.9* 1 52 1.02, 2 27 PRIVINSURANCE 68.0 57.5 1.05 .80, 1.37 MEMBER HMO 12.2 70.8 1.30 .97, 1.73 EMPLOYED 8.1 62.5 1.11 75, 1.66 RETIRED . 82.7 56 8 1.02 73, 1.41 HOMEMAKER 12.7 48 83 54, 1.27 LIVE ALONE 48.2 54.7 94 .73, 1.20 DRIVE? 59.4 57.8 1 05 .82, 1.35 CAR 62.4 59.8 1 17 89, 1.52 DRIVER 50.3 55.6 96 .75, 1.23 BUS 18.3 52.8 92 .66, 1.29 SPECTRANSP 21.8 51.2 88 .64, 1.21 ATTEND CHURCH 68.5 57.5 1 05 .80, 1.37 ATTEND SR CTR 16.2 56.3 99 .71, 1.38 SOCIAL CLUB 25.9 54.0 94 .70, 1 26 VOLUNTEER 39.1 55.3 96 .75, 1.24 AARP MEMBER 54.3 59.4 1 11 .87, 1.43 PENSION 64.0 56 8 1.01 .78, 1.30 INVESTMENTS 49.7 59.8 1.12 .87, 1.43 EARNINGS 13.7 70.4 1.29 .98, 1.71 OTHER 5.6 72 4 1.31 .89, 1.92 * Significant @ pg .05 47 TABLE 8 Significant Results of Chi-Square Analysis of Factors Related to Compliance among Black Subjects DR RECOMMEND. 60.0 68.8* 2.40 1.51, 3.82 PAST SHOT(N=79) 67.1 69.8* 2.80 1.80, 4.36 FUTURE SHOT(N=75) 53.2 81.4 13.02 3.30,50.21 * Significant @ p< .05 Future shot result is unreliable: CI very wide 48 p<.000). The very wide confidence interval indicates that this result is unreliable. Analysis of Factors Among White Subjects. Table 9 shows significant differences in compliance rates among white subjects. White subjects who reported receiving an influenza immunization in the past (n=112) had a significantly higher rate of compliance (73.8%, RR=2.95, 95% CI=1.53, 5.68, p<.000). White subjects who had a recommendation from their physician to take the vaccine had a higher compliance rate (82.1%, RR=1.95, 95% CI=1.41, 2.71, p<.000). White subjects who intended to have an influenza vaccination in the future (n=108) had a higher compliance rate (94.8%, RR=7.71, 95% CI=3.38, 17.57, p<.000). The wide confidence interval indicates the significance of this result is unreliable. Analysis of Factors among Female Subjects. Analysis of factors among female subjects (n=135) yielded significant differences based on the following factors: past immunization (64.2%, RR=4.94, 95% CI=2.46, 9.93, p=.000); doctor recommendation of vaccine (51.5%, RR=2.19, 95% CI=1.49, 3.22, p=.000) and membership in an HMO (11.1%, RR=1.68, 95% CI=1.23, 2.31, p=.04) (Table 10). Intent to receive an immunization in the future was significantly associated with compliance (53.9%, RR=10.60, 95% CI=4.57, 24.62, =.000), however the wide CI indicates that this result is unreliable. 49 TABLE 9 Significant Results of Chi-Square Analysis of Factors Related to Compliance Among White Subjects DR RECOMMEND. 49.6 82.1* 1.95 1.41, 2.71 PAST SHOT(N=112) 75.0 73.8* 2.95 1.53, 5.68 FUTURE SHOT(N=108) 62.0 94.0 7.71 3.38,17.57 -—--—--—-—-----——--—--————-———I----—-—-—-——-—-————-——-—--m--— * Significant @ p< .05 Future shot result unreliable: CI very wide TABLE 10 Significant Results of Chi-Square Analysis of Factors Related to Compliance among Female Subjects DR RECOMMEND. 51.5 69.6* 2.19 1.49, 3.22 PAST SHOT(N=134) 64.2 72.1* 4.94 2.46, 9.93 FUTURE SHOT(N=128) 53.9 89.9 10.60 4.57,24.62 HMO 11.1 80.0* 1.68 1.23, 2.31 * Significant @ pg .05 Future shot result unreliable: CI very wide 50 Analysis of Factors among Male Subjects. A significant association with compliance among male subjects was found based on physician recommendation of vaccine (57.4%, RR=2.09, 95% CI=1.32, 3.33, p=.0003) (Table 11). Intent to receive a future immunization was significantly associated with compliance (74.5%, RR=6.15, 95% CI=1.69, 22.29, p=.000), however the extremely wide confidence interval renders this result unreliable. Analysis of Factors between Male and Female Subjects. Because vaccination compliance was found to be significantly associated with gender in this study, all the variables were analyzed with.X? analysis between male and female subjects (Table 12). Many significant associations were found. Male subjects had significantly higher levels of past compliance and intent to receive an immunization in the future. Male subjects were significantly more likely to live in a house and were less likely to live in SSI apartments than were female subjects (X?=6.91, df=2, p=.03). Male subjects were more likely to have retired from an occupation 06:6.14, df=1, p=.01). Female subjects were more likely to be in the low income group and less likely to be in the higher income groups than male subjects (n=162, analysis was performed without considering those subjects who refused to answer the income question) (X?=14.14, df=2, p=.0008). Male subjects were significantly more likely to have a pension as a source 51 TABLE 11 Significant Results of Chi-Square Analysis of Factors Related to Compliance among Male Subjects DR RECOMMEND. 57.4 88.6* 2.09 1.32, 3.33 FUTURE SHOT(N=55) 74.5 87.8 6.15 1.69,22.59 * Significant @ pg .05 Future shot result unreliable: CI very wide 52 TABLE 12 Significant Results of Chi-Square Analysis of Factors between Male and Female Subjects HOUSE 83.3 65.4 .01 SENIOR APTS. 16.7 34.6 .01 PAST SHOT 85.0 64.7 .006 FUTURE SHOT 67.2 51.1 .02 RETIRED 93.0 78.2 .01 INCOME(N=160) LOW 26.5 55.0 .003 MEDIUM 59.2 37.0 MED—HIGH 14.3 7.2 PENSION 78.7 56.3 .004 INVESTMENTS 65.5 42.2 .004 EARNINGS 24.6 8.9 .006 CAN DRIVE 86.9 46.7 .000 OWNS AUTO 86.9 51.1 .000 HAS DRIVER 29.5 60.0 .000 SPECTRANSP 11.5 26.7 .03 %MARRIED 62.3 25.2 .000 %SINGLE 16.4 16.3 %WIDOWED 21.3 58.5 LIVES ALONE 29.5 57.0 .0006 %FAMILY=2 55.7 31.9 .002 %FAMILY=3+ 14.8 11.9 AGE 65-69 yrs 47.5 20.7 .001 70—74 yrs 13.1 28.9 75-79 yrs 24.6 21.5 80-84 yrs 6.6 14.1 85+ 8.2 14.8 SENIOR PROGRAM 6.6 20.7 .02 MEMBER AARP 67.2 48 1 02 53 of their income (X?=9.49, df=1, p=.002) and also were more likely than female subjects to have wage earnings (x?=7.45, df=1, p=.006) and savings or investments as a source of income (X2=8.25, df=1, p=.004). Male subjects were much more likely to be able to drive (X2=26.49, df=1, p<.000) and own a car (X2=21.38, df=1, p<.000) than were female subjects. Males were also less likely to use Spectran or Dean Service or have someone who drove for them (x?=4.81, df=1, p=.03). Marital status differed significantly by gender 08:28.14, df=2, p<.000). Female subjects were more likely to be widowed (58.5%) than male subjects (21.3%). Male subjects were more likely to be married (62.3%) than female subjects (25.2%). Female subjects were more likely to live alone (57.0%) than were male subjects (29.5%). Age differed significantly by gender (X5=18.38, df=4, p=.001). The majority of male subjects (47.5%) were aged 65- 69 years. The largest proportion of female subjects (28.9%) were aged 70-74 years. Male subjects were significantly less likely to attend a senior citizen center (6.6%) than were females subjects (20.7%, X?=5.16, df=1, p=.02) and more likely to be a member of the AARP (67.2%) than were female subjects (48.1%, }C=5.41, df=1, p=.02). Analysis of Factors between Black and White Subjects. Table 13 shows significant findings between black and white subjects. A significantly larger proportion of black 54 TABLE 13 Significant Results of Chi-Square Analysis of Factors between Black and White Subjects VARIABLE % of BLACKS % of WHITES p= (N=83) (N=113) HOUSE 80.2 64.3 .03 SENIOR APTS. 19.3 35.7 .03 HMO MEMBER 4.8 17.7 .01 INCOME(N=162) LOW 53.0 38.5 05 MEDIUM 40.9 43.8 MED-HIGH 6.1 17.7 INVESTMENTS 39.8 56.6 .03 55 subjects (80.2%) lived in single family houses than did white subjects (64.3%,X?=9.27, df=3, p=.03). Black subjects also had a significantly lower rate of residency in a SSI apartment complex (28.6%) than did white subjects (71.4%). Black subjects were members of an HMO at a much lower rate (4.8%) than were white subjects (17.7%, X?=6.24, df=1, p=.01). Black subjects were more likely to be in the low income group (53.0%) than were white subjects (38.5%) (n=162, )F=6.01, df=2, p=.05). White subjects were in the medium- high income group at a higher rate (17.7%) than were black subjects (6.1%). All the subjects (n=3) in the high income group were black. A larger proportion of white subjects had savings and investments as a source of income (56.5%) than did black subjects (39.8%, X?=4.80, df=1, p=.03). Black subjects attend church at significantly higher rates (85.5%) than do white subjects (55.8%, X?=18.28, df=1, p<.000). Analysis of Census Tracts. Data were analyzed among census tracts. X? analysis requires at least 5 expected observations per cell (Clark and Hoskings, 1986) in contingency tables in order for the results to be considered valid. For 2X2 contingency tables of factors within census tracts, Fisher's Exact test results were used to determine statistical significance in the case of <5 expected observations. In the case of 2XN contingency tables (N > 3) this 56 statistic is sensitive to very small expected values and experiences artificial increases if expected values fall below 5 in very many classes (Clark and Hoskings, 1986). Compliance rates of census tracts and of the race and gender groups within census tracts are shown (Table 14). Compliance levels did-not differ significantly among census tracts included in the study area. Table 15 shows significant results of this analysis. Significant differences were found between census tracts in the following variables: race (X?=48.42, df=5, p<.000), presence or absence of medical checkups (X?=15.43, df=5, p=.009), private insurance coverage (x?=25.62, df=5, p=.0001), marital status (X?=28.49, df=10, p=.002), whether a subject lived alone and the number of family members 06:28.65, df=10, p=.002), ability to drive, ownership of an automobile, use of special transportation services (X’=41.60, df=5, p<.000), pension (X’=30.08, df=5, p<.000), and savings or investments (X?=16.42, df=5, p=.005). The results in age group and income level between tracts is unreliable due expected frequencies less than 5. These same data were also analyzed separately by gender and race groups. Because the data were further stratified, sample sizes became smaller, again introducing the problem of validity of the X?. Cases of unreliable significance are indicated in the following tables. Table 16 shows significant differences among white subjects between census tracts. Significant differences were 57 TABLE 14 Compliance Rates of Census Tracts Tract# Total B W F M .3; """""""" 86'; """" 66'6 """ 83:; """ SECS """ £666 13 66.7 100.0 64.3 61.5 100.0 15 50 0 50.0 50 0 44 4 66.7 16 64.3 60.9 100.0 58.8 80.0 36 01 53 5 28 6 65 5 55 6 50.0 Significant Results of Chi-Square Analysis of Variables Among Census Tracts 58 TABLE 15 36.01 16 N=27 N=43 % BLACK CHECKUPS PRINSURANCE %MARRIED %SINGLE/DIV. %WIDOWED LIVES ALONE %FAMILY = 2 %FAMILY = 3+ CAN DRIVE OWNS AUTO SPECTRANSP PENSION INVESTMENTS AGE GROUP 65-69 yrs 70-74 yrs 75-79 yrs 80-84 yrs 85+ INCOME REFUSED LOW MEDIUM MED-HIGH 80. 52. 20. 32. 20. 20. CO GOO GOO COO 0090 00000 13 15 N=15 N=48 6.7 58. 86.7 70. 66.7 41 13.3 20 33.3 33 53.3 45 86.7 62. 13.3 29. 0.0 6. 40.0 41. 33.3 37. 60.0 43. 26.7 41. 46.7 25. 40.0 22. 13.3 29. 20.0 22. 13.0 10. 8.0 28. 0.0 16 53.3 68. 46.7 14. 0.0 0 04 mUIQ OIAWNW 77. 55. 18. 18. 40. 20. \ll-‘tfi mow OQQU'IU'I moo 37. 55. 11. 79. 88. 81. 62. mnb <3an moon \OUUI OWWWN 47. 39. 13. 63. 68. 21. 65. 21. 23. 10. 31. 20. U10) ohm NU'IIh (“WI-4 QNN 001de .002 .001 .003 .000 .000 .000 .005 .000 .000 59 TABLE 16 Significant Results of Chi-Square Analysis of Variables by Census Tracts among White Subjects 36.01 16 N=4 N=29 37 N=33 VARIABLE (t) N=13 PRIVATE INSURANCE 92.3 RETIRED 69.2 HOMEMAKERS 30.8 MARRIED 38.5 SINGLE\DIV 15.4 WIDOWED 46.2 LIVE ALONE 46.2 FAM. MEMBER=2 38.5 FAM. MEMB=3+ 15.4 CAN DRIVE 53.8 AUTO OWNERS 69.2 HAS DRIVER 69.2 SPECIAL TRANSPORTATION 7.7 PENSION 76.9 INVESTMENTS 61.5 AGE GROUP 65-69 yrs 7.7 70-74 yrs 30.8 75-79 yrs 23.1 80-84 yrs 15.4 85+ 23.1 EDUCATION LEVEL 0-6 yrs 15.4 7—11 yrs 15.4 12 yrs 30.8 13-15 yrs 38.5 16 yrs 0.0 16+ yrs 0.0 13 15 N=14 N=20 64.3 35. 92.9 70. 0.0 15. 14.3 10. 35.7 50. 50.0 40. 85.7 85. 14.3 15. 0.0 42.9 40. 35.7 30. 64.3 65. 57.1 55. 21.4 25. 50.0 20. 42.9 35. 14.3 20. 14.3 25. 14.3 14.3 15. 0.0 26.6 50. 42.9 15. 28.6 15. 0.0 15. 0.0 0 000000 0 00 0 000 000 000 00 00000 25. 25. 50. 25. 25. 50. 50. 50. 50. 100. 75. 00 0 000 000 000 00 0 00000 000000 41. 55. 82. 86. 31. 10. 82. 69. 51. 17. 27. 34. 00 W 0N0 \ONIh- U"IIhiI--| mm \) 000N\I 45. 48. 42. 45. 12. 72. 78. 36. 12. 69. 66. Hmp mHm oo \1 H #04 \lxl NUI—‘NN .002 .001 .02 .0001 .05 .000 .000 .01 60 found among white subjects between census tracts in the following variables: private insurance (X?=21.01, df=5, p<.000), retired (X?=13.21, df=5, p=.05), homemaker (X2=12.60, df=5, p=.03), marital status (X2=31.60, df=10, =.0005), whether a subject lived alone and number of family members (X?=28.88, df=10, p=.001), ability to drive DF=13.60, df=5, p=.02), ownership of an automobile (X2=24.90, df=5, p=.0001), having a driver (X2=11.29, df=5, p=.05), use of special transportation services (X?=27.73, df=5, p<.000), pension (X?=30.89, df=5, p<.000), and savings and investments as a source of income (X?=15.00, df=5, p=.01). Although significance could not be determined beyond a reasonable doubt due expected cell frequencies <5 in the age group and level of education variables, the proportions of subjects are presented for visual interpretation. Table 17 shows the associations revealed between black subjects among census tracts. The significance of these findings is unreliable due to low expected cell frequencies in many of the categories. The obvious visual differences in rates between tracts could not be verified by a valid X2 statistic due to small sample sizes and expected cell frequencies. One black subject was interviewed in Tract 13 and the data gathered in this interview cannot be meaningfully compared to that of the other tracts. Immunization compliance was comparatively low in Tract 36.01 among black subjects, as was history of immunization and intent to have one in the future. 61 TABLE 17 Significant Results of Chi-Square Analysis of Variables by Census Tracts among Black Subjects VARIABLE 5 13 15 16 36.01 37 p= (%) N=12 N=1 N=28 N=23 N=14 =5 FEMALE 66 7 100.0 82 1 60 9 64 3 80 0 COMPLIANCE 50.0 100.0 50.0 60.9 28.6 40.0 DRRECOMMEND. 50.0 100.0 64.3 56.5 57.1 40.0 PASTSHOT(N=82)58.3 100.0 64.3 81.8 35.7 80.0 FUTURESHOT(N=65)63.6 0.0 63.0 65.0 27.3 60.0 CHECKUPS 91.7 100.0 64.3 87.0 100.0 40.0 PRINSURANCE 75.0 100.0 46.4 82.6 71.4 20.0 EMPLOYED 16.7 0.0 14.3 4.3 0.0 0.0 RETIRED 83.3 100.0 78.6 87.0 85.7 60.0 MARRIED 33.3 0.0 28.6 47.8 35.7 20.0 SINGLE\DIVORCE 8.3 0.0 21.4 13.0 7.1 0.0 WIDOWED 58.3 100.0 50.0 39.1 57.1 80.0 LIVE ALONE 25.0 100.0 46.4 34.8 28.6 80.0 FAM. MEMBER=2 58.3 0.0 39.3 39.1 57.1 0.0 FAM. MEMB=3+ 16.7 0.0 10.7 26.1 21.4 20.0 CAN DRIVE 50.0 0.0 42.9 73.9 71.4 0.0 OWNS AUTO 50.0 0.0 42.9 78.3 92.9 0.0 DRIVER 58.3 100.0 60.7 43.5 35.7 100.0 SPECTRANSP 0.0 100.0 35.7 4.3 0.0 80.0 SENIOR PROG 8.3 100.0 14.3 21.7 21.4 60.0 PENSION 83.3 100.0 53.6 73.6 78.6 40.0 INVESTMENTS 41.7 0.0 28.6 52.2 50.0 20.0 AGE GROUP 65-69 yrs 33.3 0.0 14.3 21.7 50.0 0.0 70-74 yrs 33.3 0 0 35 7 21.7 28 6 20 0 62 Table 17 (cont.) 75-79 yrs 16.7 100.0 21.4 39.1 14.3 20.0 80-84 yrs 0.0 0.0 14.3 4.3 7.1 60.0 85+ 16.7 0.0 14.3 13.0 0.0 0.0 EDUCATION LEVEL 0-6 yrs 16.7 0.0 10.7 17.4 0.0 - 20.0 7-11 yrs 41.7 100.0 46.4 34.8 28.6 . 60.0 12 yrs 16.7 0.0 14.3 17.4 42.9 20.0 13-15 yrs 8.3 0.0 25.0 17.4 7.1 0.0 16 yrs 0.0 0.0 3.6 4.3 7.1 0.0 16+ yrs 16.7 0.0 0.0 8.7 14.3 0.0 INCOME(N=68) LOW 44.4 0.0 76.2 50.0 8.3 80.0 MEDIUM 55.6 100.0 23.8 35.0 66.7 20.0 MED—HIGH 0.0 0.0 0.0 10.0 16.7 0.0 HIGH 0.0 0.0 0.0 0.0 5.0 0.0 63 Table 18 shows significant differences between female subjects among census tracts in the following variables: race (X?:31.45, df=5, p<.000), presence or absence of medical checkups (X?=12.36, df=5, p=.03), private insurance 00:18.04, df=5, p=.003), ownership of an automobile 00:21.50, df=5, p=.0007), use of special transportation 06:25.18, df=5, p=.0001), and a pension as a source of income (X?=14.66, df=5, p=.01). The significance of age group, education levels and income levels is unreliable due to low expected cell frequencies in many cases. Proportions of factors present in males were analyzed among census tracts. Seven male subjects were interviewed in Tract 5 and two male subjects were interviewed in Tract 13. The number of low expected cell frequencies casts suspicion on the validity of any significant findings. Table 19 shows significant differences between male subjects among census tracts based on race (X?:19.53, df=5, p=.002) and use of special transportation services (X?:18.53, df=5, p=.003). Data were analyzed within census tracts. Except in the case where the X? statistic was deemed valid beyond a reasonable doubt (pg .05 and a relatively narrow 95% CI), no determination could be made about the validity of some of these results when the data were stratified by census tract. Significant results that exhibit extremely wide confidence intervals should be interpreted with caution. In some cases, significance was indicated by a p-value at or below .05, however the CI was very wide. The width of the CI indicates 64 TABLE 18 Significant Results of Chi-Square Analysis of Variables by Census Tracts for Female Subjects 36.01 N=27 16 N=17 VARIABLE (%) N BLACK 44. CHECKUPS 83. PRINSURANCE 77. LIVE ALONE 38. FAM. MEMBER=2 44. FAM. MEMB=3+ 16. OWNS AUTO 50. SPECTRANSP PENSION 72. AGE GROUP 65-69 yrs 11. 70-74 yrs 38. 75—79 yrs 16. 80-84 yrs 11. 85+ 22. EDUCATION LEVEL 0-6 yrs 22. 7-11 yrs 33. 12 yrs 22. 13-15 yrs 16. 16 yrs 16+ yrs INCOME(N=114) LOW 41. MEDIUM 58 MED-HIGH HIGH 0 5. 0. 5. NHQKDH ONOQNUJN 23. 61. 30. 30. 15. 23. 15. 15. fisht4¢>m 00HU‘IU10 66. 27. 30. 44. 38. 13. 30. 22. 13. 19. omnbtooow pmwmm 35. 35. 29. 64. 70. 17. 17. 52. 11. 11. 29. 23. 11. ll. 50. mommm ooommiboo 44. 48. 14. 81. 11. 74. ##0#me \IONWQ 66. 20. 12. 54. 29. 54. ONNMUO \IQONU‘I .005 .006 .0006 .0001 .01 65 TABLE 19 Significant Results of Chi-Square Analysis of Variables by Census Tracts among Male Subjects 16 N=10 36.01 N=16 VARIABLE 13 (%) N 7 N: BLACK 57.1 0.0 MARITAL STATUS MARRIED 71.4 50.0 SINGLE\DIV 0.0 50.0 WIDOWED 28.6 0.0 CAN DRIVE 100.0 100.0 OWNS AUTO 85.7 100.0 SPECTRANSP 0.0 50.0 PENSION 100.0 0.0 SAVINGS 85.7 100.0 # OF CHECKUPS NONE 0.0 50.0 g 1 PER YEAR 42.9 0.0 2-6 PER YEAR 42.9 0.0 1-2 PER MONTH 14.3 50.0 50. 58. 41. 50. 25. 93. 100. 93. .002 .04 .003* .002 .02 .05 * significant at 66 t:he effect of the sample size on the stability of the (estimate. A wide CI (with a small sample size) implies an increased variability in the estimate of the effect, (Hennekens, 1987). Tables 20-25 show proportions of risk factors in the various groups of subjects and their corresponding compliance rates for each census tract in the study. In Table 20, results of X? analysis of proportions of factors in subjects and compliance rates are shown for Tract 5 (n=25) for all subjects interviewed in Tract 5 and for a breakdown by race and gender. Total vaccination compliance in Tract 5 was 52%. Compliance rates differed significantly between females and males (RR=.33, 95% CI:.17, .64, p=.01). Female subjects in Tract 5 were less likely to have received an influenza immunization in 1991. Those subjects who intended to receive an influenza immunization in the future had significantly higher rates than those who did not (X?=18.63, df=2, p<.000), however small expected cell frequencies in the classes of subjects who did not intend to or did not know if they would receive one render this result unreliable. Four percent of subjects reported being members of an HMO (compliance : 100%). Immunization compliance among black subjects (n=12) in Tract 5 was 50%. Black females in this group had a significantly lower compliance rate than did black males (25% and 100%, respectively; RR=.25, 95% CI:.08, .83, p=.03). 67 TABLE 20 Proportion of Factor in Subjects/% Immunization Compliance in Tract 5 for total, race and gender groups.** -----------—--------—-------—-—--—------------—----_--——---- ----—-—---------—---_--—-—--_-_-—‘--—--------——---——-_-----— Total n=25 Compliance(%) Female Male Black White House SSI Apt. Duplex Dr Recommend. Past Shot Future Shot Checkups Priv Insurance HMO Employed Retired Homemaker Married Single Widowed, Age Group 65—69 yrs 70-74 yrs 75-79 yrs 80-84 yrs 85+ yrs Education Leve 0-6 yrs 7-11 yrs 12 yrs 13-15 yrs 16 yrs 17+ yrs Live Alone Live w/l Live w/2+ 100/52 48/58 64/81 52/92 88/55 84/57 4/100 12/67 76/53 20/60 36/67 12/0 52/54 20/60 32/38 20/40 8/100 20/60 1 16/50 28/29 24/50 24/67 8/100 36/33 48/67 16/50 67/25* 33/100 100/50 100/50 50/67 58/86 58/88 92/55 75/67 17/100 83/50 8/100 33/75 8/0 58/43 33/75 33/50 17/0 0/ - 17/50 17/ - 42/40 17/50 8/100 17/100 25/33 58/57 17/50 100/54 100/54 46/50 69/78* 46/100 85/55 92/50 92/50 8/0 56/69 31/50 39/60 15/0 46/67 8/0 31/25 23/67 15/100 67/23 15/100 15/0 31/50 39/50 39/60 46/33 39/80 15/50 44/38 50/67* 33/83* 83/33 78/36 6/100 6/0 67/25 28/60 22/25 17/0 62/46 11/0 39/29 17/0 11/100 22/50 22/50 33/17 22/25 17/33 6/100 39/14 44/50 17/33 100/100 57/100 43/100 100/100 57/100 100/100 100/100 100/100 100/100 29/100 100/100 71/100 29/100 43/100 14/100 29/100 14/100 14/100 29/100 43/100 14/100 29/100 57/100 14/100 Refused Low Income Medium Med High High Income Volunteer Can Drive Owns Auto Has Driver Rides Bus Uses Spectran Attend Church Senior Center Social Grp Member AARP Pension Savings Earnings Other No Church Baptist Methodist Catholic Lutheran AME Jehovah’s Wit COG COGIC COC Presbyterian Episcopal Pentecostal Non-Denomin. Other 32/75 24/17 36/44 8/100 48/67 52/69 60/67 64/44 20/60 4/0 72/61 16/50 0/80 56/57 80/55 52/61 16/100 28/25 8/100 20/60 12/100 12/0 8/50 12/100 68 25/67 47/20 33/75 42/80 50/83 50/84 58/29 25/33 83/61 8/0 17/100 83/60 42/60 25/100 25/0 25/0 17/100 25/0 8/100 8/100 Table 20 (cont.) 39/80 15/0 39/40 8/100 54/57 54/57 56/50 69/56 15/100 8/0 62/62 23/67 23/67 54/57 77/50 62/62 8/100 39/40 39/60 33/67 28/0 39/29 39/43 33/33 50/44 78/36 17/33 6/0 67/42 17/33 17/67 50/33 72/31 39/29 6/100 39/14 6/100 22/50 11/100 17/0 570 29/100 14/100 29/100 29/100 72/100 100/100 86/100 29/100 29/100 86/100 14/100 29/100 71/100 100/100 86/100 43/100 14/100 14/100 14/100 14/100 14/100 29/100 ** % of subjects with the factor in regular type, % compliance in bold type *significant @ pg .05 69 White subjects (n=13) in Tract 5 had a compliance rate of 53.8%. Those subjects who had an influenza immunization in the past had a significantly higher level of compliance (78%, X2=3.97, df=1, p=.02). Female subjects (n=18) in Tract 5 had a compliance rate of 33.3%. Those who had a past immunization (67%, X?=6.25, df=2, p=.005) and intended to receive one in the future (83%, X?=3.97, df=1, p=.02) had significantly higher levels of compliance. Male subjects (n=7) in Tract 5 had a 100% compliance rate with influenza immunization. Table 21 shows results of X? for respondents residing in Tract 13 (n=15) and those for a breakdown by race and gender. Total vaccination compliance was 66.7%. All respondents resided in Riverfront Apartments. Single dwelling houses and apartments are located within Tract 13 in an area of approximately four blocks. This area is notorious for drug houses and prostitution and after inquiries to apartment managers and several friendly residents about numbers and location of elderly residents, that portion of Tract 13 was passed over for interviews. No reliable significant associations were found among subjects interviewed in Tract 13. One black subject was interviewed in Tract 13 and reported being vaccinated. Compliance among white subjects (n=14) in Tract 13 was 64.3%. Two male subjects were interviewed and neither reported being vaccinated. Female 70 TABLE 21 Proportion of Factors in Subjects/% Immunization Compliance in Tract 13 for Total, Race and Gender groups** (%) Compliance Female Male Black White House SSI Apt. Duplex Dr Recommend. Past Shot Future Shot Checkups Priv Insurance HMO Employed Retired Homemaker Married Single Widowed Age Group 65-69 yrs 70-74 yrs 75-79 yrs 80-84 yrs 85+ yrs Education Leve 0-6 yrs 7-11 yrs 12 yrs 13-15 yrs 16 yrs 17+ yrs Live Alone Live w/1 Live w/2+ 60/89 67/80 60/100 87/70 67/70 20/33 0/- 93/64 13/100 33/80 53/50 40/83 13/50 20/67 13/50 13/50 1 0/- 33/60 40/40 27/27 0/- 87/62 13/100 0/- 100/100 100/100 100/100 100/100 100/100 100/100 100/100 100/100 100/100 100/100 100/100 86/58 14/100 100/64 100/64 57/88 64/78 64/78 86/67 64/67 21/33 93/62 14/100 36/80 50/43 43/83 14/50 14/50 14/50 14/50 29/50 43/67 29/75 86/58 14/100 8/100 92/58 100/62 54/86 62/75 54/100 92/67 62/63 23/33 92/58 8/100 31/75 62/50 31/75 15/50 23/67 15/50 15/50 39/60 39/60 23/67 92/58 8/100 100/100 100/100 100/100 100/100 100/100 50/50 100/100 100/100 50/100 50/100 100/100 50/100 50/100 50/100 50/100 Refused Low Income Medium Med High High Income Volunteer Can Drive Owns Auto Has Driver Rides Bus Uses Spectran Attend Church Senior Center Social Grp Member AARP Pension Savings Earnings Other No Church Baptist Methodist Catholic Lutheran AME Jehovah’s Wit COG COGIC COC Presbyterian Episcopal Pentecostal Non-Denomin. Other 0/- 53/53 47/71 53/75 40/67 33/80 67/60 33/80 60/67 80/75 27/75 40/50 33/60 27/50 47/71 13/100 7/0 20/20 13/100 13/13 33/60 7/100 7/100 71 100/100 100/100 100/100 100/100 Table 21 (cont.) 57/63 43/67 57/75 43/67 36/80 64/56 36/80 57/63 79/73 21/67 43/50 36/61 21/33 50/71 14/100 7/0 21/33 14/100 14/100 36/60 7/100 54/57 46/67 54/71 31/50 23/67 69/56 31/75 62/63 85/73 31/75 31/40 31/50 31/50 39/60 8/100 8/0 15/0 15/100 15/50 31/50 8/100 8/100 50/100 50/100 50/100 100/100 100/100 50/100 50/100 50/100 50/100 50/100 50/100 100/100 50/100 50/100 50/100 ** % of subjects with the factor in regular compliance in bold type *significant @ pg .05 72 subjects (n=13) in Tract 13 had a 61.5% compliance rate. Table 22 shows results for Tract 15 (n=48). Total vaccination compliance in the tract was 50%. Over 65.0% of respondents lived in Capital Commons Apartments. Thirty-one percent of subjects lived in single dwelling houses. Subjects who had a recommendation from their physician to receive a vaccination had significantly higher compliance (74%, RR=3.89, 95% CI=1.57, 9.66, p=.0004) as did those with a past immunization (67%, RR=5.00, 95% CI=1.35, 18.58, p=.002) and those who had checkups with their physician (62%, RR=2.88, 95% CI=1.02, 8.13, =.03). Subjects who reported intent to receive an immunization in the future (85%) had a significantly higher rate of compliance, however, low expected cell frequencies indicate that this result is unreliable. Black subjects (n=28) in Tract 15 had a compliance rate of 50.0%. Subjects who had received a recommendation from their physician of vaccination had significantly higher compliance (78%, X?=12.60, df=1, p=.0004) as did those who reported receiving an immunization in the past (72%, RR=7.22, 95% CI=1.10, 47.40, p=.006). Subjects reporting intent to receive an immunization in the future had significantly higher compliance (82%, X?=28.80, df=2, p<.000). These results are unreliable due to low expected cell frequencies and wide confidence intervals. Female subjects (n=36) in Tract 15 had a compliance rate of 44.4%. These subjects had significantly higher 73 TABLE 22 Proportion of Factor in Subjects/% Immunization Compliance in Tract 15 for total, race and gender groups** Factor Total B W F M (%) n=48 n=28 n=20 n=36 n=12 Compliance(%) 50.0 50.0 50.0 44.4 66.7 Female 75/44 82/51 39/65 100/44 - Male 25/67 19/60 35/71 - 100/67 Black 58/50 100/50 - 64/48 42/60 White 42/50 - 100/50 36/39 58/71 House 31/53 58/50 5/100 33/50 25/67 SSI 65/52 41/64 65/47 64/44 67/75 Apt. — 8/0 - - - Duplex 4/0 — - 3/44 8/0 Dr Recommend. 56/74* 67/78* 45/67 58/67* 50/100 Past Shot 69/67* 67/72* 75/60 64/65* 83/70 Future Shot 56/85 50/90 50/89* 75/78 Checkups 70/62* 63/65 80/63 75/52 58/100* Priv Insurance 42/50 48/46 35/57 39/43 50/67 HMO 10/80 4/100 20/75 8/67 17/100 Employed 13/50 11/67 10/50 14/60 8/0 Retired 75/50 78/48 70/57 75/44 75/67 Homemaker 13/33 11/67 16/0 17/33 - Married 21/70 30/63 10/100 14/60 42/80 Single 33/31 22/17 50/40 25/11 58/57 Widowed 46/55 48/62 40/50 61/55 - Age Group 65-69 yrs 23/64 24/60 35/43 14/60 5/67 70-74 yrs 29/57 28/52 20/75 31/55 25/67 75-79 yrs 23/36 26/38 25/40 22/25 25/67 80-84 yrs 10/60 11/44 5/100 14/60 - 85+ yrs 15/29 17/50 15/33 19/29 - Education Level 0-6 yrs 8/100 13/55 5/100 8/100 8/100 7-11 yrs 48/49 40/42 50/50 53/42 33/75 12 yrs 15/14 20/44 15/0 14/0 17/50 13-15 yrs 21/60 16/62 15/67 19/57 25/67 16 yrs 8/50 7/67 15/67 6/50 17/50 17+ yrs - 7/10 - — - Live Alone 63/47 44/50 85/47 67/42 50/67 Live w/l 31/57 41/55 15/67 28/60 33/50 Live w/2+ 6/67 11/67 - 3/0 17/100 Refused Low Income Medium Med High High Income Volunteer Can Drive Owns Auto Has Driver Rides Bus Uses Spectran Attend Church Senior Center Social Grp Member AARP Pension Savings Earnings Other No Church Baptist Methodist Catholic Lutheran AME Jehovah's Wit COG COGIC COC Presbyterian Episcopal Pentecostal Non-Denomin. Other 17/38 69/49 15/71 31/33 42/40 38/44 63/53 23/46 77/48 63/53 10/60 20/30 44/48 42/40 25/33 13/68 13/83 38/46 25/67 2/100 2/100 8/0 2/0 4/100 2/0 2/0 2/100 4/50 74 22/50 59/44 19/80 33/22* 44/42 44/50 59/50 22/50 33/44 78/52 15/50 30/25 41/46 52/43 30/38 8/100 17/43 42/60 170 15/33 3/50 6/60 5/25 1/0 2/100 4/0 2/100 2/100 Table 22 (cont.) 5/0 80/50 15/67 5/100 5/100 19/43 72/42 8/67 39/39 39/36 31/37 67/46 25/44 44/44 72/50 14/60 25/33 42/47 39/36 25/33 11/75 11/75 36/31 33/67 3/100 3/100 8/0 3/0 3/100 3/0 3/0 3/100 ** % of subjects with factor in regular type, % compliance in bold type * significant @ pg . 8/0 58/71 33/75 17/0 50/50 58/57 50/83 17/50 42/60 33/75 8/0 50/50 50/50 25/33 17/50 17/100 75/67 8/0 75 compliance based on a physician recommendation of the vaccine (67%, RR=5.00, 95% CI=1.33, 18.81, p=.005), a past immunization (65%, RR=8.48, 95% CI=12.6, 57.06, p=.003), intent to receive a future immunization (89%, X?=28.80, df=2, p<.000). These results must be interpreted carefully due to wide CIs and low expected frequencies. Male subjects (n=12) in Tract 15 had a compliance rate of 66.7%. No significant associations were found among these subjects. Results are shown for Tract 16 (n=27) in Table 23. Total vaccination compliance in Tract 16 was 64.3%. Those subjects who intend to receive an immunization in the future had a significantly higher rate of compliance (88%, 28:12.28, df=2,, p=.002), however low expected cell Black subjects (n=23) in Tract 16 had a compliance rate of 60.9%. Subjects who intended to receive an immunization in the future had significantly higher compliance (85%, 18:9.50, df=2, p=.009). Low expected cell frequencies indicate this result should be carefully interpreted. White subjects (n=4) in Tract 16 had a compliance rate of 100.0%. Male subjects (n=10) in Tract 16 had a compliance rate of 80.0%. Compliance of female subjects (n=17) in Tract 16 was 58.8%. Subjects who intended to receive a future immunization had a significantly higher level of compliance (89%, X?=7.83, df=2, p=.02). The significance of this result is unreliable due to low expected cell frequencies. 76 TABLE 23 Proportion of Factor in Subjects/% Immunization Compliance in Tract 16 for total, race and gender groups** Factor Total B W F M (%) n=28 n=24 n=4 n=17 n=10 Compliance(%) 64.3 60.9 100.0 58.8 80.0 Female 61/59 61/50 75/100 100/59 — Male 39/80 32/78 25/100 - 100/80 Black 86/61 100/61 - 82/50 90/78 White 14/100 - 100/100 18/100 10/100 House 100/64 - 100/100 100/59 100/80 SSI - 41/64 - - - Apt. - - — — - Duplex - - - - - Dr Recommend. 57/75 57/69 75/100 59/70 60/83 Past Shot 82/73 82/67 100/100 77/69 90/79 Future Shot 61/88 55/85 100/100 53/89 80/88 Checkups 86/65 87/60 75/100 88/53 80/88 Priv Insurance 79/64 83/58 75/100 77/54 90/78 HMO 11/100 9/100 25/100 12/100 10/100 Employed 4/100 4/100 - - 10/100 Retired 86/65 88/60 75/100 82/57 90/78 Homemaker 11/67 9/50 25/100 18/67 - Married 46/75 48/73 25/100 35/67 60/83 Single 14/100 13/100 25/100 18/100 10/100 Widowed 39/46 39/33 50/100 47/ 30/67 Age Group 65-69 yrs 21/80 22/80 - 18/67 20/100 70-74 yrs 18/60 22/60 - 18/33 20/100 75-79 yrs 39/64 39/56 50/100 53/67 20/50 80-84 yrs 4/0 4/0 - - 10/0 85+ yrs 18/80 13/67 50/100 12/50 30/100 Education Level 0-6 yrs 14/75 17/75 — 12/50 20/100 7-11 yrs 29/38 35/38 — 29/40 30/33 12 yrs 18/80 17/75 25/100 24/75 10/100 13-15 yrs 14/50 17/50 - 12/0 20/100 16 yrs 7/100 4/100 25/100 12/100 - 17+ yrs 18/100 9/100 50/100 12/100 30/100 Live Alone 32/67 35/63 25/100 35/67 30/67 Live w/l 36/60 39/56 25/100 35/33 40/100 Live w/2+ 32/75 26/67 50/100 29/80 30/67 Refused Low Income Medium Med High High Income Volunteer Can Drive Owns Auto Has Driver Rides Bus Uses Spectran Attend Church Senior Center Social Grp Member AARP Pension Savings Earnings Other No Church Baptist Methodist Catholic Lutheran AME Jehovah’s Wit COG COGIC COC Presbyterian Episcopal Pentecostal Non-Denomin. Other 10/100 36/50 32/56 14/100 7/100 50/69 71/68 75/65 43/58 18/60 4/100 82/64 21/67 36/78 54/71 79/67 57/73 11/100 4/100 18/80 46/62 11/100 10/67 4/100 7/50 4/0 Table 23 77 17/100 52/50 17/25 9/100 4/100 52/58 74/65 78/61 44/50 17/50 4/100 87/60 22/60 30/71 44/60 74/59 52/67 13/100 4/100 13/67 57/62 13/67 4/100 9/50 4/0 (cont.) 25/100 25/100 50/100 50/100 50/100 50/100 50/100 25/100 50/100 25/100 50/100 100/100 100/100 75/100 6/100 53/44 24/50 18/100 59/60 53/56 65/55 53/56 24/75 6/100 88/53 35/67 29/80 53/67 71/58 53/67 6/100 6/100 50/100 12/100 50/100 41/43 12/100 12/50 6/100 12/50 20/100 10/100 50/60 10/100 10/100 30/100 100/80 90/78 30/67 10/0 70/86 40/75 50/80 90/78 60/83 20/100 30/67 60/83 50/100 10/100 ** % of subjects with factor in regular in bold * significant @ pg . type, % compliance 78 In Tract 36.01 (n=43), vaccination compliance was 53.5% (Table 24). Subjects had significantly higher compliance based on doctor recommendation of vaccine (73%, RR=2.18, 95% CI=1.13, 4.21,p=.02), and past immunization (73%, RR=5.07, 95% CI=1.38, 18.65, p=.001). The relatively wide CI indicates unreliable significance. Intent to receive a future immunization differed among subjects, however the significance of the statistic is unreliable due to low expected cell frequencies (82%, X?=17.12, df=2,p:.0002). Compliance rates between black (28.6%) and white subjects (65.5%) approached significance (p=.05119). Black subjects (n=14) in Tract 36.01 had a compliance rate of 28.6% White subjects (n=29) in Tract 36.01 had a compliance rate of 65.5%. These subjects had significantly higher compliance rates based on doctor recommendation of the vaccine (88%, RR=1.84, 95% CI=1.03, 3.29, p=.03) and intent to receive a future immunization (84%, RR=2.81, 95% CI=1.07, 7.38, p=.006). Female subjects (n=27) in Tract 36.01 had a compliance level of 55.6%. These subjects also had significantly higher compliance based on past immunization (78%, RR=7.00, 95% CI=1.08, 45.16, p=.002) and intent to receive one in the future (87%, RR=5.60, 95% CI=1.44, 18.71, p=.001). These results are unreliable due to low expected cell frequencies and wide CIs. Male subjects (n=16) in Tract 36.01 had a compliance 79 TABLE 24 Proportion of Factor in Subjects/% Immunization Compliance in Tract 36.01 for total, Total n=43 race and gender groups** Compliance(%) Female Male Black White House SSI Apt. Duplex Dr Recommend. Past Shot Future Shot Checkups Priv Insurance HMO Employed Retired Homemaker Married Single Widowed Age Group 65-69 yrs 70-74 yrs 75-79 yrs 80-84 yrs 85+ yrs Education Level 0-6 yrs 7-11 yrs 12 yrs 13-15 yrs 16 yrs 17+ yrs Live Alone Live w/1 Live w/2+ 100/29 57/50 36/60 21/67 100/29 71/30 86/25 7/0 36/40 7/0 57/25 50/14 29/50 14/0 7/100 7/0 21/0 43/33 7/100 7/100 14/0 21/33 57/38 21/0 62/72 38/54 100/66 100/66 48/88* 83/75 66/84* 90/67 62/90 17/60 7/50 76/68 28/63 62/72 3/100 35/50 52/60 17/60 28/75 3/100 3/100 24/86 41/67 17/40 7/50 7/50 39/55 55/69 7/100 100/56 33/22 67/72 100/56 44/75 67/78* 56/87* 93/56 85/57 11/100 4/100 67/56 33/56 44/75 4/100 52/36 41/55 33/55 22/50 4/100 4/100 26/43 52/57 7/0 7/100 4/100 37/50 48/62 15/50 100/50 31/40 69/55 100/50 63/70 69/64 44/71 94/48 81/46 13/0 6/0 100/50 69/55 6/0 25/50 69/36 25/75 6/100 25/75 25/50 25/75 7/0 19/0 25/50 69/55 7/0 Refused Low Income Medium Med High High Income Volunteer Can Drive Owns Auto Has Driver Rides Bus Uses Spectran Attend Church Senior Center Social Grp Member AARP Pension Savings Earnings Other No Church Baptist Methodist Catholic Lutheran AME Jehovah’s Wit COG COGIC COC Presbyterian Episcopal Pentecostal Non-Denomin. Other 17/50 7/67 47/47 26/55 2/100 40/59 79/53 88/55 33/57 16/57 7/33 70/53 14/50 26/46 28/58 81/57 63/59 16/29 5/50 33/57 14/17 9/75 7/67 7/100 7/33 2/0 2/0 7/33 2/0 2/100 7/100 80 14/50 7/0 57/13 14/50 7/100 29/50 71/20 93/31 36/40 21/33 93/23 21/33 29/0 64/22 79/27 50/29 7/0 7/0 7/100 36/20 770 21/33 770 7/0 7/100 Table 24 (cont.) 21/50 7/100 41/75 31/56 40/64 83/67 86/69 31/67 14/75 10/33 59/77 10/67 24/71 76/72 83/71 70/70 21/33 3/100 44/54 4/0 14/75 7/100 10/100 7/50 3/100 7/100 19/40 11/67 44/50 22/67 4/100 37/70 70/58 82/59 37/50 11/33 11/33 70/53 15/50 26/57 63/65 74/60 56/67 14/50 7/50 30/63 11/0 7/50 7/100 4/100 7/50 4/0 4/0 7/50 4/0 4/100 11/100 19/67 50/50 31/40 44/43 94/47 100/50 25/75 25/75 69/55 13/50 25/25 86/50 94/53 75/50 19/0 38/50 19/33 13/100 7/0 13/100 7/0 7/0 ** % of subjects with factor in regular type, % compliance in bold type *significant @ pg . 81 level of 50.0%. No significant results were found for these subjects based on any factor. Table 25 shows results for Tract 37 (n=38). Total vaccination compliance was 60.5%. Seventy-one percent of respondents resided in single family houses, 26.3% in South Washington Park Apartments and one subject (2.6%) in Somerset Senior Citizens Community. Subjects who had a recommendation from their physician to be vaccinated against influenza had significantly higher compliance (89%, RR=2.54, 95% CI=1.37, 4.72, p=.002). Subjects who intended to receive an immunization in the future had significantly higher compliance (96%, X?=26.83, df=2, p<.000), however, because of low expected cell frequencies, this result is unreliable. Black subjects (n=5) in Tract 37 had a compliance rate of 40.0%. Low expected cell frequencies diminished the reliability of any significant results. Compliance for white subjects (n=33) in Tract 37 was 63.6%. White subjects had significantly higher compliance levels based on doctor recommendation of vaccine (82%, RR:2.66, 95% CI=1.38, 5.12), and intent to receive a future immunization (100%, X?:25.80, df=2, p<.000). Female subjects (n=24) in Tract 37 had a compliance rate of 58.3%. Females had higher compliance rates based on doctor recommendation (82%, RR:2.13, 95% CI=1.01, 4.47, p=.04) and past immunization (80%, RR:3.60, 95% CI=1.03, 12.54, p=.009). The relatively wide CI indicates unreliable 82 TABLE 25 Proportion of Factor in Subjects/% Immunization Compliance in Tract 37 for total, race and gender groups** Compliance(%) Female Male Black White House SSI Apt. Duplex Dr Recommend. Past Shot Future Shot Checkups Priv Insurance HMO Employed Retired Homemaker Married Single Widowed Age Group 65-69 yrs 70-74 yrs 75-79 yrs 80-84 yrs 85+ yrs Education Level 0-6 yrs 7—11 yrs 12 yrs 13-15 yrs 16 yrs 17+ yrs Live Alone Live w/l Live w/2+ 80/50 20/0 100/40 40/0 60/67 40/50 80/50 60/67 40/50 20/0 20/0 60/33 20/0 20/0 80/50 20/0 20/0 60/67 20/0 60/67 20/0 80/50 20/0 61/60 39/69 100/64 75/68 21/43 3/100 49/94* 72/74* 61/100* 67/73 73/63 18/83 12/75 100/66 4/0 46/53 7/100 49/68 24/53 24/50 9/33 27/89 15/40 55/72 27/44 9/100 9/33 42/64 45/60 12/75 46/82* 63/80* 58/93 58/71 67/56 13/100 4/100 92/59 8/0 25/33 8/100 67/63 13/67 29/43 42/25 17/25 58/71 33/25 4/100 4/100 67/63 20/40 13/67 100/64 7/0 93/69 100/64 50/100* 86/58 57/100* 71/70 64/67 29/50 21/67 100/64 71/60 29/75 36/80 14/50 29/25 14/100 7/100 7/0 50/71 14/100 14/100 14/0 14/50 71/70 14/50 Refused l3/80 - 15/80 8/50 21/100 Low Income 31/58 80/50 24/63 46/55 7/100 Medium 44/47 20/0 61/60 38/56 57/38 Med High 11/100 - - 8/100 14/100 High Income - - - - - Volunteer 29/36 25/0 27/44 25/33 36/41 Can Drive 63/63 73/63 46/55 93/69 Owns Auto 68/65 79/65 54/62 93/69 Has Driver 45/65 100/40 36/75 63/67 14/0 Rides Bus 8/0 40/0 3/0 4/0 14/0 Uses Spectran 21/50 80/50 12/50 29/57 7/0 Attend Church 58/50 100/40 52/53 63/53 50/43 Senior Center 18/43 60/67 12/25 25/33 7/100 Social Grp 24/56 20/0 24/63 17/75 36/40 Member AARP 55/67 40/0 58/74 49/64 71/70 Pension 66/64 40/0 67/68 54/62 86/67 Savings 61/61 20/0 67/64 50/67 79/55 Earnings 13/80 - 15/80 4/100 29/75 Other 3/100 - 3/100 - 7/100 No Church 43/67 - 55/67 46/55 50/86 Baptist 11/25 60/33 3/0 13/33 7/0 Methodist 13/40 - 15/40 17/50 7/0 Catholic 11/75 - 12/75 8/100 14/50 Lutheran 11/75 - 12/75 8/100 14/50 AME 3/0 20/0 - 4/0 - Jehovah’s Wit - - - - — COG 3/100 - 3/100 - 7/100 COGIC 3/100 20/100 - 4/100 - COC - — - - - Presbyterian - - - - - Episcopal - - - - - Pentecostal - - - - - Non-Denomin. Other 83 Table 25 (cont.) ** % of subjects with factor in regular type, %compliance in bold type * significant @ pg .05 84 significance. Females who intended to receive a future immunization had significantly higher compliance (93%, )P=16.52, df=2, p=.0003). This result must be cautiously interpreted due low expected cell frequencies. Male subjects (n=14) in Tract 37 had a compliance rate of 64.3%. Compliance was significantly higher in males who had a doctor recommendation (100%, RR=3.50, 95% CI=1.08, 11.29, p=.01). The CI is rather wide and significance should be interpreted cautiously. Hypophgsis Tgsting. The results of the X? test were used to test the null hypotheses put forth in the statement of problem. Analysis of the data for the entire sample does not allow rejection of H51 08:2.58, df=1, p=.11), although the difference in compliance between black and white subjects approached significance within Tract 36.01 (p=.05119). No significant difference exists between the compliance rates between black and white subjects in the total sample or among census tracts. IHZ cannot be rejected (X?:3.21, df=5, p=.66). No significant difference exists in compliance rates between the census tracts observed in the study. In the third case, the HTB is rejected and the alternative hypothesis is accepted (X?=32.05, df=1, RR:2.18, CI=1.62, 2.95, p<.000). Compliance rates of those subjects that received a recommendation from their physician to get a vaccination against influenza differ significantly 85 from the compliance rates of those who did not. The RR indicates that the association is a strong positive one. Further, the same results were obtained in Tracts 15, 36.01, and 37. These results strengthen the case for rejection of the null hypothesis. 1%4 is rejected and the alternative is accepted (X2=4.84, df=1, RR=1.52, CI: 1.02, 22.7, p=.02). A significant difference exists in the compliance rates of those subjects who have regular check-ups and those who do not. The RR indicates that the association is a positive one. Further, the number of check-ups that a subject had per year was associated significantly with compliance (X?=7.75, df=3, p=.05). Chi-square for linear trend was calculated for this risk factor and no significant linear trend in doctor visits was found. Compliance in Tract 15 also differed significantly according to absence or presence of checkups. IRS cannot be rejected (X?=4.19, df=6, p=.65). No significant differences exist between the compliance rates of subjects of different levels of education. }%6 cannot be rejected (N=162, X?=4.59, df=2, p=.10). No significant differences exist between the compliance rates of subjects of different income levels. 1%7 can be rejected based on the finding that female subjects who are members of an HMO have a significantly higher compliance rate than females who are not. None of the other groups in the sample have significantly different compliance based on this factor, however. 86 W Subjects who reported they were not vaccinated in 1991 were asked why (Figure 10). One third felt a vaccination for influenza was unnecessary. Twenty percent reported that they had become ill after receiving a past vaccination for influenza, 9.4% stated their physician had advised them not to take the vaccine (allergies to eggs or an existing medical condition). Subjects stated they hate needles (2.3%) and 2.3% said they did not know why they did not receive a vaccination. Over 35% of the subjects had various other reasons for non-compliance. Subjects were asked if they intended to receive an influenza immunization in the future. A total of 55.8% intended to receive an influenza vaccination in the future and 6.6% of subjects did not know if they would receive one in the future. An interesting result was obtained when the reason given for not intending to be vaccinated were analyzed according to vaccination history (Figure 11). The largest proportion of those who had received a vaccination in the past do not consider receiving one in the future because they became ill after a past vaccination (59.0%). The largest proportion of subjects who had not received an influenza immunization in the past (after Other; 45.7%) felt that the vaccine was unnecessary (37%) (Figure 12). Black subjects who did not comply with influenza immunization reported that they did not because of a 87 Hates Needles Became Ill (2.4%) (20.0%) Not Necessary Doctor's Advice (9.4%) (30.6%) Don't Know (2.4%) Other (35.3%) Figure 10. Reasons for non-compliance. Fear of Becoming 111 (59.0%) Doctor's Advice (5.1%) Other (5.1%) Don't Know (2.6%) Hates Needles (2.6%) Not Necessary (25.6%) Figure 11. Reasons for non-compliance: subjects with history of vaccination. 88 Not Necessary (37.0%) Hates Needles (2.2%) Fear of Becoming 111 (4.3%) Doctor's Advice (4.3%) Other (45.7%) Figure 12. Reasons for non-compliance: subjects with no history of vaccination. Became 111 (19.0%) Hates Needles (4.8%) Doctor's Advice (9.5%) Not Necessary (23.8%) Other (42.9%) Figure 13. Reasons for non-compliance: black subjects. 89 physician’s advice (9.5%), they had become ill from a prior vaccination (19%), they hated needles (4.8%),they did not believe it was necessary (23.8%), and other miscellaneous reasons (42.9%) (Figure 13). White subjects did not comply for the following reasons: physician’s advice (9.3%), became ill after a prior shot (20.9%), felt it was not necessary (37.2%), did not know why they hadn’t received a vaccination (4.7%) and other (27.9%) (Figure 14). None of the white subjects replied that they hated needles. Female subjects who did not comply with vaccination gave the following responses as to why they did not receive an influenza vaccination: doctor’s advice (9.1%), became ill (21.2%), hates needles (3.0%), not necessary (30.3%), don’t know (1.5%) and other (34.8%) (Figure 15). Most females with no history of vaccination compliance felt that an influenza vaccination was unnecessary (36.6%) or did not comply for various other reasons (51.2%) (Figure 16). The majority of female subjects who had a history of influenza vaccination responded that they refuse to be vaccinated because they became ill from a past vaccination (58.3%) (Figure 17). Male subjects gave the following responses to why they did not receive a vaccination: doctor’s advice (10.5%), became ill (15.8%), not necessary (31.6%), don’t know (5.3%) and other (36.8%) (Figure 18). Males with no history of immunization either believed vaccination was unnecessary 90 Became III (20.9%) Not Necessary Doctor's Advice (9.3%) (37.2%) Don't Know (4.7%) Figure 14. Reasons for non-compliance: white subjects. Hates Needles (3.0%) Became 111 (21.2%) Doctor's Advice (9.1%) Not Necessary (30.3%) Don't Know (1.5%) Other (343%) Figure 15. Reasons for non-compliance: female subjects. 91 Not Necessary (36.6%) Hates Needles (2.4%) D t ' Ad' 7.37 Don't Know (2.4% 0° ors Vice( 0) Other (51.2%) Figure 16. Reasons for non-compliance: female subjects with no history of vaccination. Became 111 (58.3%) Doctor's Advice (12.5%) Other (4.2%) Not Necessary (20.8%) Hates Needles (4.2%) Figure 17. Reasons for non-compliance: female subjects with history of vaccination. 92 Became 111 (15.8%) Not Necessary Doctor's Advice (10.5%) (31.6%) Don't Know (5.3%) Other (36.8%) Figure 18. Reasons for non-compliance: male subjects. Not Necessary (60.0%) Other (40.0%) Figure 19. Reasons for non—compliance: male subjects with no history of vaccination. 93 (60.0%) or had other reasons (40.0%) (Figure 19). Males with a history of immunizaton were not vaccinated for the following reasons: doctor’s advice (14.3%), became ill (21.4%), not necessary (21.4%), don’t know (7.1%) and other (35.7%) (Figure 20). A large proportion of the reasons for non—compliance was placed in the Other category and the largest proportions of black, female and male subjects who reported that they had not complied with vaccination answered in this category. Eight subjects do not comply with influenza because they don’t believe in it. Answers included: "I just don’t get them", "I don’t like them", and "never have, never will get one". Seven subjects gave answers that indicated that getting an influenza vaccination just was not important to them. These could generally be placed in the "not necessary" category on the questionnaire. Typical responses included: "I forgot", "too lazy to ask for it", "too busy" and "never thought about it". Three subjects reported that they were ill or hospitalized during the influenza season. Two subjects stated that their phySician had run out of the vaccine. Other responses included: "Heard people got sick", "Can’t get around too well", "I have faith in God", "Let nature take its course", "My father was a chiropractor and I never was vaccinated", "It’s too experimental ". 94 Became 111 (21.4%) Not Necessary Doctor's Advice (14.3%) (21 .4%) Don't Know (7.1%) Other (35.7%) Figure 20. Reasons for non-compliance: male subjects with history of vaccination. 95 WW Subjects were asked who was their regular physician. Over six percent said they had a physician, however the name and location of their physician was not ascertained during the interviews. The compliance rate for these subjects was 58.3%. Six percent of subjects reported having no doctor. The compliance rate for this group was 16.7%. When asked why they had no physician many replied that they hadn’t been sick in years, weren’t generally sickly, and that if one went to the doctor, he or she would find something wrong. Several subjects placed their faith in God or natural remedies. Appendix F contains a list of the physicians, the reported rate of recommendation of vaccine by each physician and the compliance rates of the subjects. Several of the respondents (n=14, 7.0%) reported that their family physician was Dr. William Harrison. The compliance rate for subjects who were patients of Dr. Harrison was 42.9%, which is double the recent national trends for compliance. Dr. Harrison was the first black physician to practice in the Lansing area and retired in 1992. A majority of subjects who reported that they were patients of Dr. Harrison live in Tracts 15 and 16. All of the respondents who reported Dr. Harrison as their physician were black. A number of subjects (6.6%) have physicians who 96 practice outside Lansing. The locations include E. Lansing, MI (just east of Lansing), Holt, MI (just south of Lansing), Ann Arbor, MI (Veteran’s Administration Hospital), and Battle Creek, MI (Veteran’s Administration Hospital). A number of subjects see physicians in Grand Ledge, MI (15 miles west of Lansing) and Eaton Rapids (10 miles south of Lansing). One respondent saw her physician in Olivet, MI, which is in southern Eaton County, MI. One respondent stated that his physician was in Chicago, IL and he made regular trips there to consult with him. Medical Conditions of Subjects. Many subjects (46.4%) volunteered information regarding their medical conditions to the investigator during the interviews. The various conditions were divided into the following categories: Hypertension, Cardio/Pulmonary, Diabetes/Renal, Arthritis/Osteoporosis, Cancer, and Other (including allergies, glaucoma, various surgeries unrelated to any of the above, and nervous conditions). A final category called Several included those subjects who reported having a cardio/pulmonary or diabetes/renal malady along with one or more conditions in the other categories. Table 26 shows the proportions of subjects who reported medical conditions and their respective compliances with influenza immunization. Of those subjects who reported having only hypertension, 27.3% were vaccinated. Sixty percent of those reporting a cardio/pulmonary condition were vaccinated and 97 TABLE 26 Medical Conditions Reported by Subjects Condition N= (% w/Condition) %Compliance No Condition - 105(53.6) 59.0 Hypertension 11( 5.6) 27.3 Cardio/Pulmonary 25(12.8) 60.0 Diabetes/Renal 13( 6.6) 84.6 Arthritis/Osteo 6( 3.1) 33.3 Cancer 5( 2.6) 20.0 Other l7( 8.7) 58.8 Several 14( 7.1) 50.0 98 84.6% of those reporting diabetes or other condition were vaccinated. Twenty percent of those who were currently undergoing chemotherapy for a cancer condition were vaccinated. One such subject was not vaccinated due to their physician's advice; two subjects did not take the vaccine because they became ill from it in the past, and one subject was on chemotherapy and did not want the vaccine. Discussion. Compliance rates in this study were relatively high compared to recent national averages (MMWR, 1986; Fedson, 1990). Compliance among these subjects (56.3%) was higher than that of the elderly population of Ingham County in 1990 (51.5%) but lower than that in 1991 (68.3%) (ICHD, 1992). According to the report compiled by the ICHD (1992), 63% of medicare B vaccines were distributed by physicians and related health care providers. In this study, 80% of subjects reported medical checkups (p=.03), 53.8% reported that their physician had recommended they be vaccinated (p<.000), and 55% of those who reported being vaccinated said they received their vaccination at their physician's office. A simple visual inspection of the rates of recommendation of vaccine of each physician or clinic and the corresponding compliance rates shows a trend toward correlation (not significant) between these two factors (Appendix F). These facts underscore the importance of physicians in the high compliance rates of Ingham County and in this sample. 99 Immunization history, medical checkups and their frequency, and intent to receive a vaccination in the future were all significantly associated with compliance in the entire sample and among groups stratified by race, gender and census tract of residence. These findings are consistent with those reported in other studies (Ennis et al, 1976; Aho, 1979; MacPherson, 1980; McKinney and Barnas, 1989; Stehr-Green et al, 1990). The most striking result of this study was the significant difference in compliance between male and female subjects. This result conflicts with many of the findings of other studies. MacPherson et a1 (1980) reported that elderly males had lower compliance. Monto and Ross (1979) reported that females had higher compliance. Aho (1979) reported that males had significantly higher rates than females and that this was contrary to findings of studies of utilization of health services at that time. The Behavioral Risk Factor Survey, however, did find that males had slightly higher levels of compliance than did females (MMWR, 1989). Several factors were considered and explored in this study to determine what effect they have on the compliance levels of the gender groups. Immunization compliance among female subjects was significantly associated with a physician recommendation of vaccine, past immunization and membership in an HMO. HMOs promote preventive health care and the Bluecare Network in Lansing holds immunization clinics as a covered benefit to 100 its members. Almost 67.0% of these subjects received a recommendation of immunization from their doctor. This is a relatively high rate in comparison to that of female subjects as a whole (51.1%) and the entire sample (53.8%). The difference in compliance rates among female HMO members and non-members is directly linked to the high rate physician recommendations reported by female HMO members. An interesting discovery was that 42.0% of female subjects who are HMO members received their vaccination at their physician's office, and 42.0% received theirs at the ICHD. Three of the 5 subjects who received their immunization at ICHD live in Tracts 36.01 and 37 which are nearer to the ICHD in distance than any of the other tracts. These subjects also reported owning automobiles and said they were able to drive. Perhaps the ICHD was more convenient in terms of location for them. No significant difference was found between the compliance rates of black and white subjects in this study. This finding conflicts with those of other studies (Aho, 1979; MMWR, 1988; Sievert, 1989; MMWR, 1989; Stehr-Green, 1990). McKinney and Barnas (1989), however, reported that in their study, compliance was unrelated to age, gender or race. Black and white subjects differed significantly in their proportions of those who were HMO members. The proportion of black subjects who were HMO members was 4.8%, and that of white subjects, 17.7%. Both these proportions 101 were relatively small. Compliance rates, however, did not differ between these two groups, in the sample as a whole, or between male and female subjects according to HMO membership. Among male subjects, a doctor’s recommendation was important for compliance. Both gender groups had doctor recommendation as a common factor that was significantly associated with compliance. Intent to receive a future immunization was probably important in both groups, however, small expected cell frequencies render these findings unreliable. Among female subjects, history of compliance was a very important factor in terms of compliance and non-compliance. Indeed a large proportion (31.0%) of the reasons why females did not take an immunization in 1991 and did not intend to in the future was because they had become ill after a previous immunization. Analysis of the differences in proportion of all the other variables between male and females subjects was performed in order to determine if any significant connections or differences were pertinent to level of compliance in either group. Income and education levels did not appear to have a significant bearing on the vaccination compliance of these subjects either as a whole or broken down into to groups by gender, race and census tract of residence. Male subjects had significantly higher levels of income than female 102 subjects, however, white subjects also had higher levels of income than black subjects and no significant difference in compliance was found between the race groups. Male subjects reported a significantly higher rate of automobile ownership and ability to drive than did female subjects. Ownership of a vehicle and ability to drive it might effect a subject's ability to get to the doctor on a frequent basis. Frequency of doctor visits was significantly associated with compliance in the total sample. Frequency of doctor visits is also significantly associated with owning a car, although no significant difference between frequency of doctor visits and compliance was found between male and female subjects. Age also did not appear to effect compliance. This finding was confirmed in one study reviewed (McKinney and Barnes, 1989). A large proportion of subjects were aged 65- 69 years. The majority of male subjects were aged 65-69 years. The ages of female subjects were more evenly distributed among the age groups (Table 6). In some cases, however, younger subjects may believe an immunization is unnecessary because of their age. Male subjects predominately lived in single family houses, as did black subjects. No significant difference was found in the compliance levels of black and white subjects based on type of dwelling. It could be assumed that those living in houses might be more isolated socially or spatially from dissemination of information (even in the way 103 of casual conversation) about influenza vaccination than those who reside in senior apartment complexes and therefore would have lower compliance levels. Vaccination coverage, however, did not vary significantly according to type of dwelling in either group. Black subjects report attending church in far greater proportions (85.5%) than do whites (55.8). This difference was highly significant (p=.0000) and indicates that perhaps churches would be a good location to reach black residents by setting up influenza immunization clinics in them. Over 78.0% of subjects reported being regularly involved with church, a senior citizen center, a social club or some type of volunteer work. The immunization compliance for those that reported participation in these activities was 57.8%. Those that reported that they did not participate in any of these activities totalled 21.4%. The compliance for this group was 52.4%. Participation in one or more of these activities was not significantly associated with compliance. In this study, those who had more active social lives did not differ significantly in compliance from those that did not. Even those who volunteered at hospitals where immunizations are offered free of charge did not differ significantly in immunization compliance from those who performed other types of volunteer work. The majority of male subjects were married (62.3%). The majority of female subjects were widowed (58.5%). Monto et al (1977) found that married subjects had higher compliance 104 rates. MacPherson et al (1980) found that single individuals had lower compliance rates. Fifty—seven percent of females reported living alone, as did 29.5% of male subjects, 54.9% of white subjects, and 39.8% of black subjects. Perhaps married male subjects are more apt to avail themselves of preventive health care because of the presence of supportive relationships in their households. Although black subjects reported a high rate of being widowed (51.8%), a relatively low proportion of them lived alone. Males and females differed significantly in their rates of past immunization and intent to receive one in the future (Table 12). Females had lower rates of both factors than did males. This fact more than any other probably accounts for the lower rates of immunization among females in this study. It is interesting, however, that a doctor’s recommendation of the vaccine did not differ significantly between groups, and would suggest that perhaps there exists a "hard core" of individuals among the female subjects that simply refuse to be immunized against influenza. A large proportion of non-compliers, both male and female, felt that immunization against influenza was unnecessary. Others refused to receive a vaccination because they had a bad reaction in the past. These findings are consistent with those of Rundall and Wheeler (1979) and Aho (1979) who found perceived necessity and efficacy of the vaccine, and perceived risks of the vaccine to be positively associated with compliance. 105 Relatively few subjects used the Capital Area Transportation Authority (CATA) for transportation, the public transportation system of the Lansing area. Several subjects felt the bus was inconvenient. In most cases, subjects said that they had to walk too far to a bus stop, that the bus near their residence would take them way out of their way, or the bus schedules were inconvenient. Several subjects did not feel they could get around well enough to use the bus. Anecdotal information given by subjects regarding their medical conditions was taken by the investigator. Those who reported having cardio\pulmonary conditions reported a 60.0% rate of doctor recommendation of vaccine. Those who reported diabetes or renal conditions reported an extremely high rate of doctor recommendation (84.6%). The physicians of these subjects appear to be aware of the necessity of vaccination of these individuals. Those individuals who reported hypertension or cancer had relatively low rates of doctor recommendation, as did those who reported more than one condition (Several category). Persons with hypertension are monitored regularly by a physician, and these visits would be excellent opportunities for education of the patients about the benefits of immunization. As mentioned, this information was anecdotal, and no statistical analysis was or should have been performed, however, the information was informative and interesting. 106 Census Tracts. Tract 13 had the highest overall compliance rate (66.7%) (Table 14). Tract 16 had the highest compliance rate among black subjects (60.9%) (one black resident was interviewed in Tract 13 [100.0%] and reported being vaccinated). Tract 36.01 had the highest compliance rate among white subjects (65.5%). Tract 13 had the highest female compliance rate (61.5%). Tract 5 had the highest male compliance rate (100.0%). All of the subjects interviewed in Tract 13 lived in Riverfront Apartments. Riverfront had an overall compliance rate of 66.7%, Capital Commons, 51.6%, and South Washington Park, 50.0%. The high compliance rate of the Riverfront residents may be due to the fact that all of these respondents were recruited from the Resident’s Council meeting. Perhaps some connection exists between active participation in the council and immunization compliance. The significant difference in distribution of black subjects among census tracts can be explained by the fact that tracts were chosen specifically on the basis of their racial makeup. The difference in income level among tracts is explained by the distribution of household income among tracts (Table 2). Tracts 13 and 15 are the least affluent in terms of household income and have the highest proportion of total residents living under the poverty level. In this study they also have the largest proportions of low income subjects (p=.000). Significant differences in age are also 107 explained by the demographic make-up of the tracts. The proportions of subjects who reported regular medical checkups differed by census tract of residence. These proportions were all relatively high, however, Tract 37 had the lowest (63.2%). Tract 15 had the significantly lowest proportion of subjects who carried private medical insurance. This tract also had the highest proportion of low income subjects. Tract 36.01 had the highest proportion of privately insured subjects and also the lowest proportion of low income subjects. Tract 13 had the significantly highest proportion of subjects who lived alone. This result in not surprising since all subjects resided in Riverfront Apartments. The majority of those living in SSI apartments live alone. Tract 13 had the lowest proportion of subjects who owned cars or could drive. In spite of these significant factors, Tract 13 had the highest overall compliance rate of the tracts included in this study. The only instance where a difference in compliance between black and white subjects even approached significance was within Tract 36.01 (Table 23). Table 16 shows significant results and those suspected of being significant for black subjects among census tracts. Tract 36.01 has the lowest (questionable significance due to low expected cell frequencies) compliance of all the tracts for blacks subjects. This is almost certainly due to the fact 108 that black subjects in Tract 36.01 have extremely low rates of past immunization and intent to receive a vaccination in the future (questionable significance due to low expected cell frequencies). Rate of doctor recommendation (57.1%) is not very low relatively and 100% of subjects reported regular checkups. Tract 36.01 had the significantly highest proportion of all the tracts of subjects reporting medical checkups (93.0%). The black subjects in this tract as a group were relatively young. Indeed among all tracts (Table 4) 36.01 had the highest proportion of subjects in the 65-69 years group (51.2%, p=.000). Fifty percent of black subjects of 36.01 were in the 65-69 years age group (7.1% were 80 years and older). Of all the sample tracts, 36.01 has the smallest proportion of elderly residents (5.0%) (Table 2). Tract 36.01 has the highest proportion of residents interviewed who were 65-69 years (51.2%). Perhaps because of the relative youth of the overall proportion of elderly residents who live in the tract, influenza immunizations are not perceived as a priority or are not discussed among neighbors. Subjects may not perceive themselves as "old" and think they are not susceptible to the complications of influenza. The high compliance of white subjects of that age group within the tract (65.5%) would not, however be explained in this manner. This theory would also be contradicted by the fact that a large proportion of all of the male subjects were 65-69 years and had relatively high compliance. The majority of black subjects in this tract 109 were high school graduates (42.9%) and 14.3% had graduate education. Black subjects in Tract 36.01 were mostly medium income (66.7%) and had the highest proportion of medium-high income black subjects (16.7%) of all the tracts. Thirty percent of black subjects in Tract 36.01 stated they felt immunization was unnecessary and 40.0% gave answers that indicate they are part of a "hard core" of individuals who have their minds made up and will not take a vaccination. The fact that a large proportion of these subjects had a recommendation from their physician underscores the probability that the low compliance in this group is due to perceived lack of necessity of the vaccine by the subjects or simple refusal to accept vaccination. Tract 36.01 also had the lowest compliance rate for male subjects (50.0%, n=16). The relatively small number of males subjects must be considered when interpreting these results. The largest proportion of males (69%) in Tract 36.01 were in the 65-69 years age group and the subjects in this age group a 36% compliance rate. This rate was lowest among compliance rates for males among the all the census tracts. No significant differences in compliance were noted based on the factors known to be associated with higher compliance in this study, probably due to the number of male subjects. Proportions of subjects who had a doctor recommendation, a past immunization and reported regular medical checkups were relatively high. A relatively small proportion of subjects intended to receive an immunization 110 in the future. The reasons subjects gave for non—compliance were almost evenly split between believing it was not necessary and various other reasons ("forgot", "doctor ran out of the vaccine", "never thought about it" and "I'm against it because people got sick"). Tract 5 had the lowest compliance rate for female subjects (n=18, 33.3%) (Table 19). This tract also had the lowest compliance among black female subjects (25%). The difference in compliance between males and females was significant in this tract as was that between black male and female subjects. Male subjects (and black male subjects) had a compliance rate of 100%. Past immunization and intent to receive an immunization in the future was significantly associated with compliance for female subjects in Tract 5. Proportions of female subjects who reported a recommendation by their physician (44%), had a past immunization (50%) and intended to have one in the future (33%) were all relatively small. Eighty-three percent of female subjects reported having regular medical checkups. Low compliance in this tract can probably partially be attributed to the relatively low rate of physician recommendation. The reasons given by female subjects for not receiving an immunization reflect that most felt it was unnecessary, and many just refuse to consider having it. Tract 15 had the lowest overall compliance rates of all the tracts (Table 21). This is probably attributable to the relatively low compliance of females within the tract 111 (44.4%). Also, older age groups had low compliance rates. This was true for subjects aged 75-79 years (36%) and those aged 85 years and older (29% in the entire tract sample; those aged 75-79 years (38%) and 80-84 years (44%) among black subjects; in the aged 65-69 years (43%) and 85 years and older (33%) among the white subjects; and in the 75-79 years age group (25%) among female subjects. Compliance was also relatively low among black (48%) and white (39%) female subjects. White (47%) and female (44%) subjects who lived in Capital Commons had relatively low compliance rates. The majority of these subjects reported that they felt immunization against influenza was unnecessary. This is especially alarming since 36.4% of these individuals were 85 years and older and highly susceptible to the complications of influenza. Perhaps these individuals would benefit from health education on this subject. Lack of transportation did not seem to be a hindrance to vaccination compliance. Owning an auto and ability to drive were significantly associated with the number of medical checkups a subjects had. The fact that males had a significantly higher rate of automobile ownership and ability to drive may be indirectly linked to their significantly higher compliance. Income level was also significantly associated with auto ownership and males were significantly more affluent than females in this study. No significant difference was found between auto ownership and compliance and ability to drive and compliance, however. 112 The wide confidence intervals encountered in many of the preceding analyses of intent to receive a future immunization as a factor in immunization compliance can be considered probably significant after careful consideration of the data. Because associations appeared to be significant time and again throughout the analyses and because in each case the majority of those subjects who did not intend to have an immunization in the future, did not receive an immunization against influenza in 1991, it is probable that future intent is a significant factor in the compliance of these groups. Chance, however cannot be ruled out as a factor in this result. Chapter Five Conclusions and Recommendations The evidence of this study overwhelmingly indicates that a history of immunization, a doctor recommendation of vaccine, and intent to receive a future immunization are strongly positively associated with vaccination compliance in this sample population. Ennis et al, 1976; Monto and Ross, 1979; Aho, 1979; Macpherson et al, 1980; McKinney and Barnas, 1989; and Stehr-Green et a1, 1990 reported similar results. These factors held for the entire sample, and for numerous groups of subjects broken down by census tract of residence and race and gender groups. Medical checkups and their frequency were positively associated with compliance in the total sample, however frequency of medical checkups had no bearing on the results in the breakdowns by census tract, gender and race, perhaps due to small numbers of subjects. This study also has suggested that relatively low compliance rates were probably due to low rates of physician recommendation of the vaccine and low rates of past immunization. A large proportion of those who did not comply with vaccination often were not aware of the necessity of being vaccinated. Others were set against being vaccinated because they had a reaction to a former vaccination. Past immunization as a strongly significant factor indicates that patients in the habit of having an immunization probably will have one in the future, 113 114 especially if their physician suggests it. The majority of subjects (93.9%) reported that they had a doctor, and eighty percent made regular visits to him or her. Almost 42% of subjects made 2 to 6 checkups per year. These visits present opportunities for recommendations from physicians, as subjects verified (56.3%). Immunization compliance within race and gender groups was significantly associated with past immunization, doctor recommendation and intent to receive an immunization in the future. These findings highlight the similarities between these groups and is convincing evidence that similar stimuli increase compliance within the groups, and among all subjects. The Influenza Vaccination Demonstration Project in Ingham County appears to have been a success in this group of subjects. Vaccination coverage in this survey appears to be uniform over income and education groups. Compliance among all subjects did not vary geographically in this study. This was also true for the subjects broken down by race and gender groups, which may have been a result of small sample sizes within census tracts. This fact underscores the success of the IVDP in getting subjects who live in predominately black census tracts vaccinated. The results of this study seem to suggest that vaccination rates of black residents may not be as low as suspected by the official at the ICHD. 115 Recommendations to the Ingham County Health Department. The Ingham County Health Department (ICHD) seems to have successfully implemented the Influenza Vaccination Demonstration Project in Ingham County in this sample of residents. The U. S. Congress approved influenza vaccination as a covered benefit of Medicare B enrollees on May 1, 1993, to begin in the 1993-94 influenza season. This study has shown that physicians play an important role in the high compliance rates of this sample population. ICHD should continue to focus their educational programs on the physicians in the area. If the general population of elderly people in Ingham County have high rates of medical checkups, this presents an important opportunity for contacts with elderly residents. Physicians should be informed that a simple recommendation can make a significant difference in the decision to receive a vaccination. If vaccination differs significantly between males and females in Ingham County elderly residents, a plan must be devised to target female patients, especially those who have no immunization history or who report an unpleasant past experience with immunization. This study shows that subjects who had a past influenza immunization were more likely to take one in the future. Emphasis should be placed on getting Ingham County elderly residents vaccinated the first time starting at age 65. Subjects who visit their physician infrequently must be reached with information about the benefits of having a 116 vaccination against influenza. Medicare B enrollees who live in apartment complexes such as those described in this study would be much benefitted by having immunization clinics set up by ICHD in their buildings during the season for vaccinations. Black subjects in this study live in single family houses at significantly higher rates than did white subjects. If this is true of the entire elderly black population, these individuals who live in single family houses also must be reached. Eighty—five percent of black respondents reported that they attend church. If black elderly residents of Ingham County also attend church at rates as high as those in this study, ICHD would do well to avail themselves of the assistance black churches could give them to reach black elderly residents, especially females. Acquiring the cooperation of the major black churches in Lansing, and especially actively involving the pastors would be a way to reach black residents who live in houses and have infrequent visits to the physician. Large proportions of both black and white non-compliers believe the influenza immunization to be unnecessary. The ICHD would do well to point their efforts at education toward these individuals. These individuals did not appear adamant in their refusal to receive the vaccine, but are merely ignorant of the liabilities of not receiving it. Physicians could be instrumental in underscoring the benefits of having an influenza immunization, assuaging the 117 fears of those who have had reactions to the former live vaccines, and to repeat these messages to the hardcore non- compliers on a yearly basis. ICHD may wish to consider compiling demographic information about elderly residents who are vaccinated and their census tracts of residence in order to more effectively target underrepresented groups and individuals for vaccination. Recommendations for Further Research. Several changes would make this study a more effective one. If addresses of elderly residents were available, a random sample, stratified based on race, income, gender, etc. would make the results of a study representative of the population. A much larger sample would facilitate an indepth analysis of risk factors and compliance within census tracts, and a more meaningful study of geographic variation could be performed. A study in which a random, proportionate, stratified sample of residents were interviewed in each of the 50 census tracts in Lansing might yield results which were representative of the urban population of Ingham County. In future studies, subjects could also be questioned more thoroughly about non-compliance attitudes and behaviors in an attempt to discern ways of counteracting these attitudes against compliance with influenza immunization. This study did not focus on the compliance behaviors of those who attend senior citizen centers. A study focussing 118 on these individuals may reveal different patterns and may present opportunities for locations for educational programs on influenza immunization. The evidence of this study suggested that relatively low compliance in Tract 36.01 might be due to the relatively young age of the subjects interviewed and the small overall proportion of elderly residents residing in that tract. Future research could determine if influenza immunization compliance among census tracts in Ingham County differs significantly according to the proportion of elderly population residing in it and the factors associated with those differences. APPENDICES APPENDIX A Introduction Dear Lansing Resident, As part of my masters degree in geography at Michigan State University, I wish to conduct short personal interviews with persons aged 65 years and older who are residents of the City of Lansing, Ingham County. I will be gathering information about whether you have received an influenza vaccination and certain factors which may relate to this subject. ALL of the information gathered will be kept in the strictest confidence and will not be associated with your name or address. The interview will take about 20 minutes of your time. I will be conducting these interviews during the next two months. Your participation in this project is completely voluntary. If this time is inconvenient please feel free to ask me to return at another time. The following people can be contacted at the Department of Geography at Michigan State University in order to confirm that my request is a legitimate one. Dr. Judy Olson, Chairperson Dr. John M. Hunter, Professor Department of Geography, MSU Department of Geography, MSU 315 Natural Science Building 311 Natural Science Building East Lansing, MI 48824 East Lansing, MI 48824 517-355-4649 517-355-1685 Your cooperation is greatly appreciated. Sincerely, Karen D. Johnson-Webb Graduate Student Department of Geography Michigan State University 119 APPENDIX B Consent Form 1. I have freely agreed to take part in a scientific study being conducted by Karen Johnson-Webb, who is a graduate student in the Geography Department of Michigan State University, working under the guidance of Dr. John M. Hunter. I will be taking a survey that asks questions about whether I have received an influenza vaccination and other personal information about myself. 2. This survey has been explained to me, and I understand the explanation that has been given and what my role in the survey will be. 3. I understand that I am free to stop my participation in the survey at any time. I may refuse to answer any question. I understand that the expected length of the survey will be about 20 minutes. 4. I understand that the results of my participation in the survey will be kept in the strictest confidence, as will those of all other participants. All participants will remain anonymous in the reporting of results. Results will be made available to me at my request. 5.I understand that, at my request, I can receive additional explanation of this study after my participation is completed. Signed DATE: 120 APPENDIX C IMMUNIZATION QUESTIONNAIRE Tract# Blkgrp# Address Gender F(1) M(2) 1.What is your race? B(0) W(1) 2.House(0) Apt(1) DupCondo(2) SSCom(3) Nursing Home(4) 3.Did you receive an influenza vaccination last year? 1 2 If no go to #6. 4.Where? Priv.Dr.(0) Hlth Dept(1) SpecClinic(2) Other(3) . Name 5.Did your Dr. recommend that you get a flu shot? 1 2 6.What is the reason you did not receive a flu shot? Dr.Advice(0) Became ill(1) Painful shot(2) Not necessary(3) Don’t know(4) Other(S) 7.Have you ever received a flu shot? 1 2 8.Do you plan to receive a flu shot in the future? 1 2 9. If not, why? 0 1 2 3 4 Other(5) 10.Do you visit the Dr. even when you are feeling well and are not sick? 1 2 11.How often? 12.Where do you see the Dr? Priv.Dr.(0) HlthDept(1) SpecClinic(2) Other(3) Name 13.Do you have any kind of health insurance? 1 2 14.What kinds? Medicare(0) Private(1) HMO(2) Other(3) None (0) 15.Does it cover your medical expenses when you are sick? None(0) All(1) Some(2) DK(3) PERSONAL INFORMATION 16.How old were you at your last birthday? 17.What was the highest year of school you completed? . 18.Are you currently: mark all that apply. Employed for wages(1) Student(4) Self-employed(2) Homemaker(5) Retired(3) 19.Are you:M(0)D(1)W(2)Sep(3)NeverM(4)Member unmarried couple(5)? 20.Do you live alone? 1 2 21.How many people live with you? Relationships? Spouse(0)__ Sibling(1)___ Child(2)__ Grandchild(3)___ Other 22.Are any of these individuals dependent on you for their support? 1 2 How many? 23.Do you perform volunteer work? 1 2 Where? 24.Do you drive? 1 2 25.Do you own a car? 1 2 121 122 26.Is there someone who drives you where you need to go? 1 2 27.Do you ride the city bus? 1 2 28.Taxi or shuttle? 1 2 29.Do you attend church? 1 2 How often? /mo Where? 30.A senior center? 1 2 How often? /wk Where? 31.Member of a social club or civic org? 1 2 /mo Name 32.Are you a member of the AARP? l 2 Since 33.Source of your income? Pension(0) Saving/invest(1) Earnings(3) SSI(4) Other(5) 34.How many persons contribute to this income? 35.How much is your income? /mo iyr 123 APPENDIX D Responses to Questionnaire Question# Category % of Responses Tract# 5 12.7 13 7.6 15 24.4 16 14.2 36.01 21.8 37 19.3 Gender Female 68.5 Male 31.5 Race Black 42.6 White 57.4 Type of House 70.1 dwelling Apt. 0.5 Duplex 1.0 SSI Apts. 28.4 Vaccinated in Yes 56.3 1991 No 43.1 DK 0.5 Where vaccinated Physician 55.5 ICHD 22.7 SpecClinic 15.5 Other 4.5 DK 1.0 Dr. recommend. Yes 52.8 of vaccine No 47.2 Reason not Dr.Advice 4.1 vaccinated Became ill 8.6 Painful Shot 1.0 Not Necessary 13.2 DK 1.0 Other 15.7 Vaccinated in Yes 70.1 the past No 28.9 DK 1.0 Intend future Yes 55.8 vaccination No 37.6 DK 6.6 9. 10. 12. 13. 14. 15. 16. l7. 18. 19. 20. Reason for no intention Regular medical checkups Physician office location Med. insurance Type of insurance Coverage of medical costs Age Education Employment Marital status Live alone 124 Dr. Advice Became ill Painful shot Not Necessary DK Other Yes No Private Dr. ICHD SpecClinic Other Yes Medicare Private HMO Other None None All Some DK Most Employed Self-employed Retired Student Homemaker Married Divorced Widowed Separated NeverMarried MemberUnmarCouple Yes No 100. 100. 68. 12. 26. 23. 43. 82. 12. 37. 11. 47. 48. 51. NUTQOQO O U1UTOLA) UJQ OONOO vhf-‘OU'IO \1U1\IU1l-' U'II—‘OOOO GDN 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. Members of Family Dependents Volunteer Drive Own auto Have a driver Ride the bus Use special. transportation Attend church Attend senior center Member of a social club Member AARP Income source # contributing to income Amount of income 125 Spouse Sibling Child Grandchild Other Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Pension Savings Earnings SSI Other 48. 30. 12. 12. 86. 39. 60. 59. 40. 62. 37. 50. 49. 18. 81. 21. 78. 68. 31. 16. 83. 25. 74. 54. 45. 64. 49. 13. 100. ONOQQO CDrP-QON 0Q \‘Iw APPENDIX E Churches Attended Church Name N= % No Church 67 50.7 Union Baptist 13 69.2 Friendship Baptist 6 36.4 Holy Cross 6 71.4 Mt. Zion Baptist 6 80.0 St. Mary’s 5 16.7 St. Casimir 4 66.7 Grace United Methodist 4 66.7 Westminster Presbyterian 4 40.0 Immaculate Heart 4 75.0 Original COG #2 4 75.0 Faith United Methodist 4 50.0 Bethlehem Lutheran 3 66.7 Holmes Rd. COC 3 33.3 Jehovah’s Witness (Waverly) 3 100.0 St. Paul’s Episcopal 3 0.0 Pentecostal Outreach Center 3 0.0 Christ United Methodist 2 0.0 Mt. Olive Baptist 2 50.0 Jehovah’s Witness (West) 2 100.0 Central United Methodist 2 100.0 Grace Lutheran 2 50.0 Inner City Bible Church 2 50.0 Mt. Hope Church (Flags) 2 50.0 Macedonia Baptist 2 100.0 Paradise Baptist 2 50.0 Unity 2 100.0 Bethany Baptist 1 100.0 Central Free Methodist 1 100.0 Delta Haven Free Methodist 1 0.0 Immanuel Baptist 1 0.0 Evangel Methodist 1 0.0 Fellowship for Today 1 0.0 First Spiritual 1 100.0 First United Methodist 1 0.0 Galilee Baptist 1 50.0 Good Shepherd Lutheran 1 100.0 Gospel Assembly 1 0.0 West Lansing COG 1 100.0 Lansing Christian Center 1 100.0 Lively Faith 1 100.0 Living Word 1 0.0 Mask Memorial CME 1 100.0 Mt. Hope United Methodist 1 0.0 Church of the Nazarene 1 100.0 New Jerusalem 1 0.0 New Mt. Calvary Baptist 1 100.0 H N m New St. Paul COGIC Olivet Baptist Penway COG Pilgrim Rest Baptist Plymouth Congregational Redeemer Lutheran Ressurection Robinson Memorial COGIC South Baptist United Universalist Zion Lutheran Lansing COGIC Out of Town H N \] mtJFJHPJFJHPJFJHFJFJH 100. 100. 100. 100. 100. 100. 100. 100. 66. \IOOOOOOOOOOOO Appendix F Physicians of Respondents Physician Name or Number of %Dr. %Immunization Office Building Subjects Recommen. Compliance No Physician 11 0.0 16.7 Ingham Medical Ctr. 25 47.8 65.2 Cedar St. HMO 16 64.7 64.7 Wm. Harrison 14 42.9 42.9 Ingham Professional Ctr. 9 37.5 62.5 Medical Arts Building 9 88.9 77.8 Medical Dental Building 9 62.5 75.0 St. Lawrence Hospital 7 66 7 66.7 MSU Clinical Center 7 57 1 57.1 901 E. Mt. Hope Ave. 6 60.0 60.0 Lansing Nephrology Grp. 6 60.0 60.0 1414 S. Pennsylvania Ave. 5 100.0 60.0 Lake Lansing Rd. HMO 4 66.7 66.7 Ingham Co. Health Dept. 4 50.0 75.0 Creyts Rd. HMO 3 33.3 100.0 Grand Ledge Clinic 3 33.3 33.3 Holt, MI 2 50.0 50.0 Family Medical Svcs. 2 0.0 0.0 Dr. C. Beals 2 100.0 100.0 Dr. R. Griswold 2 50.0 100.0 Dr. H. Roth 2 100.0 100.0 Dr. F. Hogan 2 0.0 0.0 Dr. C. Rapson 2 50.0 50.0 Dr. C. Dehlin 2 50.0 50.0 Dr. 0. Graesser 2 50.0 0.0 Dr. J. Neuman 2 0.0 0.0 Dr. R. McElmurry 2 50.0 100.0 Eaton Rapids, MI 2 50.0 50.0 Battle Creek Vet. Admin. 2 0.0 0.0 E. Lansing, MI 2 0.0 0.0 Ann Arbor Vet. Admin. 1 100.0 100.0 Dr. E. Lopez 1 100.0 100.0 Sparrow Hospital 1 0.0 0.0 Lansing General Hospital 1 100.0 100.0 Black Child&Family Instit. 1 100.0 0.0 Dr. F. Breen 1 100.0 0.0 Dr. J. Chandler 1 0.0 0.0 Dr. A. Cheblis 1 100.0 100.0 Chicago, IL 1 100.0 100.0 Dr. B. Bingham 1 100.0 100.0 Dr. S. Daniels 1 0.0 0.0 Dr. R. Darios 1 100.0 100.0 Dr. P. Davenport 1 100.0 0.0 Dr. G. Gliniecki 1 0.0 100.0 Dr. F. Horvath 1 0.0 0.0 Okemos MI 3 100.0 66.7 128 129 Olivet, MI Dr. W. Page-Echols Dr. L. Rawsthorne Dr. R. Swenson Not Ascertained 1 thFJHFJ woooo 100. 100. 58. \JOOOO B IBLI OGRAPHY Bibliography Aho, William R., "Participation of Senior Citizens in the Swine Flu Inoculation Program: An Analysis of Health Belief Model Variables in Preventive Health Behavior," Journal of Gerontology, 34, no. 2 (1979): 201. Barkdoll, Gerald L., "Influenza Vaccine and the Elderly- Low Demand from a High-Risk Group," Geriatrics, (June 1976): 34. Barker, William H., "Excess Pneumonia and Influenza Associated Hospitalization During Influenza Epidemics in the United States, 1970-78," American Journal of Ppblic Health, 76, (1986): 761. Barker, William H. and John P. Mullooly, "Influenza Vaccination of Elderly Persons," Journal of the American Medical Association, 244, no. 2 (1980): 2547. Barton, Mary B. and Stephen C. Schoenbaum, "Improving Influenza Vaccination Performance in an HMO Setting: The Use of Computer Generated Reminders and Peer Comparison Feedback," American Journal of Public Health, 80, no. 5 (May 1990): 534. Bloom, Harrison G. et al, "Increased Utilization of Influenza and Pneumococcal Vaccines in an Elderly Hospitalized Population," Journal of the American Geriatrics Society, 36, no. 10 (October 1988): 898. Buchner, David M. et al, "The Relationship of Attitude Changes to Compliance With Influenza Immunization," Medical Care, 23, no. 6 (June 1985): 771. Clark, W. A. V. and P. L. Hosking. Statistical Methods for Geographers (New York, N. Y.: John Wiley & Sons, 1986). Ennis, Francis A., "Acceptance of Vaccination by the Elderly," in Influenza: Virus, Vaccines, and Strapegy: Proceedings of a Working Group on Pandemic Influenza, Rougemont, 26-28 January 1976, edited by Philip Selby, 311—318 : Academic Press. Fedson, David S., "The Influenza Vaccination Demonstration Project: An Expanded Policy Goal," Infect Control Hosp Epidemiol, 11, no. 7 (1990): 357. "Influenza and Pneumococcal Immunization in Medical Clinics, 1971-1983," Journal of Infectious 130 ‘ 131 Diseases, 149, no. 5 (May 1984): 817. Forbes, Robert Cowan, "Immunization of the Elderly," Postgraduate Geriatrics, 84, no. 2 (August 1988): 217. Frank, John W. et al, "Influenza Vaccination in the Elderly: 1. Determinants of Acceptance," Canadian Medical Association Journal, 132 (February 1985): 371. Ganguly, Rama and David Cameron, "Factors Affecting Immunization Rate in a Cohort of Elderly Veterans: A Retrospective Pilot Study of Influenza Vaccine Compliance," Vaccine, 7 (October 1989): 462. Ganguly, Rama et al, "Reasons for Nonimmunization Against Influenza in the Aged," Journal of the American Geriatrics Society, 37 (1989). Gemson, Donald H. et al, "Differences in Physician Prevention Practice Patterns for White and Minority Patients," Journal of Community Health, 13, no. 1, (Spring 1988): 53. Ingham County. Health Department. "Ingham County Influenza Report, 1990 - 1991," Disease Control Office. Lansing, MI. Hennekens, Charles H. and Julie E. Buring, Epidemiology in Medicine ed. Sherry L. Mayrent (Boston\Toronto: Little, Brown and Company, 1987), XXX. Howells, C. H. L. et al, "Influenza Vaccination and Mortality from Bronchopneumonia in the Elderly," Lancet, (15 February 1975): 381. Hutchinson, Brian G. and Harry S, Shannon, "Effect of Repeated Annual Reminder Letters on Influenza Immunization Among Elderly Patients," The Journal of Family Practice, 33, no. 2 (1991): 187. Hutchinson, Brian G., "Measurement of Influenza Vaccination Status of the Elderly by Mailed Questionnaire: Response Rate, Validity and Cost," Canadian Journal of Public Health, 80 (July\August 1989): 271. Kurinczuk, J. J. and K. G. Nicholson, "Uptake of Influenza Vaccination by Patients with Serious Cardiac Disease," British Medical Journal, 299 (August 1989): 367. Larson, Eric B., "The Relationship Between Health Beliefs and a Postcard Reminder to Influenza Vaccination," Journal of Family Practice, 8, no. 6 (1979): 1207. 132 Lefkowitz, Alan et al, "Preventive Care in a Veterans Administration Continuity Clinic," Journal of Community Health, 15, no. 1 (February 1990): 7. Leirer, Von O. et al, "Increasing Influenza Vaccination Adherence Through Voice Mail," Journal of the American Geriatrics Society, 37, no. 12 (December 1989): 1147. Lui, K. J. and A. P. Kendall, "Impact of Influenza Epidemics on Mortality in the United States from October 1972 to May 1985," American Journal of Public Health, 77, (1981): 712. MacPherson, Brian V. et al, "A Cohort Study of Preventive Health Behavior: Swine Flu Vaccination Acceptance in Vermont," Delaware Medical Journal, 52, no. 8 (1980): 427. McDowell, Ian et al, "A Follow-up Study of Patients Advised to Obtain Influenza Immunizations," Family Medicine, 22, no. 4 (July\August 1990): 303. McKinney, W. Paul and Gary P. Barnas, "Influenza Immunization in the Elderly: Knowledge and Attitudes do not Explain Physician Behavior," American Journal of Public Health, 79, no. 10, (October 1989). Monto, Arnold S. and Helen W. Ross, "Swine Influenza Vaccine Program in the Community: Acceptability, Reactions and Responses," American Journal of Public Health, 69, no. 3 (1979): 233. Monto, Arnold S. et al, "Utilization and Acceptability of Influenza A/New Jersey/76 Virus Vaccine in Oakland County, Michigan," Journal of Infectious Disease, 136, supplement (December 1977): S693. Mullooly, John P., "Increasing Influenza Vaccination among High-Risk Elderly: A Randomized Controlled Trial of a Mail Cue in an HMO Setting," American Journal of Public Health, 77, no. 5 (May 1987): 626. Nicholson, Karl G. et al, "Influenza Vaccination of the Elderly: Perceptions and Policies of General Practitioners and Outcome of the 1985-86 Immunization Programme in Trent, UK," Vaccine, 5 (December 1987): 302. Patriarca, Peter A., "Use of Influenza Vaccine in Nursing Homes," Journal of the American Geriatrics Society, 33, no. 7 (July 1985): 463. 133 Perotta, D. M. et al, "Acute Respiratory Disease Hospitalizations as a Measure of Impact of Epidemic Influenza," American Journal of Public Health, 122, (1985): 468. Pyle, Gerald F. and K. David Patterson, "The Geography of Influenza," Focus, (Fall 1987): 16. Riddiough, Michael A., "Factors Affecting the Use of Vaccines: Considerations for Immunization Program Planners," Public Health Reports, 96, no. 6 (December 1981): 528. Rives, Norfleet W. and Anne Mooney, "Attitudes of Nonparticipation Toward the Swine Influenza Immunization Program in Southern Delaware," Delaware Medical Journal, 50, no. 10 (October 1978): 533. Rundall, Thomas G. and John R. C. Wheeler, "Factors Associated with Utilization of the Swine Flu Vaccination Program among Senior Citizens in Tompkins County," Medical Care, XVII, no. 2 (February 1979): 191. Schoenbaum, Stephen C., "Influenza Vaccine- Unacceptable or Unaccepted," American Journal of Public Health, 69, no. 3 (March 1979): 219. Setia Usha et al, "Factors Affecting the Use of Influenza Vaccine in the Institutionalized Elderly," American Geriatrics Society, 33, no. 12 (December 1985): 856. Siewart, Mavis et al, "High Rates of Immunization in a Nursing Home," American Journal of Infection Control, 16, no. 5 (October 1988): 228. Stehr-Green, Paul A. et al, "Predictors of Vaccination Behavior Among Persons Ages 65 Years and Older," American Journal of Public Health, 80, no. 9 (September 1990): 1127. Tranter, Alan W., "Acceptance of Vaccination by Doctors and the Public," in Influenza: Virus, Vaccines, and Strategy: Proceedings of a Working Group on Pandemic Influenza, Rougemont, 26-28 January 1976, edited by Philip Selby, 319-322 : Academic Press. U. S. Department of Health and Human Services, Public Health Service, Centers for Disease Control, "Prevention and Control of Influenza," Morbidity and Mortality Weekly Report, 41, no. RR-9 (15 May 1992). 134 "Influenza Vaccination Levels in Selected States Behavioral Risk Factor Surveillance System, 1987" Morbidity and Mortality Weekly Report, 38, no. 8 (3 March 1989): 124. "Adult Immunization: Knowledge, Attitudes, and Practices- DeKalb and Fulton Counties, Georgia, 1989," Morbidity and Mortality Weekly Report, 37, no. 43 (4 November 1988): 657. U. S. Department of Commerce. Economics and Statistics Administration. Poverty in the United States: 1991. Current Population Reports, Consumer Income, Bureau of the Census, Series P-60, no. 181, (August, 1992). Wakefield, I. R., "Influenza Vaccination and the Elderly," British Medical Journal, 309 (1990). Wyngaarden, James B. et al, eds. Cecil Textbook of Medicine. (Philadelphia: Harcourt Brace Jovanovich, 1991), 1815.