UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Differences between participants and nonparticipants of a community-based senior health education program Berg, Shannon 1992

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1992_fall_berg_shannon.pdf [ 9.6MB ]
JSON: 831-1.0098791.json
JSON-LD: 831-1.0098791-ld.json
RDF/XML (Pretty): 831-1.0098791-rdf.xml
RDF/JSON: 831-1.0098791-rdf.json
Turtle: 831-1.0098791-turtle.txt
N-Triples: 831-1.0098791-rdf-ntriples.txt
Original Record: 831-1.0098791-source.json
Full Text

Full Text

DIFFERENCES BETWEEN PARTICIPANTS AND NONPARTICIPANTSOF A COMMUNITY-BASED SENIOR HEALTH EDUCATION PROGRAMbySHANNON BERGB.Sc.O.T., The University of Alberta, 1982A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF SCIENCEinTHE FACULTY OF GRADUATE STUDIESDEPARTMENT OF HEALTH CARE AND EPIDEMIOLOGYHEALTH SERVICES PLANNING AND ADMINISTRATIONWe accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIASeptember 1992© Shannon Berg, 1992In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of myor by his or her representatives. It is understood that copying orof this thesis for financial gain shall not be allowed without my writtendepartmentpublicationpermission.(Signature)Department of  HEALTH CANE AND EPIDEMIOLOGYThe University of British ColumbiaVancouver, CanadaDate OCTOBER 1 0 , 1992DE-6 (2/88)ABSTRACTThe purpose of this study was to determine which factors, if any, are associatedwith participation in a formal seniors' health education program. It was hypothesisedthat three identifiable groups, two of which would be nonparticipants of the healtheducation program of interest, would be found. One group of nonparticipants wouldbe very active, and score more favourably than the other two subsamples on multi-dimensional health scales. This group were expected to be actively engaged in asatisfying lifestyle, and would perceive no need for the health education program. Asecond group of nonparticipants was expected to score less favourably than programparticipants on scales measuring multi-dimensional health and to be less active in thecommunity.The second nonparticipant group was of primary interest to this study, as lowerfunctioning on health and social support scales might constitute both risk factors whichwould target this group as possible participants, and identify barriers to theirparticipation in the program.One hundred eighty respondents were surveyed from three sources: (1)participants of a seniors' health education program based in a senior centre; (2)residents of single family dwellings and (3) apartment buildings in a five-block radiusaround the senior centre. Respondents completed a standardised tool to measurephysical, mental and social health, and a series of questions relating to communityinvolvement and participation in the health education program of interest.The 63 program participants were found to be younger, and to have higherscores on measures of physical health and mobility, than nonparticipants.Nonparticipants were found to fall into two groups. One group was younger, hadhigher scores on measures of health, and was more likely to choose predisposingreasons for not attending the health education program, such as "Not interested" andii"Don't need it". A second group of nonparticipants was found to be older, with lowerscores on measures of health and social support. This second group ofnonparticipants was more likely to choose enabling or reinforcing reasons for notattending, such as "Poor health", "Can't do exercise", "No transportation", "Can't affordit", or "Don't know anyone there".The information gained in this study suggests than direct outreach in some formwould be required to recruit the group of nonparticipants who are identified in theliterature as the target population for health education programs; that is, those who areeconomically or socially disadvantaged, older, less mobile or have chronic healthproblems.iiiTABLE OF CONTENTSABSTRACT^ iiLIST OF TABLES viiiLIST OF FIGURES^ ixLIST OF APPENDICES xACKNOWLEDGEMENTS^ xiCHAPTER 1: INTRODUCTION1.0 STATEMENT OF PROBLEM^ 11.1 PURPOSE OF THE STUDY 31.2 STUDY SIGNIFICANCE^ 31.3 CONCEPTUAL FRAMEWORK 41.4 OPERATIONAL DEFINITIONS^ 4CHAPTER 2: LITERATURE REVIEW2.0 HEALTH EDUCATION AND HEALTH PROMOTION DEFINED^72.1 MODELS OF HEALTH EDUCATION^ 112.2 HEALTH EDUCATION FOR OLDER ADULTS^ 152.3 COMMUNITY-BASED HEALTHPROMOTION/EDUCATION PROGRAMS FOR THE ELDERLY^172.4 EFFECTIVENESS OF SENIOR HEALTH PROMOTION PROGRAMS 212.5 TARGET POPULATION FOR SENIOR HEALTHPROMOTION PROGRAMS^ 252.6 CHARACTERISTICS OF PARTICIPANTS OFHEALTH PROMOTION PROGRAMS^ 282.7 SENIOR CENTERS AS LOCATIONSFOR HEALTH EDUCATION/PROMOTION^ 32iv2.8 CHARACTERISTICS OF SENIOR CENTER ATTENDERS^332.9 SUMMARY^ 36CHAPTER 3: METHODOLOGY3.0 OBJECTIVE^ 393.1 RESEARCH QUESTIONS^ 393.2 THE PROGRAM^ 403.2.1 Choice of Program^ 403.2.2 Description of Program 403.2.3 Demographics of the Area^ 423.3 RESEARCH DESIGN^ 443.3.1 Subjects 443.3.2 Choice of Questionnaire^ 463.3.3 The Iowa Self-Assessment Inventory^ 523.3.4 Questions Specific to the Kerrisdale Health Drop-In^563.4 SPECIFIC HYPOTHESES^ 613.5 ANALYSIS^ 633.6 PILOT OF QUESTIONNAIRES^ 64CHAPTER 4: RESULTS4.0 DATA COLLECTION^ 684.1 DETERMINATION OF PARTICIPANTS AND NONPARTICIPANTS 704.2 DEMOGRAPHIC CHARACTERISTICS^ 724.2.1 Age and Place of Residence 754.2.2 Age and Participation^ 754.2.3 Gender and Place of Residence^ 754.2.4 Gender and Participation 764.2.5 Education and Place of Residence 764.2.6 Education and Participation^ 764.2.7 Living Arrangement and Place of Residence^774.2.8 Living Arrangement and Participation^ 77v4.3 KNOWLEDGE OF THE HEALTH DROP-IN^ 774.3.1 Knowledge and Previous Attendance AmongNonparticipants^ 774.3.2 Source of Knowledge 784.4 LENGTH OF PARTICIPATION^ 794.5 PROGRAM COMPONENTS^ 804.5.1 Past and Present Participants^ 804.5.2 Current Participants^ 824.6 FREQUENCY OF ATTENDANCE 834.7 ISAI SCORES^ 844.7.1 Economic Resources (ER)^ 864.7.2 Emotional Balance (EB) 874.7.3 Physical Health (PH) 874.7.4 Trusting Others (TO)^ 884.7.5 Mobility (MO)^ 884.7.6 Cognitive Status (CS) 894.7.7 Social Support (SS)^ 904.8 COMMUNITY INVOLVEMENT 904.8.1 Other Activities at Kerrisdale Senior Centre^914.8.2 West Main Health Unit^ 924.8.3 Church^ 934.8.4 Church Groups and Organisations^ 934.8.5 Cultural Centres or Organisations 944.8.6 Symphony, Opera, etc.^ 954.8.7 Recreational Activities 954.8.8 Other Senior Centres 964.8.9 Community Centre^ 974.8.10 Other^ 984.9 PATTERNS OF ACTIVITY^ 984.9.1 Total Hours of Activity 1014.9.2 Household Activity 1024.9.3 Exercise^ 1034.9.4 Hobbies 1044.9.5 Visiting Friends and Relatives^ 1044.9.6 Community Activities^ 1054.9.7 Gardening^ 1064.9.8 Sleep 107vi4.10 REASONS FOR ATTENDING THE HEALTH DROP-IN^1084.10.1 Predisposing Reasons^ 1124.10.2 Enabling Reasons 1154.10.3 Reinforcing Reasons 1164.11 REASONS FOR NOT ATTENDING THE HEALTH DROP-IN^1184.11.1 Predisposing Reasons^ 1204.11.2 Enabling Reasons 1244.11.3 Reinforcing Reasons 1264.12 SUMMARY OF RESULTS^1294.12.1 Hypothesis 1 1294.12.2 Hypothesis 2 1304.12.3 Hypothesis 3^132CHAPTER 5: DISCUSSION AND CONCLUSIONS5.0 LIMITATIONS^ 1375.1 DISCUSSION RE METHODOLOGY ISSUES^1405.1.1 Return Rates^ 1405.1.2 ISAI Trusting Others Scale^ 1415.2 DISCUSSION OF RESULTS 1415.2.1 Place of Residence as a Possible Confounder^1415.2.2 Demographic Characteristics and Participation 1425.2.3 Measures of Health^1445.2.4 Community Activity 1465.2.5 Reported Hours of Activity 1475.2.6 Reasons for Attendance or Nonattendance^1495.3 IMPLICATIONS OF RESULTS^1515.4 SUGGESTIONS FOR FURTHER RESEARCH^154THESIS REFERENCES^158viiLIST OF TABLESTABLE 1: CHARACTERISTICS OF PILOT STUDY SAMPLE ^ 63TABLE 2: RETURN RATES FOR STUDY ^ 66TABLE 3: DEMOGRAPHIC CHARACTERISTICS BY PLACE OF RESIDENCE ^ 70TABLE 4: DEMOGRAPHIC CHARACTERISTICS BY PARTICIPATION ^ 70TABLE 5: DEMOGRAPHIC CHARACTERISTICS BY PROGRAM COMPONENT • ^ 71TABLE 6: DEMOGRAPHIC CHARACTERISTICS OF ATTENDERS WHODID AND DID NOT PERCEIVE THEMSELVES AS PARTICIPANTS ^ 71TABLE 7: KNOWLEDGE OF HEALTH DROP-IN AND PASTATTENDANCE AMONG NONPARTICIPANTS ^ 77TABLE 8: ATTENDANCE PATTERNS OF PROGRAM PARTICIPANTS ^ 79TABLE 9: MEAN ISAI SCORES BY PLACE OF RESIDENCE 81TABLE 10: MEAN ISAI SCORES FOR PARTICIPANTS AND NONPARTICIPANTS  ^82TABLE 11: ISAI SCORES FOR PROGRAM COMPONENT ATTENDERS ^ 82TABLE 12: MEAN ISAI SCORES OF ATTENDERS WHO DID ANDDID NOT PERCEIVE THEMSELVES AS PARTICIPANTS ^ 82TABLE 13: PERCENTAGE OF COMMUNITY INVOLVEMENT AMONGRESPONDENTS BY DEMOGRAPHIC CHARACTERISTICS ^ 87TABLE 14: COMMUNITY ACTIVITY PATTERNS BY PARTICIPATION 88TABLE 15: MEAN WEEKLY HOURS OF ACTIVITY BY DEMOGRAPHIC GROUP ^ 96TABLE 16: MEAN REPORTED WEEKLY HOURS OF ACTIVITY BY PARTICIPATION •^•^• 96TABLE 17: MEAN WEEKLY HOURS OF ACTIVITY BY PROGRAM COMPONENT 96TABLE 18: REASONS GIVEN FOR ATTENDANCE ^ 107TABLE 19: CHARACTERISTICS OF ATTENDERS CHOOSING AT LEASTONE PREDISPOSING, ENABLING OR REINFORCING REASON ^ 108TABLE 20: PARTICIPANT AGE AND ISAI RESULTS BY NO. OF PRED. REASONS  ^ 110TABLE 21: PARTICIPANT AGE AND ISAI RESULTS BY NO. OF ENAB. REASONS  ^112TABLE 22: PARTICIPANT AGE AND ISAI RESULTS BY NO. OF REINF. REASONS  ^114TABLE 23: REASONS GIVEN FOR NONATTENDANCE ^ 116TABLE 24: CHARACTERISTICS OF NONATTENDERS CHOOSING AT LEASTONE PREDISPOSING, ENABLING OR REINFORCING REASON ^ 117TABLE 25: NONPARTICIPANT AGE AND ISAI RESULTS BY NO. OF PRED. REASONS -119TABLE 26: NONPARTICIPANT AGE AND ISAI RESULTS BY NO. OF ENAB. REASONS -121TABLE 27: NONPARTICIPANT AGE AND ISAI RESULTS BY NO. OF REIN. REASONS-- 123TABLE 28: MEAN AGE AND ISAI SCORES FOR PARTICIPANTS ANDFOR NONPARTICIPANTS BY REASON ^ 131TABLE 29: COMPARISON OF SELECTED CHARACTERISTICS BETWEEN:COLUMN A: PARTICIPANTS AND NONPARTICIPANTS BELONGINGTO THE TARGET GROUP COLUMN B: PARTICIPANTS AND NON-PARTICIPANTS NOT BELONGING TO THE TARGET GROUP ANDCOLUMN C: NONPARTICIPANTS BELONGING AND NOT BELONGINGTO THE TARGET GROUP ^ 157VIIILIST OF FIGURESFIGURE1: CONCEPTUAL SCHEME SHOWING THE RELATIONSHIPSBETWEEN PARTICIPATION IN A HEALTH EDUCATIONPROGRAM AND VARIABLES AFFECTING PARTICIPATION^38ixLIST OF APPENDICESAPPENDIX A: ORDER OF VISITATION OF KERRISDALE BLOCKS^170APPENDIX B: IOWA SELF-ASSESSMENT INVENTORY ANDITS SCORE PROFILE^173APPENDIX C: STUDY QUESTIONNAIRE^176APPENDIX D: LIST OF "OTHER" ACTIVITIES SPECIFIED BYRESPONDENTS AS COMMUNITY ACTIVITIES^187APPENDIX E: PROFILE OF PARTICIPANTS BY REASON CHOSEN^190APPENDIX F: PROFILE OF NONPARTICIPANTS BY REASON CHOSEN^193xACKNOWLEDGEMENTSMy sincere thanks go to my thesis supervisor, Dr. Samuel Sheps, for his interestand encouragement throughout the process of formulating, researching and writingthis thesis study. Thanks also go to my thesis committee members, Dr. LawrenceGreen and Dr. James Thornton for suggestions, assistance and encouragement, andto Dr. Brenda Morrison for consultation regarding statistical analysis.To the people at West Main Health Unit, especially Dr. Veronica Doyle, SarahBoyd and Judy Curran, and to Grace Pirie at Kerrisdale Senior Centre, thank you foryour assistance, cooperation and support. Special thanks are due the members of theConcerned Citisens for Affordable Housing, who cheerfully acted as my pilot group,and to the many apartment managers, especially John Tufts, who volunteered theirtime to assist with the study.The cooperation and support of the participants of Kerrisdale Senior CentreHealth Drop-In is much appreciated. To these participants, and to the residents ofKerrisdale who participated in the study, thank you.I would particularly like to thank my beloved and very supportive husband,Terry, and my children, Karl, Kyra and Mark, who pitched in and helped in order tomake the thesis possible.This research was supported in part by the National Health Research andDevelopment Program (NHR DP) through National Health Fellowship #6610-1946-47.xiCHAPTER 1INTRODUCTION1.0 STATEMENT OF PROBLEMThe increasing proportion of elderly people (Health and Welfare Canada, 1989)and increasing costs of medical care (Cox and Monk, 1989) have been credited withinfluencing the development of health education and promotion for older people. Theelderly are high users of health services, as they suffer from more chronic disease thanyounger persons (Butler, Gertman, Oberlander and Schindler, 1979). Healtheducation and promotion are viewed as a means of decreasing costs of care throughprevention of illness and decreasing institutionalisation (Schneider, Chapman andVoth, 1985).A successful health promotion program must be designed to meet the needs ofits target population (Fallcreek, Warner-Reitz and Mettler, 1986). The majority ofcommunity-based health education programs involve voluntary registration byparticipants. Suggested recruitment procedures include advertisements at seniorcentres, personal letter from physicians, word-of-mouth recruitment by pastparticipants and the media (Simmons, Roberts and Nelson, 1986). With self-selectionof participants, however, there is limited control over characteristics of the participantpopulation. Some sources suggest use of a pre-enrollment screening assessment toensure that course participants are appropriate, or to allow instructors to respond tothe needs of participants (Simmons, et. al., 1986; Fallcreek et. al., 1986).The issue of whether the appropriate individuals are registering for the programis not addressed, however, by pre-screening. It may be that individuals who choose toparticipate are those who have the greatest need for the intervention. It is alsopossible that those individuals who choose to, or are able to participate in the programare not those who might obtain the greatest benefit from the intervention.Mass enrollment campaigns are frequently successful in recruiting highnumbers of participants, but cannot locate the people who are most in need of theservice, particularly isolated older people (Stuen, 1985). Factors such as activitylevel, financial status, existence of a chronic health problem or perceived health statusmay prevent or discourage appropriate individuals from accessing the services ofcommunity-based programs (Toseland and Sykes, 1977). Health problems andneeds of seniors have been related to other needs in the social, financial, nutritionaland educational spheres (Leanse, 1986). The very problems which identifyindividuals as part of the target population for the intervention, therefore, may act asbarriers to their participation. Outreach, defined as an effort to link people in need withthe appropriate community resources, is often a necessary step in reaching isolatedolder persons (Stuen, 1985).Recently in senior centres, emphasis has been placed upon providing a widerrange of programming, including health services and education (Hanssen et. al.,1978). It is not clear, however, that this shift has been accompanied by outreach, orthat the new programming has involved older people who were not previouslyinvolved with senior centres (Hanssen et. al., 1978). A lack of integration of frailelderly into senior centres has been observed, along with a feeling among seniorcentre directors that they are unable to accomodate frail older people in proportionsgreater than ten percent of their total participants (Cox and Monk, 1990).If health education programs are not reaching the individuals who could mostbenefit from the services, then the identification of variables that act as barriers mayassist with enhancement of program design, recruitment strategies and supportservices in order to better serve the target population. In order to be effective,programs must match the intervention strategies with the needs and resources of thepeople the programs are intended to serve (British Columbia Ministry of Health, 1989).Patterns and correlates of service use must be understood by program planners in-2-order to reach the intended target population and, subsequently, provide effectiveservice (Krout, 1983).1.1 PURPOSE OF THE STUDYThe purpose of this study is to determine which factors, if any, correlate withparticipation in a formal seniors' health education program. More specifically, itexplores potential significant differences in demographic, health, psychosocial anddaily activity pattern characteristics between participants and nonparticipants of aspecific community based health education program.1.2 STUDY SIGNIFICANCEBased upon the literature and discussions with professionals in West MainHealth Unit, it is hypothesised that this study will find evidence of three identifiablegroups, two of which will be nonparticipants of the health education program ofinterest. One group of nonparticipants may be very active, and score more favourablythan the other two subsamples on scales measuring economic resources, emotionalbalance, level of trust, mobility, cognitive status and social support. This group will beactively engaged in a satisfying lifestyle, and may perceive no need for the healtheducation program.It is hypothesised that a second group of nonparticipants will be more likely tobe male, and score less favourably on emotional balance, physical health, level oftrust, mobility, cognitive status and social support scales than program participants.The second nonparticipant group is of primary interest to this study. If such agroup is found, the lower scores on the specified scales may constitute both riskfactors which would target this group as possible participants, and identify barriers totheir participation in the program. The information gained in this study might provehelpful in enhancing programming to reach this group of nonparticipants, who areunable or unwilling to participate at present but could benefit from the program.-3-1.3 CONCEPTUAL FRAMEWORKLaLonde's publication of "A New Perspective on the Health of Canadians"(1974) is partially responsible for a recent widespread change in outlook towardhealth care delivery, with greater emphasis being placed on health promotion, self-help and illness prevention (Green, Kreuter, Deeds and Partridge, 1980). The growingproportion of older people, movement toward independence rather than illness cureas a goal of health service delivery and publicity received by health promotion"successes" (Weiler, 1986) are also factors in the rise of health promotion programs forthe elderly.It is not desirable, however, to provide health education without channeling theprograms to those who can benefit from them. Past research has demonstrated that"high risk groups and optimum age ranges can be clearly identified for most preventiveand self-care actions which health education can facilitate" (Green, 1978, p.30).1.4 OPERATIONAL DEFINITIONSActivities of Daily Living (ADL): functions which are essential to a person'sability to do their self-care (eg. dressing, bathing, feeding) (Spector, Katz, Murphy andFulton, 1987).Health education: "any combination of learning experiences designed to facilitatevoluntary actions conducive to health" (Green and Kreuter, 1991, p. 17). It consists ofplanned activity which promotes health or illness-related learning, and may producechanges in understanding or ways of thinking, bring about some shift in belief orattitude, influence or clarify values, facilitate the acquisition of skills or even effectchanges in behaviour or life-style. (Tones, 1990).Health promotion: "any combination of educational, informational, organisational,service, legislative and regulatory, and/or economic methods designed to motivatehealth-related behaviour changes among a target population in a specifiedgeographic area during a defined time period". (Windsor, 1986, p. 436).-4-Iowa Self-Assessment Inventory (ISAI): a 56-item self-report instrumentdesigned to measure the resources, needs, statuses and abilities of older persons. Itis intended for use either with individuals or to gain a better understanding of groups ofelderly people through survey research methods (Morris et al., 1991)Community-based health education for the well elderly: programs designedto provide health maintenance activities and early detection of disease (Moneyhamand Rider, 1988) in order to prevent illness in older people, assist them to achieve theability to manage independently their chronic disease and provide them withinformation to become more effective consumers of health services (Rimer, Keintz,Glassman and Kinman, 1986). In Vancouver, such programs generally consist ofweekly meetings organised around a format of exercise, social interaction and healthdiscussion (Martin, 1986).Multi-dimensional measures of health: comprehensive evaluations of health,which measure distinct dimensions such as "physical health, mental health, everydayfunctioning in social and role activities, and general perceptions of well-being" (Ware,1987, p. 474).Participant: a current attender of the health education program, who has attendedmore than one session (not a first time attender).Senior centre: "a community focal point on aging where older persons asindividuals or in groups come together for services and activities which enhance theirdignity, support their independence and encourage their involvement in and with thecommunity.""As part of a comprehensive community strategy to meet the needs of olderpersons, senior centre programs take place within and emanate from a facility. Theseprograms consist of a variety of services and activities in areas such as education,creative arts, recreation, advocacy, leadership development, employment, health,nutrition, social work and other supportive services."-5-"The centre also serves as a community resource for information on aging, fortraining professional and lay leadership and for developing new approaches to agingprograms" (In Leanse, 1986, p. 105-6).Well elderly: people who are 55 years of age or older, not living in a nursing homeor other long-term care facility (adapted from Lalonde, Hooyman andBlumhagen,1988).CHAPTER 2REVIEW OF THE LITERATURE2.0 HEALTH EDUCATION AND HEALTH PROMOTION DEFINEDCentral to any definition of health education and promotion is the concept of"health". The traditional, and most common, definition of health is "the absence ofdisease or the ability to perform vital functions normally" (MacRae, 1986). This is inkeeping with the word's roots, as "health" is derived from the Anglo-Saxon word"haelth", meaning safe, sound or whole (Shirreffs, 1984.) Such a definition treatshealth as a static state. One is either healthy or unhealthy.The concept of health includes lack of disease, which Sherriffs (1984) definesas "largely a scientific, medical and technical territory encompassing what is knownabout biologic impairment" (p. 40). Absence of illness, viewed as "culturally defined;the territory in which social judgment defines that deviance for which the physician andother health practitioners are considered the official remedial agents", is alsoencompassed under the umbrella of health (Shirreffs, 1984, p. 40).Contemporary views of health, however, treat it as a process rather than a staticstate (Shirreffs, 1984). It is seen as multi-dimensional, including physical and mentalstates, the capacity to perform normal daily living activities, and the social, economicand environmental resources required to maintain a quality life style (Duffy andMacDonald, 1990, Kane and Kane, 1987). Health is perceived as encompassing notonly a lack of disease and illness, but also the positive aspects of "wellness" (Bergner,1985). Wellness has been characterised as the "attitudes and activities which improvequality of life and expand the potential for higher levels of functioning" (Mullen, 1986,p. 34). Efforts to achieve wellness are motivated by a desire to improve quality of life,rather that a desire to avoid disease (Mullen, 1986). Health is thought of as a qualityor a state of mind, rather than a condition (Sherriffs, 1984, p. 41).-7-Bergner (1985) identified five dimensions of health status: genetic or inheritedcharacteristics, biochemical/physiological/anatomic characteristics, functionalcondition including performance of activities of daily living (ADL), mental stateincluding self-perception of mood and emotion, and health potential (longevity,prognosis of disease, functional potential). A somewhat different approach toidentification of health dimensions by Ware (1987) includes physical health, mentalhealth, social functioning, role functioning and general perceptions of well-being. Thisphilosophy of health is captured in the definition proposed by WHO (1946):Health is a state of complete physical, mental and social well-being, and not merely theabsence of disease or infirmity (In Downie et. al., 1990, p. 9).The same change in focus is apparent within the discipline of health education,which has progressed from a negative approach in the late nineteenth and earlytwentieth centuries, when educators were known as "Warriors against pleasure", to apositive focus on quality of life (Rubinson and Alles, 1984).Current definitions of health education stress the voluntary involvement ofindividuals in learning activities which are meant to promote health (French, 1990;Green, 1978). Thus, Green and Kreuter (1991) define health education as "anycombination of learning experiences designed to facilitate voluntary actions conduciveto health" (p. 17). Acknowledgement of the multiple determinants of health is evidentin the use of "combination", which refers to the necessity to match health determinantswith appropriate interventions or supports. The use of the word "designed" is meant toencompass only systematically planned activities in a health education context.A more complex definition of health education, which incorporates theseelements is:Health education is any planned activity which promotes health or illness related learning,that is, some permanent change in an individuals's competence or disposition. Effectivehealth education may produce changes in understanding or ways of thinking; it may bringabout some shift in belief or attitude; it may influence or clarity values; it may facilitate theacquisition of skills; it may even effect changes in behaviour or life-style (Tones, 1990).-8-French (1990), in addition, believes that health education should be concernedwith the provision of high quality information in a readily understandable form, and thatsocial and economic determinants of health should not be neglected. He supports acommunity empowerment view of health education, which aims at enabling andsupporting individuals or groups to set and implement their own health agendas. Thisis in accord with the widely held belief that community health programs are mosteffective when they are established by the community (B.C. Research, 1986).Shirreffs (1984) lists several qualities of health education, including thefollowing:• a professional field and academic discipline,• strategies which provide a bridge between scientific discovery and applicationfor everyday healthful purposes,• an integral part of school curriculum and community based health programs,• contribution to total education through providing meaningful experiences thatcan positively influence health behaviour,• principles and strategies which are based upon and improved through basicand applied research,• facilitates primary prevention of health problems. (p. 41).It has, as its goal, to maintain, promote and improve individual and communityhealth through educational processes (Shirreffs, 1984).Health promotion grew out of health education (Green and Kreuter, 1991) andhas wider boundaries. It includes, but is not limited to, the voluntary activities of healtheducation (Green and Kreuter, 1991) and can consist of any planned activities aimedat healthy individuals that are intended to increase their health (Duncan and Gold,1986).The Canadian Public Health Association has defined health promotion as, "theprevention of physical and mental ill health through the promotion of better personal-9-health habits and the elimination of self-imposed risks" (in Labonte and Penfold,1981). Such a definition is limited, as it does not encompass environmental factors(eg. neighbourhood safety, pollution). It has been criticised for promoting an attitude of"victim blaming" (Labonte and Penfold, 1981, Minkler and Pasick, 1986). Broaderdefinitions which include these factors are the following:Health promotion is any combination of educational, informational, organisational,service, legislative and regulatory, and/or economic methods designed to motivatehealth-related behaviour changes among a target population in a specified geographicarea during a defined time period (Windsor, 1986, p. 436) orHealth promotion comprises efforts to enhance positive health and prevent ill-health, through the overlapping spheres of health education, prevention and healthprotection (Downie, Fyfe and Tannahill, 1990, p. 2).McGinnis (1982) identified five strategies which are aimed at promotion ofhealth:•Regulation - is generally used when the potential exists for behaviour of anindividual to impinge upon the rights of others. Some examples includelegislation of the use of seat belts, restriction of alcohol and drugs while driving,provision of smoke-free sections in public places.• Economic Incentives - such measures can be positive, such as tax breaks toindustries who engage in environmentally friendly practices, or negative, suchas increased insurance premiums for smokers.• Technology Development and Application - these can be innovations whichenhance health, such as less toxic cigarettes or computerised health riskappraisal forms, or environmental manipulations such as fluoridation of water.•Service Delivery - Fitness programs, nutritious meal programs, smokingcessation programs and telephone counselling are specific examples ofservices offered.-Education and Information - carefully designed efforts to provide informationsystematically to targeted populations in order to facilitate health behaviour-10-changes.The areas which are generally included under health education include someservice delivery and education and information.2.1 MODELS OF HEALTH EDUCATIONThere are many models used as a framework for health education andpromotion programs. Most are founded on social learning theory. Some alsoincorporate environmental factors into their design and at least one model appliesepidemiological and planning concepts.A well known and widely researched model of health education based on sociallearning theory is the Health Belief Model (Janz and Becker, 1984). This model wasdeveloped in the 1950's as a way to explain people's behaviour with regards topreventive screening tests, and was based upon Lewin's theory of positive, negativeand neutral valances (perceptions of value) as the determining factor for engaging in abehaviour (Rosenstock, 1974). Preventive behaviour was explained in terms of anindividual's perceptions about the relevant disease (perceived susceptibility to andseverity of the disease), which leads to their valuation of the threat posed by thedisease. The perceived threat is modified by demographic and sociopsychologicalvariables (age, sex, race, ethnicity, personality, peer pressure, etc.), which influencethe perceived benefits of the behaviour as well as barriers to action. However, in orderto engage in preventive action, the individual must also experience a cue to action (eg.media campaigns, advice, reminder by physician, etc.). The net result of theinteraction of all these factors is the degree of likelihood of the individual to engage inthe preventive behaviour (Janz and Becker, 1984; Maiman and Becker, 1974;Radecki and Cowel1,1990; Rosenstock, 1974).Research generated by the discipline of adult education is related tobehavioural models in that it focuses on individual motivation and requirements forlearning. Although not specific to health education, the concepts derived are useful ina wide range of programming, including health. It is thought that learning outcomeswith older individuals depend upon many factors, including learning capabilities,previous learning experiences, and the social contexts in which thought andbehaviour have developed (Thornton, 1986). Previous learning can, therefore,interfere with the acquisition of new knowledge. To facilitate transfer of learning (theextent to which a learning event contributes to or detracts from subsequent learning orproblem solving) instruction must be arranged so that relevant past learning canfacilitate the acquisition of new learning (near transfer) and new learning can be usedto deal with everyday problems or events (far transfer) (Uman and Richardson, 1986).In addition, older adults appear to learn more efficiently when they can control thepace of instruction, when attention is given to compensation for sensory deficits (eg.visual and auditory), and when the need for sufficient practice is balanced with thedetrimental effects of fatigue (Whitbourne and Sperbeck, 1982).Because educational experiences for older adults are often motivated by aproblem which must be solved, strategies must identify the problem and develop thenecessary knowledge, skills and attitudes to solve them. There is a need for initiallearning, practice and integration of the new knowledge for long term use (Uman andRichardson, 1986).Boshier (1991) has done a great deal of work in identifying the motivationalfactors for adults to participate in education. His Education Participation Scaleconsists of seven motivational factors for adult education attendance: communicationimprovement, social contact, educational preparation, professional advancement,family togetherness, social stimulation and cognitive interest. An earlier version of theEducational Participation Scale, used with a sample of retired older adults, found foursignificant factors; escape stimulation, social welfare, social contact and cognitiveinterest (Boshier and Riddell, 1978).An approach which incorporates environmental factors into a model is the-12-"Ecological Model for Health Promotion" (McLeroy, Bibeau, Steckler and Glanz, 1988).This model attributes the causes of individual behaviour to five sets of factors:• intrapersonal characteristics of the individual such as knowledge, attitudes,behaviour, self-concept, skills, etc.,• interpersonal processes and primary groups, which include the formal andinformal social support systems such as families, work associates and friends,• institutional factors, which include the organisational characteristics, rules andregulations (formal and informal) of social institutions,• community relationships between organisations, institutions and informalnetworks and• public policy, which includes regional, provincial and national laws andpolicies.It is assumed that the full range of possible strategies for health promotionprogramming is included in these five levels, as techniques must be based uponcurrent understanding of causes of behaviour. Each of the five levels can be usedappropriately in health promotion programming. For example, interpersonal factorsmay be used in educational programs, mass media, support groups or peercounselling, whereas public policy targets the health of populations rather thanindividuals.A model of health education which incorporates both social learning theory andenvironmental factors, but based in epidemiology, is Green and Kreuter's (1991, p. 22-31) PRECEDE-PROCEED model. It contains nine phases, and begins the planningprocess with the determination of desired outcomes. Phase 1, Social Diagnoses,involves subjective definition of the problems and priorities of the target community.The identification of "quality of life" issues leads to the second phase, epidemiologicaldiagnosis, in which specific health goals are identified and ranked. In the Behaviouraland Environmental phase, specific health related factors which can be linked to the-13-problems or goals chosen in the first two phases are distinguished. The fourth phase,Educational and Organisational Diagnosis, consists of determining the factors whichinfluence the health behaviours identified in phase 3. There are three broadcategories under which factors can be classified:• Predisposing Factors: include the knowledge, attitudes, beliefs, values andperceptions which can facilitate or inhibit motivation for change,• Enabling Factors: include the skills, resources and barriers that assist orobstruct behavioural and/or environmental change,• Reinforcing Factors: include the rewards and feedback received fromengaging in the health behaviour, which will encourage or discouragecontinuation of the behaviour.Once the factors which influence health behaviours are classified according tothese categories, it is possible to determine which of the categories deserves highestpriority as the focus of intervention. Administrative and policy diagnosis, phase 5,involves the determination of resources and capabililities for developing a program atthe organisational level. Limitations in resources, policies, time and abilities areidentified, as are opportunities, and the methods and strategies for intervention arechosen. The intervention is then implemented (phase six). Process, impact andoutcome evaluations (phases 7 to 9) should be a continuous process throughout themodel, and evaluation criteria should be determined from the objectives identified apriori, during the diagnostic phases.This model originated as a framework for health education cost-benefitevaluations (Green, 1974) and evolved into a model for program evaluation which hasbeen widely applied in a variety of settings (Green, Kreuter, Deeds and Partridge,1980).2.2 HEALTH EDUCATION FOR OLDER ADULTSTraditionally, health education has been targeted at younger populations, withthe result that older people have often been overlooked (Barr, 1983; Cox and Monk,1989; Estes, Fox and Mahoney, 1986; Gilbert, 1986; Minkler and Pasick, 1986).Health education was viewed as an activity that could be presented as a one-timeeffort early in life to ensure healthy choices throughout the lifespan (MacRae, 1986).Attitudes were also a barrier; older adults were generally believed to be too old toparticipate in vigorous activity, for example (MacRae, 1986).Factors credited with influencing the development of health promotion andeducation for older adults include growth in the number of older people (Cox andMonk, 1989; Lalonde and Fallcreek, 1985; Weiler and Lubben, 1989) and increasingper capita proportion of older people (Gilbert, 1986; Rimer, Keintz, Glassman andKinmar, 1986). Projections by Statistics Canada indicate that by the year 2020 onefifth of the Canadian population will be over 65 years (Health and Welfare Canada,1989). Other factors include increasing costs of medical care (Barr, 1983; Butler,Gertman, Oberlander, Schindler, 1979; Cox and Monk, 1989; Lalonde and Fallcreek,1985; McLeroy et. al., 1988; Reynolds, 1975; Weiler, Chi and Lubben, 1989),increased expectations and educational attainment of elderly health care consumers(Butler et. al., 1979) and lack of coordination of health care services and research(Slivinske and Kosberg, 1984). An additional catalyst contributing to the developmentof health promotion/education programs for older persons is the increasing realisation,in an era of cost containment, that the elderly are high users of treatment services, asthey suffer from more chronic conditions than younger people (Butler et. al., 1979;Health and Welfare Canada, 1989). For instance, Gilbert (1986) quotes high U.S.percentages of older people suffering from arthritis (44%), hypertension (30%),hearing loss (29%) and heart conditions (27%). On average, older Americans areconfined to bed for 14 days of each year and must restrict their activity for 39 days of-15-each year (in Gilbert, 1986). The Canadian experience is that 80% of persons over 65years report at least one health problem, compared to 54% of Canadians as a whole.Twenty-four percent of persons aged 55 to 64 years and 32% of persons over 65 yearsreport that they experience activity limitations, compared to 10% of persons under 55years (Health and Welfare Canada, 1989). The proportionately greater rates ofchronic illness and disability in older populations have resulted in large proportions ofhealth care budgets devoted to treatment of elderly persons, providing additionalmotivation for health education among the elderly (Lalonde and Fallcreek, 1985).Health promotion activities are often viewed by health care professionals as a methodto prevent institutionalisation of elderly people, resulting in decreased costs of careand more humane treatment (Moneyham and Rider, 1988; Schneider, Chapman andVoth, 1985).The provision of health promotion programs for the elderly has been endorsedby society, government and by professional organisations (Krout, 1983). For example,the British Columbia Minister of Health and Minister Responsible for Seniors, inToward a Better Age (1989), recommended greater emphasis on preventive activities,stating that many of the major causes of ill health and disability in old age may bepreventable through healthy lifestyle. The Institute for Health Care Facilities of theFuture, in the 1988 Aging Report,_ also endorsed health promotion and diseaseprevention as ways to improve the quality of life of Canadian elders, and stressed theneed to move away from a medical model of health care toward a focus on socialsupport, self-care and creation of healthy environments. The Canadian NursesAssociation (1989) emphasised that encouraging wellness practices should become amajor concern in health care, to maintain and enhance seniors' quality of life. Mostrecently, the British Columbia Royal Commission on Health Care and Costs (1991)stated that the traditional health care system, where ninety-seven percent of thebudget goes toward institutional care, medical services and pharmacare, must be-16-transformed into a system where more money is spent on prevention of illness andprotection of health.Cox and Monk (1989) described the purpose of health education for olderadults as illness prevention. However, others portrayed it as including a range ofactivities, from primary prevention of disease to self-care for chronic disease andmaintenance of an individual's ability to function in their natural environment (Benson,Nelson, Napps, Roberts, Kane-Williams and Salisbury, 1989; Rimer et. al., 1986).Rimer et. al. (1986) listed the goals of programs as increased health, increased qualityof life and decreases in days of restricted activity. They affirmed that the intendedoutcome is an extension in years of active, independent pleasurable life, rather thatmerely an increase in years of life. Secondary analysis of research data by Fries,Green and Levine (1989) supported this; termination of bad health habits can reversehealth risks in relatively short periods of time, and development of good habits canquickly provide beneficial effects. Accessibility of medical care was also seen as agoal by some (Reynolds, 1975), as was prevention of institutionalisation (Schneider et.al., 1985; U.S. Department of Health, Education and Welfare, 1979). Health educationand health promotion may also be viewed as a vehicle for achieving community-basedempowerment to achieve social and/or environmental change (Caplan, 1990; French,1990; Taylor, 1990).2.3 COMMUNITY-BASED HEALTH PROMOTION/EDUCATION PROGRAMSFOR THE ELDERLYCommunity-based programs for the well elderly provide healthpromotion/maintenance activities and early detection of disease (Moneyham andRider, 1988). Health education programs are used as means to prevent illness inolder people, assist them to achieve the ability to manage their chronic diseaseindependently and provide them with information to become more effective consumers-17-of health services (Rimer, Keintz, Glassman and Kinman, 1986). They occur in a widerange of settings, including senior citizens' centers, meal sites, recreational facilities,subsidised housing units, older adult residences (Moneyham and Rider, 1988),churches and other religious organisations, health care delivery sites such ashospitals, colleges and universities and voluntary organisations (Rimer et. al., 1986).Staffing for older adult health education programs can consist of health educators,nurses, occupational therapists, social workers, students and older persons as peereducators (Rimer et. al., 1986). Programs typically offer health assessment,counselling, education, monitoring of chronic conditions, screening and referral(Moneyham and Rider, 1988). Often the focus is on instruction for management ofcommon chronic diseases, such as hypertension, heart disease, diabetes, arthritis andhearing and vision loss (Meeks and Johnson, 1988). In addition, medication use,nutrition, exercise, accident prevention and depression are common topics for healthpromotion among the elderly (Meeks and Johnson, 1988).Several examples of large community-based senior health promotion/educationprograms can be found in the literature. Some programs focused on a single issue foreducation and behaviour change. For example, the SRx-Senior Medication EducationProgram in San Francisco, begun in the late 1970's, provided education forprescription drug use to seniors through theatre presentations, mini-classes,multilingual educational materials and individual consultations (Eng and Emlet, 1990).A project in Buffalo, New York provided dietary education and instruction to a randomsample of elderly persons who were recipients of noon-time meals at the SalvationArmy Center and judged to have inadequate nutritional intake (Mitic, 1985). ThePreventive Health Care for the Aging Program (PHCAP) in California furnishedmedical screening and referral by a public health nurse to self-referred elderly personsat 24 local screening sites (Weiler, Chi and Lubben, 1989).Other programs were more comprehensive, including a wide range of health-18-promotion topics and activities. The Senior Actualisation and Growth Explorationprogram (SAGE), for example, was founded in 1974 in California, and spread to otherU.S. states (Butler et. al., 1979). It was based on the premise that the older years of lifeshould continue to involve personal growth, and consisted of two to three hours perweek of deep breathing, massage, exercise, biofeedback and counselling (Butler et.al, 1979). In June, 1975, Sehnert began a Medicare program out of Georgetown'sSchool for Summer and Continuing Education in Virginia consisting of healtheducation lectures and self-help skills instruction (Butler et. al., 1979).The Self-Care for Senior Citizens Program (SCSC) was developed andsponsored by the Dartmouth Institute for Better Health (DIBH) in New Hampshire in thelate 1970's (Simmons, Roberts and Nelson, 1986) with the aim of helping the elderlylearn how to achieve better physical and mental health, use the health and socialservice system more effectively and maintain their independent living status. Theprogram consisted of thirteen two hour sessions with a set curriculum developed byDIBH. This project led to the Staying Healthy After Fifty (SHAF) program, acollaborative venture of DIBH, the American Association of Retired Persons (AARP)and the American Red Cross in 1985 (Simmons, Nelson, Roberts, Salisbury, Kane-Williams, Benson, 1989). SHAF was a revision of SCSC, and consisted of eleven two-hour sessions which cover health concerns and emergency situations, lifestyle andconsumer planning.The Wisdom Project was another collaborative arrangement between theAmerican Red Cross and municipal organisations in New York to provide screeningand treatment, health education and evaluation/follow-up through thirty senior centersfor persons over 65 years with health disorders (Lederman and Farrar, 1986). TheGROWING YOUNGER program was begun in 1979 through Healthwise, a non-profithealth promotion centre which had previously only worked with young families(Kemper, 1986). It was developed and piloted at the Boise Senior Center, and aimed-19-to improve the health of older people living in the community. Its program wascomprised of four two-hour workshops covering fitness, nutrition, stress managementand medical self-care. The Tenderloin Senior Outreach Project (TSOP) in SanFrancisco was a university sponsored project, developed in 1979 with the aim ofcombating social isolation, poor health and powerlessness, which were common tothe elderly low income residents of the Tenderloin area. It provided health educationabout smoking, nutrition, exercise and alcohol/drug abuse, and screening and referralon a weekly basis in each of eight Tenderloin hotels, but it also emphasised socialgathering and mobilisation of the elderly to take collective actions in relation tolandlords, merchants and others to improve their living conditions. The WallingfordWellness Project was a research project in health promotion for persons over 54 yearsof age, which consisted of a 21 week program of physical fitness, stress management,nutrition and environmental awareness and action (Lalonde and Fallcreek, 1985).Unlike other senior health promotion programs, participants ranged in age from 13 to87 years, although the research centred around those who were over 54 years of age.Meeks and Johnson (1988) described the development of a comprehensivehealth promotion program based upon a literature review and community needsassessment at the North Shore Senior Center in Winnetka, Illinois. This resulted in ahealth promotion "browsing area", with a library of pamphlets and books, publication ofa cookbook of nutritious recipes, health topics inserted into the member newsletter, amonthly health lecture series and ongoing exercise programs. The "Feeling Great!"Wellness Program for Older Adults was developed at a Louisiana YMCA as a fitnessprogram for older adults, and included a complete physical workout, social outlet,health education lecture/discussion and recreational activity in each session for adultsover fifty years of age (Weiss, 1988). The Personal Health Management System(PHMS) was developed in the Ohio Presbyterian Homes in 1979 with the goals ofimproving quality of life for participants and decreasing costs of health care through-20-application of wellness concepts (Slivinske and Kosberg, 1984). The programincorporated educational classes, physical exercise and individual professionalconferences.In Vancouver, Seniors Wellness positions were established within each healthunit in 1984 (Martin, Robertson, Altman, 1988). This led to a process of establishinghealth promotion activities for the elderly through existing community agencies orgroups. Once the sponsorship of a group of seniors in a neighbourhood wasestablished, a health professional worked with the sponsoring group to design andimplement a program, which generally consisted of weekly meetings for two hoursoraganised around a format of exercise, social interaction and health discussion(Martin, 1986). In 1990, there were twenty-three such programs in operation inVancouver (Calsaferri, 1990).2.4 EFFECTIVENESS OF SENIOR HEALTH PROMOTION PROGRAMSMuch of the literature describing senior health promotion/education programswas descriptive in nature, such as the "Feeling Great" program (Weiss, 1988).Process evaluative information was provided for others, such as the PHCAP, whichprovided quantitative data on numbers of participants, what types of services wereprovided, and resulting referrals (Weiler et. al., 1989). Likewise, SRx suppliednumbers of persons reached by the program, but outcome effectiveness of theprogram in terms of behavioural change and health measures were not reported (Engand Emlet, 1990). Qualitative information regarding effectiveness was provided forSAGE (Butler et. al., 1979), the Wisdom project (Lederman and Farrar, 1986) andTenderloin (Wechsler and Minkler, 1986) in terms of participants' satisfaction with, andperceived benefits of, the programs.Mitic (1985) utilised a random control group design to evaluate theeffectiveness of nutrition education, and found that , using a 24 hour dietary recall, 62percent of the experimental group (n=34) reported eating adequately immediately-21-following intervention, compared to only nine percent of the control group (n=32). Sixweeks after completion of the program, 73 percent of the experimental group and ninepercent of the control group exhibited adequate eating behaviours.Slivinske and Kosberg (1984) used a modified non-equivalent control groupdesign (the experimental group was self-selected, whereas the controls were recruitedthrough random selection from one of the Ohio Presbyterian Home facilities). Self-administered questionnaires disclosed that on scales of physical health, morale,economic resources, Activities of Daily Living (ADL), spirituality, social resources andan overall Wellness index, scores for experimental subjects (n=53) increased overtime whereas scores for controls (n=14) remained relatively stable. Muscular strengthand flexibility increased significantly for the experimental group (p<.001), but therewas no significant change in number of times medical care was sought, number ofdays of illness, amount of money spent on medical care or number of days ofhospitalisation.The Wallingford Wellness Project (Lalonde and Fallcreek, 1985) recruited acomparison group (n=44) from church groups, social groups and housing complexesin the same geographic area from which the experimental group (n=90) was recruited.The controls were matched for age, sex, socioeconomic status (SES), marital statusand physical/psychosocial health levels. Measures of health knowledge, attitude andbehaviour change were taken at baseline, immediately following the 21 weekprogram, six months after completion and two years following completion. There wereno significant pre-test differences between groups. At immediate post-test, theexperimental group had significant improvement in overall mental health and positivewell-being, motivation and confidence in their ability to initiate and sustain lifestylechanges in the areas of nutrition and stress management, behavioural changes inphysical fitness, stress management, nutrition and environmental awareness, andhealth information related to environmental awareness and stress management. A-22-significant reduction was found in risk of heart attack and stroke. The majority ofsignificant post-test changes were sustained at the six month followup (Lalonde andFallcreek, 1985). There was also a significant decrease in depression and increase inacceptance of responsibility for health, knowledge of available social and healthresources and nutrition related health information at the six month post-test. At the twoyear follow-up, behavioural changes among the experimental group in physicalfitness, stress management and nutrition were sustained, but not at the same level asthe six month followup (Lalonde, Hooyman and Blumhagen, 1988). Healthinformation was retained without significant decline for all areas except nutrition, andthere was still significantly greater knowledge in all four areas compared to thecontrols. The mental health benefits, increased health responsibility, and decreasedrisk of heart attack and stroke were, however, not sustained at the long term followup.The authors concluded that the project was most effective in the short term andsuggested that continued intermittent program intervention might be required tosustain benefits over the long term (Lalonde, Hooyman and Blumhagen, 1988).The effectiveness of SHAF was examined using a pre and post evaluation studyusing friends or neighbours of participants (n=164) as the controls for the experimentalgroup (n=161). Data were obtained from twenty SHAF courses held in sixteencommunities over a five month period in 1986. Course completion rate was high(92.5%), as was satisfaction with the course (97%) (Simmons et. al., 1989). The testgroup, at baseline, perceived themselves as significantly less healthy and more limitedby health problems than the control group; they were also less likely to be white andless educated than the controls (Benson, Nelson, Napps, Roberts, Kane-Williams andSalisbury, 1989). At immediate post-intervention, the test group scored higher thanthe controls on self-assessed ability to perform health skills such as takingtemperature, blood pressure and pulse, performing first aid for choking, understandingfood labels, using a medical reference book, checking for safety hazards at home and-23-owning a medical reference book. Test group members also reported significantimprovements in health actions such as use of a seatbelt, performing vigorous andstretching/strengthening exercise, reducing stress levels, consuming calcium richfoods. The test group scored significantly higher on a health care costs scale,displaying significantly greater ability in trying to save money on health care,comparing prices and services and comparing health insurance policies. At the sixmonth post-test, differences in the health skills scale were sustained and the test groupcontinued to report improvement on individual health skills. The differences on thehealth actions scales were also sustained, but there was no longer a significantdifference in the health care costs scale. The positive results that were obtained mustbe viewed with caution, as there is a potential for regression to the mean due to lowerbase-line scores in the test group.Cox and Monk (1989) evaluated the effectiveness of comprehensive healtheducation services at six senior centers in the New York area. The participant sampleconsisted of 104 individuals, and a comparison group, comprised of senior centreattenders who did not participate in the health education program, was used (n=36).The comparison group was less likely to be foreign born, had higher levels ofeducation and had higher annual incomes than the test group. There were nosignificant differences in types of chronic conditions or in use of health services. Thecomparison group was more likely to rate their health as good or excellent (51% ofparticipants compared to 84% of non-participants). The comparison group also feltsignificantly more in control of their health than did participants, but the authorspointed out that this could have been a reflection of differences in education and SES.More participants than nonparticipants felt they were getting an appropriate amount ofexercise, however, and more participants than nonparticipants had made changes inhealth behaviour over the previous year, including dietary changes (p<.006), exercisehabits (p<.001) and weight changes (p<.05). These changes were attributed by-24-participants to the health education program. Because this was a cross-sectionalstudy, a causal relationship between the program and the differences observed cannotbe inferred.It is, perhaps, surprising that so little formal evaluation has been attempted,given the popularity of community-based programs for the elderly. Nevertheless,based upon the research results available in the literature reviewed, it seems that suchprograms are effective in producing at least short-term benefits in health knowledge,attitudes and behaviours. There is evidence that at least some of these changes aresustained up to two years following completion of the program.2.5 TARGET POPULATION FOR SENIOR HEALTH PROMOTIONPROGRAMSTo be successful, a health promotion program must be designed to meet theneeds of its target population (Fallcreek, Warner-Reitz and Mettler, 1986; Meeks andJohnson, 1988; Trela and Simmons, 1971). Choice of target populations should begrounded on knowledge of the prevalence of disease and the potential for healthpromotion and disease control (Green and Kreuter, 1991; Larson, 1988). Identificationof risk factors for development of disease or disablility, through epidemiology, canallow programs to be targeted to high risk individuals (Breslow, 1983). Recruitmenttactics, promotional materials, curriculum and evaluation are all dependent uponcharacteristics of the target population (Fallcreek et. al., 1986). The local environmentmust also be considered in determining recruitment techniques (Simmons et. al.,1986).Individuals over 70 years are known to be more at risk for malnutrition and thisis exacerbated by socioeconomic factors, physical health, mental status, and socialisolation (Anderson, 1982). Diet is an important strategy for prevention ofcardiovascular disease, which is a major cause of death and disability in people over-25-65 years (Larson, 1988). In addition, and more important to an elderly targetpopulation, diet and exercise have been found to reverse the effects of atheroscerosis(Fries, et. al., 1989). However, the majority of elderly people do not link diet withprevention of illness and disability (Health and Welfare Canada, 1989; Larson, 1988).Elderly persons exercise less, yet require exercise to maintain joint flexibility,muscle strength, balance, tendon strength and cardiovascular reserves (Anderson,1982). Elderly persons who do not exercise regularly have been found to besignificantly more at risk for decline in functional status, after demographic andmedical conditions have been controlled for (Mor, Murphy, Masterson-Allen, Willey,Razmpour, Jackson, Greer and Katz, 1989). Lack of exercise is also linked toexacerbation of cardiovascular disease (Larson, 1988). Regular exercise is apotential deterrent of osteoporosis and low bone mass, which are leading causes offractures, especially in elderly women (Larson, 1988). Yet greater than one third ofelderly Canadians believe that more exercise would not improve their health (Healthand Welfare Canada, 1989).In the area of avoidance of health risks, although the frequency of smoking anddrinking alcohol decreases with age, the use of sleeping pills and tranquilisersincreases dramatically (Health and Welfare Canada, 1989). More elderly people useseatbelts and less elderly persons report drinking and driving (Health and WelfareCanada, 1989).Depression, which increases in incidence with age (Butler, 1975), is correlatedwith low income and less social activity (Anderson, 1982; Health and Welfare Canada,1989). Suicide also increases with age, and has been related to physical stressessuch as illness or loss of finances (Butler, 1975). Physical stresses such as illnessmay induce mental confusion, depression, weakness or sleep disturbances in elderlypersons. A large percentage of elderly people do not receive help for psychiatricproblems (Butler, 1975).-26-The extent of one's connections to family, close friends, relatives, social groupsaffects health status (Breslow, 1983). Frequency of interaction with friends andrelatives for males, and number of organisational memberships for females, havebeen correlated with positive health status (Green and Gottlieb, 1989). A progressiveloss of social roles through retirement, decreased physical mobility, loss of parentingresponsibilities and death of friends/spouse occur as a person ages, and these rolelosses have the potential to cause isolation of the older person (Toseland, Decker andBliesner, 1979). Social isolation is recognised by many health care professionals as acause of some elderly persons who require community assistance "falling through thecracks" (Bennett and Killeffer, 1989).Schneider, Chapman and Voth (1985) suggested that community-basedprograms for the elderly should direct services at priority groups of older persons suchas those who are advanced in age, of low SES, who live alone and have functionaldisabilities which impede ADL. They further stated that, if a goal of the program is todecrease institutionalisation, targeted individuals should be those who are older than75 years, live alone, have recently been hospitalised and have a chronic healthproblem. Support for this is found in Canada's Health Promotion Survey completed in1985, which determined that seniors perceived themselves as less healthy, reportedmore chronic disability and more activity restrictions due to disability when they wereolder, less wealthy and less educated (Health and Welfare Canada, 1989; Health andWelfare Canada, 1987). Green and Gottlieb (1989) also found education and incometo be positively correlated, and life events and age to be negatively associated, withhealth status. They suggested that interventions which promote healthy lifestyleshould target specific groups defined by income, education and age. Krout (1983)identified the "at risk elders" who require formal government support as those whosuffer chronic illness, live in substandard housing, are subject to a large reduction inpersonal income and lack access to personal or public transportation. Low SES and-27-levels of community involvement have also been correlated with low life satisfaction,which, in turn, is related to level of well being (Kearney, Plax, Lentz, 1985). The 1973amendments to the U.S. Older Americans Act recognised the need to target the mostvulnerable groups in planning and service delivery, including the low income, minoritygroup members and socially or geographically isolated elders who were in need ofservices to maintain functional abilities and lessen deterioration (Jacks, 1975).Yet the poor and the socially isolated, who may be at high risk for healthproblems, are often the most difficult to reach with voluntary health educationprograms (McLeroy et. al., 1988). Community-wide approaches often do not reacholder age groups, rendering them ineffective with this age group (Green and Gottlieb,1989). People are often not aware of existing community services, are afraid to askfor help, or deny that they are not coping (Stuen, 1985).Recently, professionals have become aware that making services available atsenior centers and other locations does not ensure that the targeted population willutilise them (Krout, 1983). With no outreach component, an agency places the burdenfor initiating service requests on the older person, who may not know about the serviceor understand how to access it (Stuen, 1985). In a review of the literature, Krout(1983) determined that knowledge of services by the elderly was variable from surveyto survey, many elderly people do not view the services they are aware of in a positivelight, and there are very low utilisation rates by a small minority of elderly people.2.6 CHARACTERISTICS OF PARTICIPANTS OF HEALTH PROMOTIONPROGRAMSPrograms which have recorded baseline data for their participants provide agood indication of what kinds of people are being reached by present programming.Several documented programs for seniors have provided their baseline data in theliterature, which can then be contrasted with the characteristics which are perceived tobelong to the target group, as outlined in the previous section.-28-The Wallingford Wellness Project employed media such as newspaper, radioand group presentations in recruitment (Lalonde and Fallcreek, 1985). Participants(n=90) ranged in age from 55 to 87 years, with a mean age of 70. Twenty-one percentof participants were male and over half (n=38) were married. All participants wereCaucasian. The average yearly income was $11,866. No demographic informationfor seniors in the recruitment area was provided for comparison purposes. Thecomparison group, although recruited from the same geographical region, does notallow for comparison to determine whether the program reached the more "at risk"group due to the fact that recruiters tried to match on demographic characteristics forpurposes of outcome evaluation of the project. However, the control group (n=44) wasolder (range of 61 to 98 years with mean age of 73), less likely to be married (34%were married) and less affluent (mean yearly income of $9792). None of thesedifferences was significant. Similar to the participant group, 22% were male and allwere Caucasian.SHAF (Benson et. al., 1989) showed an opposite pattern between participants(n=161) and the friendship/control group (n=164). Once again, the majority of bothgroups was female (83% for participants and 88% for friendship group). Both groupswere younger (mean age of 65.9 years for participants and 64.2 years for controls). Alarge percentage of both groups was also married (46% for participants and 53% forcontrols). The participants were significantly less educated, with 7.0 years versus 7.4years of education (p=.05) and less likely to be white (83% versus 93%, p=.01), andthough no income figures are provided, this would indicate that participants were likelyless affluent than controls in the SHAF program. In addition, the test group feltsignificantly less healthy and more limited by health problems than the friendshipgroup, and considered themselves less capable in performing health skills. Theparticipant group also reported less social interaction than controls, and more of thetest group were dissatisfied with their lives, although these results were not significant.-29-None of these demographic or health variables were compared to the characteristicsof seniors in the areas from which individuals were recruited.In contrast to both SHAF and the Wallingford Wellness Project, Cox and Monk(1989), who drew their comparison group from the same senior centers that housedthe health education programs they were evaluating, found that only one quarter ofparticipants (n=104) and controls (n=36) were married, with approximately 70% ofboth groups living alone. Participants were significantly more likely to be foreign born(28% versus 7%). There was a dramatic difference between groups in annual income,with participants significantly less affluent that controls (p<.0001). Likewise,participants had significantly less years of education (p=.002). Although this was across-sectional study so the measures of health status are not baseline data, it isinteresting that there were no significant differences between groups in numbers ortypes of chronic conditions reported but, as with SHAF, the participant groupconsidered themselves to be significantly less healthy and less in control of theirhealth than the comparison group. As with preceeding studies, no data were providedto allow comparison with the area demographics.Slivinske and Kosberg (1984) did not differentiate the demographiccharacteristics between participants (n=53) and controls (n=14). Statistics for the twogroups suggest that they were very well educated (mean of 14 years of formaleducation) and financially able to cope (mean annual income of $13,500, with threequarters having retired from, or with a spouse who was retired from, a professional orwhite collar job). The average age was 76 years, 23 % of study participants weremarried and the majority were Caucasian female. It is also noteworthy that 43 of the53 participants were residents of Ohio Presbyterian Homes, which might be indicativeof a social support network.Fitch and Slivinske (1988) randomly selected 84 residents from three retirementcommunities and then randomly assigned them to participant and control groups. The-30-average age was 77.5 years. Once again, this was a relatively more affluent andeducated group, with a mean annual income of $16,485 and average years of formaleducation of 14.8. 73% of the samples were female and 39% were married.Weiler et. al. (1989), in a sample of 5,454 participants of the Preventive HealthCare Program for the Aged, found that two thirds were female, nearly 85% were white,and close to half were married. 41% of the sample lived alone. 38% were 60-69years of age, 49% were 70-79 years and 14% 80 years old or more. Forty-four percentof the sample had no chronic conditions, 23% had one, 17% had two and 16% hadthree or more chronic conditions. Using medical insurance coverage as a proxy forfinancial status, the researchers conjectured that those who were eligible for Medicaid(13%) and those who had no insurance (5%) were probably low income. In contrast,approximately 60% had private medical insurance in addition to Medicare.In a study of five of the Vancouver Health Department Seniors' WellnessPrograms, it was found that the participant ethnic mix generally mirrored the ethnic mixof the surrounding neighborhood, but males were underrepresented, with a range of 0to 33% and a mean of 27% of participants being male (Calsaferri, 1990).Schneider et. al. (1985) chose, as one of their main research objectives, todetermine whether the senior program was reaching the intended audience of those atrisk for institutionalisation. They selected a quota sample through canvassingenumeration districts based upon 1970 census data. Data from this sample (n=152)could then be compared to information obtained from seniors who took part in amobile medical screening program (n=251). The quota sample, which should haveaccurately reflected the characteristics of seniors in that geographical region, wereolder (73 years versus 71 years), less well educated (8.4 years versus 9.3 years), lessaffluent ($375 per month versus $380 per month) and less healthy (3.7 versus 1.6annual hospitalisation days, 17.3 versus 2.8 annual disability days, 6.3 versus 4.1annual doctor visits) than the medical screening sample. A smaller proportion of the-31-quota sample attended church weekly or were involved with the senior centre. Theywere also less likely to expect an interesting future (68.4% versus 87.5%). Themedical screening sample was, however, more likely to live alone and had lesscontact with their children than the quota sample.The research reviewed suggests that there is wide variation between programswith regard to characteristics of recruited individuals. However, it seems that programsare only partially successful in reaching targeted "at risk" seniors. Schneider et. al.(1985), which is the most discouraging, suggesting that participants of voluntarymedical screening are more healthy, more affluent, more educated, younger and moreinvolved in the community than the norm for the geographical region, is likely the mostaccurate, as no other studies employed a random sample of seniors in thegeographical area for their baseline. The studies which seem to have reached a more"at risk" group of participants utilised matching techniques in order to have anappropriate comparison group for determining effectiveness of the program. Thematching techniques are ineffective in determining whether or not the appropriatetarget population was reached.2.7 SENIOR CENTERS AS LOCATIONS FOR HEALTHEDUCATION/PROMOTIONA common location for community based seniors health education programs isthe local senior citizens' centre (Leanse, 1986, Rimer et. al., 1986). The definition of aSenior Center, as developed by the U.S. National Council on the Aging's NationalInstitute of Senior Centers in their Senior Center Standards (1975) is as follows:A senior centre is a community focal point on aging where older persons as individuals orin groups come together for services and activities which enhance their dignity, supporttheir independence and encourage their involvement in and with the community.As part of a comprehensive community strategy to meet the needs of older persons,senior centre programs take place within and emanate from a facility. These programsconsist of a variety of services and activities in areas such as education, creative arts,recreation, advocacy, leadership development, employment, health, nutrition, social workand other supportive services.-32-The centre also serves as a community resource for information on aging, for trainingprofessional and lay leadership and for developing new approaches to aging programs.(Quoted in Leanse, 1986, p. 105-6)Senior centers operate with the philosophy that aging is a normal process,individuals require interaction with and support from peers, and adults have the right toinvolvement in issues which concern them (Leanse, 1986). Historically, senior centershave provided primarily social and recreational activities (Hanssen, Meima, Buckspan,Henderson, Helbig and Zarit, 1978). The philosophy of senior centers, however, issupportive of health promotion/education philosophies of individual responsibility anddetermination for health behaviours (Leanse, 1986), and this has led to increases inhealth programming in recent years (Hanssen et. al., 1978).A further reason for development of health promotion programs in SeniorCenters is the fact that large numbers of seniors attend them (Cox and Monk, 1989).The assumption is made that senior centers are a natural setting to reach the targetpopulation of older adults (Cox and Monk, 1989; Meeks and Johnson, 1988;Schneider et. al., 1985). However, many senior centers do not have the resources forcareful development of health programming (Meeks and Johnson, 1988).Furthermore, if senior centers are to be effective in providing health services, they mustbe able to attract a population that needs such services and can be assisted by thetechniques utilised by the centre (Schneider et. al., 1985). With a shift fromrecreational to multi-purpose programming, there must be an accompanying movetoward outreach or programming which attracts older people who did not previouslyattend, and there is some question whether this change has happened (Hanssen et.al., 1978).2.8 CHARACTERISTICS OF SENIOR CENTER ATTENDERSA great deal of research has been undertaken to determine the characteristicsof participants/non-participants of senior centers, particularly in the United States.-33-Trela and Simmons (1971), in a sample of 210 new senior centre members and 110non-members of a senior centre, determined that reasons for joining included desirefor outside interests besides work and family (43%), the need for companionship(12%) and the persuasiveness of the volunteer recruiter (greater than one third).Reasons given for not joining included competing activities and interests, ambivalencetoward organisational activities, poor health or lack of transportation. Non-memberswere more likely to state they had a health problem and less likely to perceive theirhealth as excellent or good (significant at .01 level). Poor health also played a largerole in membership attrition, with the mortality rate for those who were no longermembers after two years at 15%, and the percentage of those who gave poor healthas a reason for lapsed membership at 16%.Toseland and Sykes (1977) found that measures of life satisfaction were notcorrelated with attendance or nonattendance at a senior centre. Hanssen, Meima,Buckspan, Henderson, Helbig and Zarit (1978) found that senior centre participantswere more "socially oriented", enjoying more structured, community activities, hadfewer physical limitations and reported less depression than nonparticipants. Incontrast to Trela and Simmons (1971), they found no correlation between lack oftransportation and non-participation, and unlike Toseland and Sykes (1977), therewere no significant differences in measures of health. Krout (1983b) also found thatusers (n=97) reported less mobility problems due to health than non-users (n=125).He also determined that attenders were more likely to see their friends often, hadlower incomes, had an average of 1.5 years less education and were more likely to befemale, not married and live alone. Using multiple regression, this study found that20% of the variation between users and non-users was accounted for by lowerincome, lower education, seeing friends more frequently and desiring more contactwith children. Reasons given for attendance were "something to do" (50%) and"invitation from friends or others" (25%). Reasons for lack of involvement were being-34-"too busy" (40%) and "lack of interest" (20%), followed by "no need" (8%). Only 2%identified lack of transportation as a reason for nonattendance. Ralston and Griggs(1985) found that lack of time, home responsibilities, lack of interest in attending asenior centre, job responsibilities and lack of interesting activities were the mostcommonly listed obstacles to participation in a survey of 110 senior homeowners.Schneider et. al. (1985) found that more women and more individuals who attendedchurch regularly were senior centre participants (n=500). There was no significantdifference in age or health status. An interesting finding was that participants enterednursing homes in greater proportions than non-participants.In looking at frequency, duration and stability of senior centre attendance, Krout(1988) found that female, unmarried and low income attenders and those who sawtheir friends daily reported higher frequency of use than male, married or high incomeattenders. Length of attendance was only significantly correlated to increased age.Unmarried persons were more likely to report variable frequencies of attendance,while renters and those reporting excellent health status were more likely to haveincreased their frequency of attendance over time.A secondary analysis by Krout, Cutler and Coward (1990) of data drawn fromthe 1984 Supplement of Aging (SOA) to the National Health Interview Study (NHIS),an annual survey of the civilian, noninstitutionalised U.S. population, used informationfrom 13,807 Americans 60 years of age and older. Of this sample, 13.7% had used asenior centre in the preceding twelve months. Use of multivariate analysis andregression techniques determined that social interaction was the variable which moststrongly predicted senior centre participation. Age and education had curvilinearrelationships, with middle ages and middle levels of educational attainment correlatedmore highly with participation. Participants were more likely to be female, have lowerfamily incomes, to live alone and to have less difficulty with Activities of Daily Living.Non-central city area residents and rural non-farm residents were more likely to be-35-participants. There were no questions on the SOA to the NHIS that allowed theresearchers to measure differences in life satisfaction, morale or access seniorcenters.An interesting finding by Cox and Monk (1990) was that senior centers do notintegrate large proportions of frail elderly individuals into centre activities (10% ofparticipants were described by directors as frail, in a survey of 282 centers), and 48%of the directors felt that 10% was the upper limit of frail individuals that could beabsorbed into their programs. Only 27% of directors felt that they could integrate more,with 20% being the upper limit of frail seniors which was perceived possible. Over onethird of the disabled participants were considered frail due to hearing loss, followed byvisual, memory and mobility impairments. The authors found evidence of littleplanning for such integration, although over half had a policy of acceptance of frailelderly and three quarters encouraged the well members to welcome them. Trainingfor staff and interagency cooperation to provide specialised services were notprevalent in the study.2.9 SUMMARYThe literature seems to show that community-based health education programsfor seniors are effective in increasing various measures of health. For example,changes have been noted in dietary habits (Mitic, 1985), improved mental health andpositive well-being have been demonstrated (Lalonde and Fallcreek, 1985),behavioral changes resulting in reduced risk of illness have been found (Lalonde andFallcreek, 1985) and increased knowlege of health behavior performance andeffective health service utilisation have been demonstrated (Benson et. al. 1989).Many of these changes are sustained at short term follow up, but tend not to besustained over the long term, suggesting the need for some continued intermittentprogram intervention (Lalonde, Hooyman and Blumhagen, 1988).The people who are identified as part of the target population, however, are-36-often not aware of existing community services (Krout, 1983) and are afraid to ask forhelp, or deny that they are not coping (Stuen, 1985). Senior centres seem to reach agroup of elders who are healthier, with less ADL limitations and more social interaction(Hanssen et. al., 1978; Krout, 1983b; Schneider et. al., 1985). Two of the reportedreasons for not attending senior centres, "lack of transportation" and "poor health",could be classified as barriers to attendance, or enabling reasons on the PRECEDE-PROCEED framework (Green and Kreuter, 1991).Little research has been carried out to determine whether this pattern is alsotrue for community-based health education programs. The studies which haveemployed a control group have generally utilised matching techniques in order todetermine the effectiveness of the health education intervention (Benson et. al. 1989;Lalonde and Fallcreek, 1985). Determining whether or not their target group wasbeing reached was generally not a research objective, so the matched controls werenot chosen to be representative of the population of seniors in the community.Schneider et. al. (1985), however, found that participants of a mobile screeningprogram were younger, healthier and more educated than a representative controlgroup. The differences in characteristics between participants and nonparticipants, aswell as the reasons for attending or not attending a community-based seniors healtheducation program have yet to documented.Based upon the literature review, a simplified conceptual scheme can bedeveloped to show the expected relationship between various characteristics ofattenders of a senior health education program, their environment, and participation.The conceptual scheme is presented in Figure 1, and is adapted from the Green andKreuter (1991) PRECEDE-PROCEED model. The framework, as presented by Greenand Kreuter (1991) already includes all elements presented in Figure 1. The elementsare simply realigned in the scheme presented here, for purposes of the present study.The addition of bi-directional arrows indicates that this cross-sectional study does not-37-1\1 EnablingBehaviorandLifestyle• health behaviors• communityinvolvement• etc.A,Demographics•age•gender• educationHealth.4-Health-Related^.44,,• mobility• economic status• social network• Physical• Mental11%PredisposingParticipation (orNonparticipation)in a HealthEducationProgram-4111-10-Reinforcing, Environment•allow for causational analysis.Participation in a health education program is expected to be influenced bypredisposing, enabling and reinforcing factors. The types of motivational factorspossessed by an individual will be determined by their behavior and lifestyle(including health behaviours, degree of community involvement, extent to whichlifestyle revolves around home-based activities, cultural factors, etc.).Individual behaviour and lifestyle is determined by intervening variables suchas demographic characteristics of the individual (for example, gender, age,educational status) and the environment (e.g., accessibility of the health program,availability of transportation, safety issues, etc.). In addition, there is a group of health-related variables which are influenced by health status and, in turn, affect behaviourand lifestyle. These variables are often considered aspects of health in a multi-dimensional model, and include factors such as mobility, social network and economicstatus of the individual.The three sets of intervening variables are influenced by, and influence, thephysical and mental health of the individual. They also may act as enabling variables,which provide motivation or barriers to participation in the health education program.FIGURE1: CONCEPTUAL SCHEME SHOWING THE RELATIONSHIPS BETWEENPARTICIPATION IN A HEALTH EDUCATION PROGRAM AND VARIABLES AFFECTINGPARTICIPATION*^ Intervening VariablesFactors Influencing^ Between Behavior andBehaviour^ Health* Adapted from Green and Kreuter (1991)^ -38-CHAPTER 3METHODOLOGY3.0 OBJECTIVEThe objective of this research project was to determine which factors, ifany, correlate with participation in a formal seniors' health education program.Such information might prove helpful in enhancing programming to reachnonparticipants who have been unwilling or unable to participate, but who mightbenefit from such a program. At a more general level, it might provide insight intosocial marketing for such programs.3.1 RESEARCH QUESTIONS1) Are there demographic differences in gender, age, educational attainmentor living arrangements between participants and nonparticipants of a community-based seniors' wellness program?2) Are there differences in measures of economic resources, physical health,mobility, emotional balance, level of trust, social support or cognitive status betweenparticipants and nonparticipants of a community-based seniors' wellness program?3) What are the reasons given by participants for their attendance at acommunity-based seniors' wellness program?4) What are the reasons given by nonparticipants for not attending acommunity-based seniors' wellness program? Are there specific barriers to theirparticipation which would also target them as potential participants?5) Are there differences in involvement in other community activities betweenparticipants and nonparticipants of a community-based seniors' wellness program?6) Are there differences in activities performed by, or time spent on performing,daily living activities between participants and nonparticipants of a community-basedseniors' wellness program?-39-3.2 THE PROGRAM3.2.1 Choice of ProgramContact was initially made with Dr. Veronica Doyle, a researcher in VancouverWest Main Health Unit. Dr. Doyle is conducting research to determine the reasons forparticipation in a health related seniors' program and is interested in determiningreasons for non-participation in the future. It was reasonable, therefore, to conduct thisresearch in the same geographical area, to allow comparability of results with regardto population samples.In consultation with West Main Health Unit staff, the Kerrisdale Seniors' CenterHealth Drop-In was determined as a first choice of a health education program for thisstudy for the following reasons:• it is well established and stable,• there are sufficient participants to achieve an adequate sample size,• the program is typical of health education programs for Vancouver seniors,including exercise, education, blood pressure monitoring and massagecomponents.The Program Coordinator of Kerrisdale Seniors' Center was then approached,and was receptive to use of the Health Drop-In for this research project.3.2.2 Description of ProgramKerrisdale Seniors' Center is attached to the Kerrisdale Community CenterComplex and located in an area of high rise apartments, self-owned apartments andsingle family dwellings. It is funded by the Vancouver Parks Board and operated bythe Kerrisdale Community Center Seniors Operational Committee. Members must be55 years or older. An annual membership fee of $7.00 for 55 to 65 year olds and$4.00 for 65 year olds and up is required. A full-time Program Coordinator and dayand evening secretarial support are employed. The Center relies mainly on volunteerassistance by members.-40-Kerrisdale Senior Center, and all of its programs, operate under the VancouverParks and Recreation Mission, which is, "To maintain and enhance the quality of life ofall citizens of Vancouver by ensuring the provision of a parks and recreation system."The established Aims, Objectives and Goals of the Kerrisdale Seniors' Center are1) To encourage involvement and participation in senior center activities.2) To promote healthy social interaction.3) To supply services at many different levels as required, and in different areas.4) To utilise the special skills of volunteers. (Kerrisdale Senior Centre Policy Manual, 1987).The Center operates a variety of programs for seniors seven days a week,including social activities, music, dancing, arts, crafts, educational activities andworkshops on a variety of subjects. In addition, a nutritious lunch is provided six daysper week at a low cost to members (Kerrisdale Senior Centre Promotional Pamphlet,1991).The Health Drop-In is one of several support services offered, and includesblood pressure monitoring, exercise and neck/shoulder massage. In the past, healthlectures and discussions were also a part of the program, but they were discontinueddue to a perceived lack of interest (Pirie, 1991-2). The blood pressure monitoring is afree service to members, operating one morning per week. Volunteer retired nursestake blood pressures, retain a record and provide members with a record. Massage isprovided by a volunteer free of charge two mornings per week. Both the bloodpressure monitoring and the massage are provided on a "drop-in", first-come first-served basis. The atmosphere as members wait is relaxed. Many members sit andvisit with each other while waiting for their turn, thus it is evident that there is also aninformal social component to the program.The exercise program is more structured as members are required to sign up fora twelve week period. It was, in the past, provided free of charge to members. Theloss of a volunteer instructor has compelled the Center to hire a qualified fitnessinstructor, prompting a participant fee of $9.00 per session (once per week for twelve-41-weeks). This fitness instructor conducts regular fitness classes three times per weekand Gentle Fit once per week.The exercise program is evaluated at the end of each session by participants.However, the results of the evaluation are not retained by the Program Coordinator.The Program Coordinator reported that in December, 1991, the participants evaluatedthe exercise program positively, with the exception of some specific comments aboutthe types of exercise they disliked (eg. mat exercises) (Pirie, 1991-2).The blood pressure monitoring component serves 30 to 40 members per week,the neck and shoulder massage component serves 18 to 20 members per week. Theexercise programs have spaces for 40 in each class. Currently, exercise classes arealways full, with members having to be turned away (Pirie, 1991-2).3.2.3 Demographics of the AreaCensus tracts 022 and 009 most closely approximate the area covered in thisstudy (Statistics Canada, 1987). Vancouver had a population over 65 years of agethat represented 11.0% of the total city population in 1986 (Statistics Canada, 1987).The percentage of males 55 years of age and over in the city was 19.6% of the totalmale population, while the percentage of females 55 years of age and over was 23.6%of the city's female population.Seniors (65 years and older) who lived alone were 30.6% of the total seniorpopulation, while 58.4% were classified as "family persons" (probably living with aspouse in this age group), 8.7% lived with a relative and 2.3% lived with a non-relative. The city as a whole had 53.3% of persons living in single detached dwellings,9.9% living in apartments over 5 storeys and the rest living in other types of dwellings.Those individuals with less than high school education represented 36.7% ofthe total city population, while 11.9% had a high school diploma and 51.3% had someform of education or training above the high school level. The average income for cityresidents was $36, 086 and the median income was $30,477.-42-Census tract 022, bounded by Trafalgar, W. 37th Ave., Arbutus and W. 41stAve., had a male population 55 years and older that was 47.3% of the total malepopulation for the area, and 16.5% of the area's total population. Females 55 years ofage and over represented 58.8% of the total female population, and 38.1% of thearea's total population.Overall, 43.4% of the area's population was 65 years and older. Among thisage group, over half (52.2%) lived alone, 41% were listed as "family persons"(probably living with spouses in this age group), 4.4% lived with a relative and 1.9%lived with a non-relative. These figures are consistent with residential types for thearea: a full half of the people lived in apartments over 5 storeys, while 8.8% lived insingle detached dwellings and 41.7% lived in some other type of dwelling.The people in this area were relatively more educated than the city as a whole:23% had less than high school, 13.1% had a high school diploma and 63.7% hadsome education or training beyond the high school level. The average income wasslightly above the city average, at $37,492, but the median income was lower, at$26,678 (perhaps indicative of a greater number of older people living in high riseapartments on fixed incomes).Census tract 009, bounded by McDonald, W. 41st Ave., Granville, and W. 49thAve., had a different profile (Statistics Canada, 1987). Approximately one quarter(24.5%) of the population was 65 years or older, which is higher than the city averagebut lower than census tract 022. Males 55 years and older represented 28.4% of thetotal male population for the area and 12.0% of the area's total population, whilefemales 55 years of age and older represented 40.2% of the area's total femalepopulation and 23.4% of the area's total population.Among those who were 65 years of age and older, 47.8% lived alone, 44.1%were family persons, 5.7% lived with a relative and 2.7% lived with non-relatives. Thearea includes 40.9% single detached dwelling residents, 15.9% residents of-43-apartments over 5 storeys and 43.2% residents of other types of dwellings. A largeproportion of people (70.8%) had education or training beyond the high school level,while 7.2% had a high school diploma and 21.9% had not completed high school.The average income ($45,936) and median income ($31,106) were both higher thanthe totals for Vancouver.3.3 RESEARCH DESIGN3.3.1 SubjectsSamples were drawn from three sources:• participants of the Kerrisdale Seniors' Center Health Drop-In, including theBlood Pressure Monitoring, Neck and Shoulder Massage and Exercisecomponents,• a sample of senior citizens residing in single family dwellings within a five-block radius of the Kerrisdale Seniors' Center and• a sample of senior citizens residing in apartment buildings within a five-blockradius of the Kerrisdale Seniors' Center.It was hoped that a sample approaching 100 participants could be drawn fromthe Health Drop-In attenders, 50 from single family dwellings in the surrounding areaand 100 from apartment buildings in the surrounding area.Health Drop-In attenders were approached by the researcher at the KerrisdaleSeniors' Center during times that the programs were running. The research study wasexplained to them and they were asked to complete the questionnaire. They weregiven the choice of completing it immediately and returning it to the researcher, ortaking it home and leaving it in a folder at the Kerrisdale Seniors' Center office at theirconvenience.To attain a random sample of nonparticipants from the geographical areasurrounding the Seniors' Center, a block map of the area including a five-block radiussurrounding the Seniors' Center (reasonable walking distance) was traced. A-44-computer-generated table of random numbers was utilised to assign a number to eachblock in the area, which determined the order in which the block was visited. (Blocklist included in Appendix A.) At each residence where there was someone home, theresearcher introduced herself, presented University of British Columbia identificationand asked if any senior citizens were in residence. Where there were seniors residingat the dwelling, the research purpose and method were briefly described and thesenior residents were asked to participate. If they agreed, the questionnaire was leftwith them overnight and picked up the next day by the researcher. A covering letterwas also left, which described the study and stated that participation in the study wasvoluntary, refusal would not jeopardise receipt of any services and completion andreturn of the questionnaire would be assumed to indicate consent. Individuals wereinvited to receive a summary of the study results; a separate form was included onwhich they could provide their name and address for this purpose if they wished.The entire five-block radius was covered before attaining a sample of fiftyparticipants from single family dwellings. This design also allowed various areas to bevisited at different times of the day, so if there were differences in activity patternswithin the geographical area, certain types of individuals would not be systematicallyexcluded by the time of day their residence was visited.To contact residents of apartment buildings, the same introduction andidentification routine was followed with apartment managers. Managers were asked todetermine the most appropriate means of contacting senior residents, observingsecurity protocol for that particular building. In some apartments, the manager took theresearcher door to door to the senior residents, which provided assurance to theseniors that the research was legitimate and permitted face to face contact withpotential participants. In other cases, the managers put the questionnaires under theresidents' doors or into their mailboxes. In two cases, the researcher leftquestionnaires in an open area beside the apartment mailboxes, so that residents-45-could pick up a questionnaire if they wished. There were six apartment managers whowere unwilling to participate in any form. In these cases, access could not be gainedto the seniors. There was also no practical way to access residents of ownedapartment buildings, as there was usually no manager and the buildings utilisedsecurity systems. Where face-to-face contact was possible, apartment participantswere given the choice of returning the questionnaires to the researcher on thefollowing day, or leaving them with the apartment manager, if that was moreconvenient for them. Where apartment managers acted as a "go-between",participants were asked to leave the questionnaires with the apartment manager.Return rates were found to be very low unless face-to-face contact could bemade with the potential participants. Therefore, three apartment managers wereapproached through contacts within the West Main Health Unit, ensuring theopportunity to go door to door and, thus, increasing participation rates while allowingestimates of non-participation rates. It was recognised that this counteracted therandom selection process, and therefore, was less than ideal. However, thealternative of low participation rates with no idea of the characteristics or numbers ofnonparticipants was considered more problematic (ie. would have compromisedrandom selection even more).3.3.2 Choice of QuestionnaireFactors identified previously in the Literature Review section as targetcharacteristics for health education programs directed to the elderly included physicalhealth, mobility, mental health, social network, health behaviours and economicfactors. Therefore, these were the variables of interest to this study. There are anumber of general health status measurement tools which include all or most of thesefactors. Choice of an instrument to be used for this study was based upon a number ofcriteria.Bergner (1985) stated that a health status measure should be meaningful,-46-understandable, sensitive to change, theoretically justifiable and intuitivelyreasonable. Component parts should be clear, and the data necessary to calculatethe measure should be obtainable. Feinstein, Josephy and Wells (1986) stated thatthe first step in choosing a scale is to determine the specific purpose and setting forwhich it is to be used. If there is a choice of suitable tools for the intended purpose,then the issues of statistical reliablility and validity can be addressed. This wassupported by Bergner, Kaplan and Ware (1987), who indicated that choice of a tool isdependent on the goals and design of the study, as well as the outcomes expected.Ware (1987) noted that the broader the instrument the more it increased in length. Thepurpose of information-gathering must be ascertained to focus the instrument andkeep it lean. Pfeiffer (1975) supported this, suggesting that only information itemswhich will be used and have action implications should be collected.In addition, the degree of ease with which an index can be administered andanalysed must be considered (Bergner et al., 1987; Feinstein et al., 1986; Pfeiffer,1975). In evaluating three measures of health, Read, Quinn and Hoefer (1987)compared the following factors: training time for interviewers, difficulty ofadministration, coding and scoring difficulty, comprehensibility for subjects andadministration time.The assessment methodology must be valid; that is, it must measure what itpurports to measure (Pfeiffer, 1975). Evaluating the validity of health status measureswas said to be difficult because there is no gold standard for health (Read et al., 1987).Recorded methods for assessment included content validity (careful inspection of theitems to confirm that they address the appropriate domains), construct validity(comparison of results between the tool of interest and other measures to confirm thatthey are related in a logical way based on assumptions of cause and effect),convergent construct validity (the measure correlates with other measures with which itis causally related, based upon an accepted theory), discriminant validity (the measure-47-correlates more highly with other measures of the same variable than with measuresof related but different variables). (Read et al., 1987; Ware, Brook, Davies and Lohr,1981; Jette, 1980 ).Other considerations discussed were that the measure is reliable, producing thesame results for the same populations over time, different locations and numerousadministrators (Pfeiffer, 1975). It should also provide quantitative data, permittingaggregation of data from the individual to the population level (Pfeiffer, 1975).In the choice of measurement instruments for this study, the framework utilisedwas by Ware et al. (1981). This framework suggested that the first step in choosing atool is to define the reasons for measurement of health status, which fall into five broadcategories: (1) measuring efficiency or effectiveness of medical interventions, (2)assessing quality of care, (3) estimating the needs of a population, (4) improvingclinical decisions or (5) understanding the causes and consequences of differences inhealth. When studying general populations, use of positively defined health measureswas perceived as providing information about the greatest percentage of subjects.The second step, according to Ware et al. (1981) is to define the aspects ofhealth which are being addressed in the particular study. The components of healthwhich may be of interest include physical health (the physiologic and physical status ofthe body), mental health (the state of mind, including basic intellectual functions) andsocial functioning (the quantity and quality of social contacts and resources). Jette(1987) added general health perceptions to this list and defined this category as self-ratings of health based upon the notion that "your health is what you think it is."Once it is clear what is to be measured, possible instruments are evaluatedusing the criteria of practicality, reliability, validity and subjectivity/objectivity. Practicalconsiderations include affordability of interviews versus self-administered instruments,respondent burden (as indicated by refusal rates, missing responses or administrationtime), and choice of the least complicated instrument and method possible. Shorter-48-scales and any scales applied to disadvantaged groups (eg. lower income oreducation) tend to be less reliable.Important issues in validity are ensuring that the measures chosen include allrelevant components of health as well as the specific aspects of each componentwhich must be measured, and ensuring that the measures are not confounded withother variables which will confuse or bias the results (Ware, 1981). This may meanthat a subset of a comprehensive battery is utilised, rather than the entire instrument, ifthe subset's indicators pertain most closely to the studied variable. Objectivemeasures have generally been more preferred than subjective measures, based onthe assumption that they are more reliable and that they agree most closely withratings by trained professionals. Subjective ratings, however, are more useful ingeneral populations where undiagnosed conditions may be a factor. They also allowfor finer discrimination among people who fall along the full range of the health statuscontinuum. Subjective ratings also have more relevance for quality of life and healthpromotion planning, because they take into consideration the personal meaning andimportance of the condition to the individual (Green, 1992).Jette (1987) and Ware (1985) argued that subjective measures of health tapadditional facets of health that are missed by objective means. Individual differencesin health such as level of effort required, pain, difficulty, worry, wellness, lifesatisfaction, psychologic distress and wellbeing, vitality and concern about health canonly be measured through self-report. Ware (1985) considered these subjectiveratings to be the most sensitive measures of age-related deterioration in health andfound them to be the best predictors of future medical care expenditures and the bestpredictors of survival. Agostino (1985) also stated that subjective ratings are aneconomical way of gaining health information about the elderly, and, following areview of relevant literature, concluded that there is evidence of a positive relationshipbetween self-ratings and physician ratings of health; that they measure something-49-different from physician evaluations but are persistently related to objective measuresof health and, therefore, could be used in survey research of older persons whereobjective ratings are not feasible. Morris and Boutelle (1985) determined that self-responses on a questionnaire were as reliable as information derived from aninterview for a group of relatively independent, well elderly people. And Dorevitch,Cossar, Bailey, Bisset, Lewis, Wise and MacLennan (1992) found that, for a group of150 elderly patients attending a geriatric day care, self ratings of ADL performancewere more accurate and less biased than ratings by either health professionals orcaregivers, judged according to direct observation of the patient undertaking theactivity.The present study falls primarily into the category, among the five purposes inthe Ware et al. (1981) framework, of "estimating the needs of a population."Secondary purposes include improvement of planning decisions and understandingthe causes/consequences of differences in (perceived) health. Physical, mental andsocial health were all important, as were the subjective components discussed byJette (1980) and Ware (1985). Because a general population was to be studied,respondent burden was an important consideration: the time, complexity andcommitment required needed to be kept to the minimum possible. Recognising thatreliability must sometimes be sacrificed for brevity, the more important consideration inthis case was to choose a short, simple survey, as groups of people were beingcompared rather than individual health status being measured. Subjective ratingswere the method of choice, as objective ratings were not feasible with a generalpopulation within the resources available for this study.With these considerations in mind, a number of possible instruments werereviewed, including the HRCA Vulnerability Index (Morris, Sherwood, Mor, 1984), theSickness Impact Profile (SIP) (Bergner, Bobbitt, Carter and Gilson, 1981; Stahl, 1984),the Duke-UNC Health Profile (Parkerson, Gehlbach, Wagner, James, Clapp and-50-Muhlbaier, 1981), the Life Satisfaction Indexes (Neugarten, Havighurst and Tobin,1961), the OARS Multidimensional Functional Assessment Questionnaire (Fillenbaumand Smyer, 1981; Pfeiffer, 1975; Ernst and Ernst, 1984), an abbreviated version ofRotter's Internal-External Locus of Control (Valecha and Ostrom, 1974), theComprehensive Older Persons' Evaluation (COPE) (Pearlman, 1987), the MOS Short-form General Health Survey (Stewart, Hays and Ware, 1988), the FunctionalAssessment Screen (Pannill, 1991), the Multilevel Assessment Instrument (MAI)(Lawton, Moss, Fulcomer and Kleban, 1982), the Comprehensive Assessment andReferral Evaluation (CARE) (Gurland, Kuriansky, Sharpe, Simon, Stiller and Birkett,1977-78; Eustis and Patten, 1984), the Assessment of Living Skills and Resources(ALSAR) (Williams, Drinka, Greenberg, Farrell-Holtan, Euhardy and Schram, 1991)and the Iowa Self Assessment Inventory (ISAI) (Morris, Buckwalter, Cleary, Gilmer,Hatz and Studer, 1990; Morris, Andrews, Gilmer, Buckwalter, Cleary, Boutelle, Hatz,1991; Gilmer, Cleary, Lu, Morris, Buckwalter, Andrew, Boutelle, Hatz, 1991).Instruments were eliminated on the basis of length, complexity, usefulness foran elderly population, usefulness for a general, healthy population (versus a patientpopulation) and measurement of only one component of health status.A short list of instruments included the Multilevel Assessment Instrument (MAI)(Lawton et al., 1982), the Duke-UNC Health Profile (Parkerson et al., 1981), theComprehensive Older Persons' Evaluation (COPE) (Pearlman, 1987), the MOS Short-form General Health Survey (Stewart et al., 1988) and the Iowa Self AssessmentInventory (ISAI) (Morris et al., 1990; Morris et al., 1991; Gilmer et al., 1991). Thesewere further evaluated based on the same criteria, with particular attention to inclusionof all relevant health status components and respondent burden.The final choice of an instrument consisted of two parts:• the ISAI (Morris et al., 1990), a standardised multi-dimensional healthmeasurement instrument which includes measures of self-perceptions of economic-51-resources, emotional balance, physical health, mobility, trusting others/alienation,cognitive status and social support, and• a series of questions determining involvement or lack of involvement in theKerrisdale Senior Centre Health Drop-In, reasons for participation or lack ofparticipation and involvement in other community and daily living activities.3.3.3 The Iowa Self-Assessment InventoryThe Iowa Self-Assessment Inventory (ISAI, Morris et al., 1991) is a 56-item self-report instrument designed to measure the resources, needs, statuses and abilities ofolder persons. It is intended for use either with individuals or to gain a betterunderstanding of groups of elderly people through survey research methods.The ISAI is based on the Multidimensional Functional AssessmentQuestionnaire (OMFAQ; Pfeiffer, 1975), also referred to as the OARS questionnaire(Morris and Boutelle, 1985). It was originally designed to provide a method ofassessing large numbers of elderly individuals in a less costly, time-intensive mannerthan is required with the OARS.Initially, the OARS questions were reworded so that respondents could provideself-report answers, and, using a double blind method, it was concluded that self-administered assessment by literate, relatively independent elderly persons was asreliable as results obtained from the interview format (Morris and Boutelle, 1985;Morris and Buckwalter, 1988). An inventory was then developed which encompassedsix scales tapping social resources, economic resources, mental health, physicalhealth, ADL ability and cognitive status (Morris, Buckwalter, Cleary, Gilmer, Hatz andStuder, 1989). A Likert-type scale, using True, More Often True Than Not, More OftenFalse Than Not and False was utilised for all questions, making the inventory relativelyeasy to use for respondents. The original scale, containing 120 items, was tested onan American sample of elderly people who resided in public housing projects,congregate meal sites, community-based groups and retirement homes. Of the 2,347-52-surveys sent out, 1,850 were returned and after removing incomplete surveys or thosewhich were not appropriate (less than 60 years old, patients in a geriatric clinic), 1,153surveys were analysed for reliability of the subscales. Acceptable reliabilities werefound, so the researchers continued with the next step, which was to reduce thenumber of questions through factor analysis (Morris et al., 1989).Morris and his colleagues subjected their original 1,153 surveys plus anadditional 420 from similar sources to factor analysis, leading to the isolation of fourmeaningful factors: perceptions of economic resources, perceptions of physical health,perceptions of cognitive status, and a new factor which pertained to mobility (Morris,Buckwalter, Cleary, Gilmer, Hatz and Studer, 1990). Social support and mental healthdid not emerge as viable factors. The researchers, therefore, developed newexperimental inventories for these two factors, based upon existing instruments. Theypresented the new inventories to their 420 respondents, then performed another factoranalysis which confirmed the existence of clear factors relating to anxiety/depression,alienation and social support.They selected eight items with the highest loading for each of the seven factorsfor the final version of the ISAI, as an eight-item solution had slightly higher reliabilitiesthan six or seven-item scales, but similar reliabilities to a nine-item solution.Reliabilities for the seven factors were as follows: anxiety/depression, 0.84; alienation,0.74; social support, 0.75; economic resources, 0.86; physical health, 0.80; mobility,0.79; cognitive status, 0.82.Morris et al. (1990) suggested the following definitions for the factors:• Anxiety/Depression: a high score indicates a relatively worry-free, calmindividual who sleeps well and enjoys a tranquil life;• Alienation: a high score indicates an individual who believes they havereliable, trustworthy friends, is friendly toward others, and is generally friendly ininterpersonal relationships;-53-• Social Support: a high score indicates a perception of a comfortable socialenvironment and close relationships with friends and relatives;• Economic Resources: a high score indicates a perception of adequateincome and assets, with no need for outside financial assistance;• Physical Health: a high score indicates a perception of excellent health, withfew doctor visits or few prescribed medications;• Mobility: a high score indicates an ability to carry on usual ADL and to get outand visit friends and relatives or participate in other social activities;• Cognitive Status: a high score indicates a perception of intact intellect, goodmemory, orientation and a continued ability to learn. (Morris et al., 1990).Gilmer, Cleary, Lu, Morris, Buckwalter, Andrew, Boutelle and Hatz (1991)developed a measurement model for the seven factors, to confirm that the specifiedrelationships between items and dimensions, in fact, existed. They administered theISAI to 511 elderly persons in a predominantly rural area. Respondents wereresidents of retirement homes, participants of congregate meal programs, or residentsof public housing projects for low income elderly. Respondents were 84% female, withan average age of 74 years. Due to missing data, they used 484 surveys in theanalysis. Their factor analysis confirmed the existence of six to eight factors. Theseven-factor solution fit the data the best, and the loadings for individual items in theseven factor solution indicated that 53 (of the 56) items had their highest loadings onthe predicted dimensions, with only nine items having loadings of less than 0.40.Reliabilities for the seven factors were: emotional balance (anxiety/depression), 0.84;trusting others (alienation), 0.71; social support, 0.79; economic resources, 0.84;physical health, 0.76; mobility, 0.78; cognitive status, 0.79. Intercorrelations betweenscales were moderate, suggesting separate factors, but were highest wherecorrelations would be expected (eg. trusting others, emotional balance and socialsupport).-54-Scores for each scale are derived by coding each item from one (for lowestfunctioning) to four (for highest functioning). Each of the seven scales ends up with araw score between 8 and 32, with 8 denoting poorer functioning on that particularscale and 32 indicating high functioning. Negatively worded items are reversed topreserve the direction of scoring.Because the means and standard deviations of the raw scores vary significantlybetween scales, they converted the raw scores to standard T-scores, with a mean of50 and standard deviation of 10 (Morris, Andrews, Gilmer, Buckwalter, Cleary, Boutelleand Hatz, 1991). The ISAI and its score Profile are included in Appendix B.The validity of the ISAI stems from its derivation from the OARS questionnaire,which may be considered the "gold standard" for geriatric multidimensional functionalassessment, as it was the first integrated system for evaluating older people and,therefore, provided a model against which possible improvements could be compared(Moss et al., 1982; Morris and Buckwalter, 1988). The original version of the ISAI wasfound to provide results which were as reliable as those obtained from the OARS(Morris and Boutelle, 1985). The additional scales added to the ISAI (social support,emotional balance and trusting others) were reviewed by a panel of 20 experts fromacross the U.S. (from the disciplines of geriatrics, geriatric nursing, anthropology,psychology, education, social work and public health) for content validity, withemphasis upon relevance and inclusiveness of items, clarity, readability and format(Morris et al., 1989). In terms of construct validity, predicted correlation betweenspecific demographic variables and the appropriate scales were confirmed (eg. strongpositive relationship between ER and income, strong negative correlation betweenADL and age) (Morris et al., 1989). It was suggested by the authors that, where resultsof the ISAI are to be used to make decisions about individuals, they should be treatedas one piece of information along with other relevant sources of personal informationabout financial and health status (concurrent and convergent validity) (Morris et al.,-55-1991).Other validity considerations in administration of the ISAI include "faking",acquiescence and negative response bias, random responses and special populationconsiderations (Morris et al., 1991). Due to the self-report nature of the ISAI, there maybe a tendency for individuals to distort their ratings in a socially desirable direction. Itwas suggested that scores deviating 0.5 or more standard deviations from the meanbe interpreted cautiously. The researchers defined acquiescence as "the tendency torate all or almost all items "true", and negative response bias as "the tendency todisagree with the items regardless of their content" (Morris et al., 1991, p. 13).Random responses may occur if there is a lack of motivation on the respondents part, ifsome of the items are not understood or if the respondent has mis-scored some of theitems.The ISAI has not yet been standardised with ethnic populations, although suchwork is presently underway. It is intended for use with adults age 60 and over and isnot recommended for individuals who are unable to read or cognitively unable tomake a response.3.3.4 Questions Specific to the Kerrisdale Health Drop-InIn addition to measures of health, the following questions were thought to beimportant to obtain information about:• age, gender, educational status, whether living alone or with someone,• whether living in apartment or single family residence,• knowledge of the Health Drop-In,• attendance or nonattendance at the Health Drop-In, and (for participants)specific parts of the program which were attended,• for participants, length and frequency of attendance, source of knowledge andreasons for attendance,• for nonparticipants, reasons for nonattendance,-56-• involvement in other community activities,• time spent on various ADL, such as housework, hobbies, visiting, physicalactivities and community activities.These questions would serve the purposes of: determining eligibility for thisstudy, determining category of the study (participant or nonparticipant of the HealthDrop-In), describing characteristics of participants and nonparticipants of the HealthDrop-In, determining reasons for attendance or nonattendance including possiblebarriers to participation for nonparticipants, and determining differences in patterns ofactivity for participants and nonparticipants.The original draft consisted of both open-ended and categorical responsequestions. It was presented to members of the thesis committee (Drs. Sheps, Greenand Thornton) for review of content and format, and revised twice. The final formatprior to piloting the questionnaires consisted of 16 questions, all requiring categoricalresponses. Generally, the questions were ordered with the easiest to answer andleast threatening ones at the beginning, and the most difficult to answer at the end.The first five questions provide demographic data (age, gender, educationalattainment, place of residence and whether living alone or other arrangement).Questions 6 to 12 determine knowledge of the Health Drop-In, participation in programand in specific program components, length and regularity of attendance and sourceof knowledge about the program. The last four questions ask for reasons forattendance or nonattendance, involvement in other community activities and hours perweek spent on ADL.A) INCLUSION AND EXCLUSION CRITERIA: As membership in the KerrisdaleSenior Center begins at 55 years of age, eligibility for inclusion in this study wasdefined as, "persons 55 years of age or older, not living in a nursing home or otherlong-term care facility." This was similar (exept for the age chosen) to a definition of"well elderly" utilised by Lalonde, Hooyman and Blumhagen (1988) in a study-57-determining the long-term effectiveness of the Wallingford Wellness Project, a healthpromotion program for the elderly. The first question, asking for birthdate, and the fifthquestion, asking for place of residence, provided the necessary information fordetermining eligibility for inclusion. Exclusion criteria included inability tocommunicate in the English language or cognitive impairment to the extent that theindividual was unable to complete the ISAI.B) DETERMINATION OF CATEGORY: Participants were defined as those whoanswered "Yes" to current attendance at the Health Drop-In, and any category except"Once" to length of attendance. By excluding first-time attenders, individuals who werechecking out the program but decided that they would not participate were notcaptured as participants.C) REASONS FOR NONATTENDANCE: Eighteen categorical answers plus an"Others (please list)" category were chosen, based upon the literature review.Particularly helpful in deriving individual items was a list of barriers to physical activityby older adults derived by O'Neill and Reid (1991). This consisted of 20 possiblebarriers to physical activity based upon a framework of "knowledge, psychological,physical/health and administrative" reasons.The eighteen items were classified according to where they best fit on Greenand Kreuter's (1991) "Predisposing, Enabling, Reinforcing" model of Educational andOrganisational diagnosis within the Precede-Proceed framework. Predisposingfactors are defined as "those antecedents to behavior that provide the rationale ormotivation for the behavior" (Green and Kreuter, 1991, p. 151.) They includeknowledge or awareness, beliefs, values, attitudes, self-efficacy, behavioural intentionand existing skills (pp. 154-61.) Enabling factors are "the antecedents to behavior thatenable a motivation to be realised " (p. 151). They are often conditions of theenvironment and include availability, accessibility and affordability of the behavior orprogram. Also included in this category are acquisition of new skills which facilitate-58-performance of the health behaviour (p. 161-65). Reinforcing factors are defined as"factors subsequent to performing, or contemplation of, a behaviour that provide thecontinuing reward or incentive for the behaviour and contribute to its persistence orrepetition" (p. 151). The individual receives positive or negative feedback as a resultof engaging in the behaviour, including social support, peer influences, advice andfeedback from health care professionals, feelings of wellbeing or pain, etc. (p. 165).Six predisposing items were included (would not enjoy it, don't like to leavehome, won't do me any good, don't need it, not interested and health is not importantto me). Seven enabling items (don't know about it, too far to centre, can't do exercise,can't afford it, not enough time, poor health and no transportation) and five reinforcingitems (doctor says not to, don't know anyone there, friends/spouse don't go, don't feelwelcome there and friends/spouse think it's silly) were included. The items chosen andthe classification for items were checked with Dr. Green, who suggested some of theitems and provided assistance in coding according to categories. The "my doctor saysnot to" item, which was coded under reinforcing factors, could also be an enablingfactor like "poor health" and it was decided to check the data to determine whetherindividuals who chose "doctor says not to" also tended to check "poor health."D) REASONS FOR ATTENDANCE: The same framework of predisposing,enabling and reinforcing factors was utilised to provide structure to reasons forattendance. This framework was chosen over two other possibilities. Boshier's (1991)Education Participation Scale (E.P.S.) was considered, but also included a number ofnon-relevant factors that would have increased the length of the questionnaire withoutproviding any more useful information, and might have confused or biased the resultsby adding non-relevant variables (Ware, 1981). (However, an earlier version of theE.P.S. (Boshier and Riddell, 1978) which was specific to older people was of greatassistance in deriving individual items.) Green and Kreuter's loose breakdown ofreinforcing factors into Physical Benefit vs. Social Benefit and Tangible Reward vs.-59-Vicarious Reward was also considered.Eighteen items were chosen, along with an "others - please list" category. Sixpredisposing items were included (health is important to me, interested in learninghow to keep healthy, need the exercise, want to stay active, want to continue living athome and to participate in volunteer or community work). There were four enablingreasons (have lots of time, centre is close and easy to get to, already there for otheractivities and good at exercise) and eight reinforcing reasons (relief from boredom,enjoy exercising, break in routine, to make friends, to participate with a group, to beaccepted by others, friends and/or spouse go and doctor told me to go). "Doctor toldme to go" could be classified as a predisposing reason as well as a reinforcing one,and it was decided that responses would be compared to determine whetherindividuals who chose this particular item tended to choose other reinforcing items orother predisposing items.E) DESCRIBING PATTERNS OF ACTIVITY: Two questions were includedwhich related to patterns of activity. One question asked about involvement in othercommunity programs. Respondents were asked to state whether they were attenders,or leaders/volunteers for various programs in the Kerrisdale area. The list wascompiled with assistance of West Main Health Unit staff and the Kerrisdale SeniorCentre program coordinator, and included: other programs at Kerrisdale SeniorCentre, West Main Health Unit activities, church, church groups or organisations,cultural centres or organisations, symphony, opera, etc., recreational activities, othersenior centres or organisations and the Kerrisdale Community Center. Spaces wereleft for other programs not listed.The second question requested hours of activity per week spent on housework,physical activities, hobbies, visiting friends or relatives and community activities.Information from a sample of 100 people over 50 years of age from an EasternCanadian city suggested that the majority of time would be allotted to activities within-60-the home (Singleton, Mtic and Farquharson, 1986). These authors, however, basedtheir interviews on activities of the past 24 hours, and noted in their discussion thatsome activities occur infrequently, so a weekly, rather than daily, time frame should beexplored. Singleton et al. (1986) found gender to be a determining factor in actual andfavorite activity choices. Age (with community involvement increasing until the age of72 years), income, educational attainment (positively correlated with communityinvolvement), marital status (married individuals more involved in community) and typeof community (retirement communities showing more integration than either planned"new" or control communities) were also found to be significant factors in activitypatterns, in a secondary analysis of a in North Carolina environmental preferencesand quality of life survey (n=873) (Spakes, 1979).A secondary analysis of 1398 senior citizens who participated in a 1986Canadian survey on time use, social mobility and language use determined that olderindividuals spend an average of 13.4 hours per day on self care (including sleep, ADLand shopping and services), 3.1 hours on productivity (paid employment, child care,school and education, organisations) and 7.5 hours on leisure (entertainment, sportsand hobbies, media) (McKinnon, 1992).The above information from the literature was considered to be important ascomparison data to that which is derived from this study.3.4 SPECIFIC HYPOTHESES1. a) Scores in the normal range (less than one standard deviation above orbelow the mean) on the ISAI economic resources, emotional balance, trusting others,mobility, cognitive status, physical health and social support scales will be associatedwith participation in the Health Drop-In.b) Gender will be associated with participation (participants will be more likelyto be female).-61-c) Participation in other Kerrisdale Senior Centre activities will be associatedwith participation in the Health Drop-In, as will involvement in other communityactivities, but generally as a attender rather than a leader or volunteer.d) Choice of predisposing and reinforcing reasons for attending will be mostcommon among participants.2. a) Higher mean scores than those of the participants on the economicresources, mobility and social support scales will be associated with one subset ofnonparticipants.b) Physical health, trusting others, emotional balance and cognitive status scalescores will be equivalent to participant scores.c) Gender (more females) and educational level (greater than high school) willbe associated with this subset of nonparticipants.d) Active involvement in other community activities, especially in a leader orvolunteer (as opposed to attender) capacity, and reporting of more hours of activity perweek will be associated with this group.e) Past participation in the Health Drop-In may be associated with this subset ofnonparticipants.f) Reasons for nonattendance such as lack of interest, lack of need, lack ofenjoyment of or lack of time for the activity will more often be chosen by this subset ofnonparticipants.3. a) Lower scores on all the ISAI scales will be associated with a second subsetof nonparticipants.b) Gender and educational attainment will be associated with this subset, withgreater representation of males and higher proportions of individuals with less thanhigh school education being represented.c) Lower levels of community involvement and lower attendance at otherKerrisdale Senior Centre activities will be associated with this subset of-62-nonparticipants.d) Reasons for nonattendance will more likely be enabling factors, and will beassociated to results in the ISAI (for example, lack of knowledge associated with lowerscores in social support, alienation and/or cognitive status scales; too far to the center,no transportation or poor health associated with low scores on mobility and physicalhealth scales; can't afford it associated with low scores on the Economic Resourcesscale; don't like to leave home, don't feel welcome there associated with low scores onthe Trusting Others scale).3.5 ANALYSISComparison of results was completed using a Macintosh LC computer andStatview SE + Graphics statistical software (Feldman, Hofmann, Gagnon andSimpson, 1987). Categorical data were compared using rxc contingency tableanalysis. Age, hours of activity per week and results on the ISAI scales werecompared using nonparametric techniques (Mann Whitney U for 2 groups or KruskalWallis for more than two groups) as data results did not approximate normaldistributions.In calculating ISAI scores, where responses were missing, a score wascalculated if there were one or two item responses missing for that particular scale, butif more than two were absent, a score was not calculated. For example, if only sixresponses were given by an individual for the Economic Resources scale, and thesum of the six responses given was 21, then a score out of 32 would be calculated(21*32/24) to be 28. The Trusting Others (TO) scale seemed to pose difficulties forsome of the respondents. Many individuals did not complete some of the items, suchas "I believe I am being plotted against" and "People secretly say bad things aboutme." Consequently, this scale had the greatest number of unusable scores (8).Each of the independent variables was first subjected to analysis with regard to-63-place of residence (also an independent variable). Where place of residence wassignificant, other results must be interpreted with caution, as place of residence maybe a confounding variable.Each independent variable was analysed according to its relationship withparticipation in the Health Drop-In, and in each of the three components (BloodPressure Monitoring, Exercise and Neck/Shoulder Massage). The variable was alsocompared between Exercise participants who answered as nonparticipants (P2)versus those who perceived themselves as participants (P1).Results of the statistical tests are reported if they were below p = .05. Due to thelarge number of tests that were carried out, it is possible that some of the results whichare reported here were due to chance rather than true differences between groups.Reporting results which were significant below the p = .01 or .001 level wasconsidered, but it was decided to report all results in order to show trends due to apotential lack of power resulting from low numbers. Marginal results should beinterpreted with caution.3.6 PILOT OF QUESTIONNAIRESThe Concerned Citizens for Affordable Housing, a seniors' political andlobbying group with support from the West Main Health Unit Wellness Coordinator,volunteered their support and time to pilot the questionnaires. Questionnaires wereprovided to the Wellness Coordinator, who distributed them to the members. Theresearcher attended a regular meeting to pick up the completed questionnaires andreceive verbal feedback from members. Five questionnaires, plus one from theWellness Coordinator were returned, which represents all individuals who attendedthe meeting at which the forms were collected.Pilot respondents consisted of one male and four females. One individual hadless than high school level, one had finished grade twelve, while the other three had-64-post-secondary education. Four lived alone, one with a spouse. Four resided inapartments or suites, while one lived in a single family dwelling. The group ranged inage from 60 to 81 years, with a mean age of 70.4 years.Feedback with regard to the questionnaire included:• One statement on the standard "Directions" for the ISAI provided in themanual ("In this way we hope to understand some of your problems and needs") wasthought to imply a perception of elderly people as "sick" or "needy". The groupsuggested either changing the wording to the positive or dropping the statement. Thestatement was dropped in the final version distributed to study participants, as it wasfelt to relate more to individual assessment than to group comparisons.• Question number 50 of the ISAI ("I use food stamps") is not applicable to aCanadian sample. Suggestions for revision were solicited from the group, andincluded the food bank, the Guaranteed Income Supplement (GIS, a supplement tothe Old Age Pension) and GAIN. The GIS was chosen for the final version, as it is anorganised government program similar to food stamps in the United States.• The group was asked to pay particular attention to the questions which werenot part of the ISAI, as they had not been tested. There were no voiced concerns withthe wording of any questions, group members felt that all possible reasons forparticipation or non-participation had been included and all possible communityactivities were covered.• Members felt that the categories in Question 16 of the Kerrisdale-specificquestions could be revised for ease of understanding and to provide more usefulinformation. They were concerned that there might be some confusion between"housework" and "physical activity" and suggested changing these categories to"household activities" and "exercise" to make the distinction more clear. They alsosuggested the addition of "gardening", which is not necessarily considered eitherhousework or exercise, and "sleep". These changes were incorporated into the final-65-version.Pilot results on the ISAI showed mean values on each scale within the averagerange. The economic resources scale had a range of values from 23 to 31, with amean of 28.4, which is a standardised score of approximately 52.5 (slightly above thestandard mean of 50). The emotional balance scale showed a range of 14 to 30, witha raw mean of 23.2, or a standardised score of approximately 47. This was the onlyscale with a result below the standard mean of 50. The Physical Health scale had arange of 19 to 32 and a raw mean of 25.8, which converted to a standardised score of52. On the Trusting Others scale, the range of raw scores was from 29 to 32, with araw mean of 30.4 and a standardised score of 52. The Mobility scale showed a rawmean of 28, with a range from 20 to 32. The standardised score was 52. On theCognitive Status scale, the raw mean was 24.2, with a range of values from 18 to 30,and a standardised score of approximately 52.5. Finally on the Social Support scale,the range of values was 26 to 32 with a raw mean of 30.2, which converted to astandardised score of approximately 53.5. Table 1 summarises the data from the pilotgroup.TABLE 1CHARACTERISTICS OF PILOT STUDY SAMPLE (N = 5)RANGERAWMEANSTANDARDISEDSCOREAGE (years) 60-81 70.4EDUCATION Gr. 10 - PhDECONOMIC RESOURCES 23-31 28.4 52.5EMOTIONAL BALANCE 14-30 23.2 47PHYSICAL HEALTH 19-32 25.8 52TRUSTING OTHERS 29-32 30.4 52MOBILITY 20-32 28 52COGNITIVE STATUS 18-30 24.2 52.5SOCIAL SUPPORT 26-32 30.2 53.5These results were encouraging, as it was expected that the ConcernedCitizens for Affordable Housing would be a comparatively high functioning group of-66-senior citizens, and would, therefore, score in the high ranges on the ISAI. Their ISAIprofile, with the exception of the emotional balance scale, supported this expectation.The revised version of the Kerrisdale specific questionnaire was presented tothe Kerrisdale Senior Centre Program Coordinator, who suggested some changes tocategories on the question related to length of attendance, based upon the fact thatmany Health Drop-In participants have been attending for years. The final version ofthe questionnaires, as presented to study respondents, is included in Appendix C.CHAPTER 4RESULTS4.0 DATA COLLECTIONData collection was carried out between the period of May 13 and June 17,1992. Survey questionnaires were distributed to 357 Kerrisdale Senior Centreparticipants, residents of single family dwellings and residents of apartment buildings.A total of 102 surveys were distributed to participants at Kerrisdale SeniorCentre. Fifty-eight of these surveys were returned completed: a return rate of 56.9%.Two additional surveys were returned uncompleted. Seven individuals refused toparticipate and were not provided with a survey. The reasons given for refusalincluded: poor English (3), no time (1), not liking surveys (1), first time participant (1)and not good at questionnaires (1).Sixty-two residents of single family dwellings who were eligible to participatewere approached. Fifty-six surveys were distributed. Four were not returned (oneindividual stated that she would mail it, but failed to do so, the other three wereunavailable when the survey was to be picked up). One additional survey wasreturned uncompleted. This resulted in the return of 51 of the 56 surveys, for a returnrate of 91.1%. There were 8 refusals, of which one had poor English skills and onepresented as confused and would, therefore, not be eligible to participate underinclusion criteria. Five of the refusals were female, three were male. It was notpossible to estimate the number of single family dwelling resident seniors who werenot contacted because they were not at home for initial contact to be made.In the case of apartment residents, 199 surveys were distributed. However, insome cases, where the surveys were left with the manager or in the front entrance tothe apartment, it was not possible to determine how many were received by potentialparticipants.-68-There were six apartment buildings where managers accompanied theresearcher door to door or provided information which allowed complete calculation ofrates of contact and return rates. In those six buildings, 61 surveys were distributed tosenior residents. Fifty-one surveys were returned completed, for a return rate of83.61%. One resident returned a questionnaire which was not usable due to too fewquestions being completed (had been completed, then most erased), and fifteenresidents refused to participate and were not provided with questionnaires. Twoindividuals had already received a survey as participants of Kerrisdale Senior Centreand were not provided with a second one. In addition, with the information providedby managers it was possible to determine the number of eligible residents who couldnot be contacted because they were out of the city, not home or the manager did notfeel comfortable bothering them. Thirty-four individuals were not contacted, for a totalof 112 individuals who were eligible to participate in these six apartment buildings.Therefore, 69.64% (61 provided with surveys + 15 refusals + 2 contacted at KerrisdaleSenior Centre equals 78/112) of all eligible residents were contacted and asked toparticipate in the study, and usable data was received from 47.32% (51 surveys + 2 atKerrisdale Senior Centre equals 53/112) of all persons eligible to be participants ofthis study from these six buildings.Of the remaining 138 surveys distributed among apartment buildings, 73 wereleft at the front entrances of two large apartments, and 53 were left with apartmentmanagers to distribute to elderly residents. It is, therefore, not possible to calculate thecontact rate or the return rate, as it is not known how many eligible residents receiveda survey. Of the surveys left at the front entrances, only nine were returned, one fromone building and eight from the other. Of those which were left with managers, elevenwere returned. There is some question about whether one of the managers, who wasprovided with twenty surveys, actually distributed them (she agreed to do so but wasless than enthusiastic). Three telephone contacts with the manager after leaving the-69-surveys determined that she had not distributed them until at least two weeks afterreceiving them, and none were returned from this building. Although it was veryquickly apparent that return rates would be low from buildings where residents couldnot be contacted directly, the judgment was made that it was important to try, in casethere were differences between residents of buildings where the manager was moreor less approachable and involved with the residents. Table 2 portrays the return rateinformation from all sources.TABLE 2:^RETURN RATES FOR STUDYSOURCESURVEYSPROVIDEDSURVEYSRETURNEDRETURNRATE (%) REFUSALSNOTCONTACTEDA) KerrisdaleSenior Centre 102 58 57 9B) Single^Family^Dwellings 56 51 91 8C) Apartments(Direct^Contact) 61 51 84 15 34D) Apartments(No Direct Contact) 138 20TOTALS 357 180 A-C:^77*A-D:^50* No true return rate could be calculated for apartments where there was no direct contact (D), asit is not known how many eligible residents actually received a survey. Therefore, total returnrate is shown in two ways: for the three sources for which a true return rate could be calculated(A-C) and for all surveys distributed and returned (A-D).4.1 DETERMINATION OF PARTICIPANTS AND NONPARTICIPANTSParticipation in the Health Drop-In was to be limited to study respondents whoanswered that they attended the Health Drop-In, and had done so more than once.However, participants of the program did not necessarily consider the programs theywere attending to be part of the Health Drop-In. This was most common amongexercise program participants, perhaps because the exercise sessions are spread outover four days rather than all taking place on Wednesday morning.-70-Thus, twelve exercise participants completed surveys as though they werenonparticipants of the Health Drop-In. These individuals could be identified by: (1) thecoding on survey forms (all surveys distributed to participants at Kerrisdale SeniorCentre were coded with a number starting with "K"); (2) comments in various locationson the surveys in which the respondents stated they participated in an exerciseprogram, and (3) checking off of the "exercise program" in the question that requestedinformation about which components participants attended. All 12 could be identifiedby the coding, 5 had checked off the "program component" and 5 included identifyingcomments.Dropping these individuals was considered. However, it was thought that theremight be some differences between individuals who considered themselvesparticipants of a health program versus those who attended the same program but didnot consider themselves participants. Therefore, it was decided to compare results ofthese twelve with the rest of the participants, in addition to comparing true participantsand nonparticipants. Unfortunately, in surveys filled out by respondents whoconsidered themselves nonparticipants, some potentially interesting data areunavailable, especially regarding the reasons for participating in a health program.Comparisons could be made, however, on demographic data, results of the ISAIscales and activity patterns.In addition to the 12 participants who described themselves as nonparticipants,8 respondents gave incomplete or contradictory information regarding theirattendance. One individual stated that he did not attend the Health Drop-In andprovided no response to length of attendance, but completed the survey as aparticipant of the Blood Pressure Monitoring component. A second individualprovided no response to the question about current attendance, but completed thequestionnaire as a weekly attender of the Neck and Shoulder Massage component.Four individuals provided no response to length of attendance, but each were-71-attenders of two of the three components. Two individuals checked that they hadattended the program once, but filled the rest of the survey out as regular weeklyattenders of more than one component. These 8 individuals were included asparticipants, rather than being dropped.Once the respondents had been stratified according to participation/non-participation, there were 63 participants, 12 of whom described themselves asnonparticipants. The majority were from the Kerrisdale Senior Centre sample; 2 camefrom the single family dwelling sample and 3 were from the apartment residentsam pie.There were 117 nonparticipant questionnaires (one questionnaire was returnedwith less than 50% of questions completed and was, therefore, dropped). Forty-ninenonparticipant respondents were residents of single family dwellings and 68 residedin apartments.4.2 DEMOGRAPHIC CHARACTERISTICSTable 3 summarises demographic characteristics with respect to place ofresidence (which represents a possible confounding variable due to samplingtechnique). Table 4 presents demographic data for participant and nonparticipantgroups. Table 5 provides a profile of the demographic characteristics of participants inthe three Health Drop-In program components, while Table 6 provides the sameinformation for participants according to whether they classified themselves asparticipants (P1) or nonparticipants (P2).TABLE 3:^DEMOGRAPHIC CHARACTERISTICS BY PLACE OF RESIDENCE*APARTMENTSINGLE FAMILYDWELLING OTHER SIGNIFICANCEAge:^(mean years) 78.5 72.7 71.3 0.0004Gender: ^M F^ 2 2 882 7 390 3 0.006Education:^< HS HS > HS^ 28 47 291 6 272120 1 NSLive:^Alone SpouseRelative/Friend Other82 2 2 5114 4 38121 0o 0.0001* Total N will vary due to missing responses in some categories.TABLE 4:DEMOGRAPHIC CHARACTERISTICS^BY^PARTICIPATION*PARTICIPANTS^INONPARTICIPANTS1^SIGNIFICANCEAge: (mean yrs.) 74.8 77.1 0.0293Gender: ^M F1350(20.6%) (79.4%)3780(31.6%)(68.4%)- NSEducation:^< HS HS > HS132918(21.7%)(48.3%)(30.0%)3345 33(29.7%)(40.5%)(29.7%) NSResidence: Apar. SFD Other42173(67.7%)(27.4%)(4.9%)6849(58.1%)0 (0%)(41.9%) ^0.0138Live:^Alone Spouse Rel./Fr. Other3523 50(55.5°k)(36.5%)(8.0%)(0.0%)644382(54.7%)(36.8%)(6.8%)(1.8%)* Total N will vary due to missing responses in some categories-73-TABLE 5:^DEMOGRAPHIC CHARACTERISTICS BY PROGRAM COMPONENT^PROGRAMB P M Sig. E Sig. N/SM Sig.Age:^(years) 74.7 NS 74.8 NS 75.51^ NS Gender:^M F1130 NS 35 0.0082 17 NSEducation:^< HS H^S > HS91614 NS10 178 NS8 11 4 NSResidence:^Apar. SFD Other27112 NS27 8 2 0.019317 5 1 NS*Live: ^Alone ^ ^Spouse Rel./Fr. Other2020 ^19 25 8401392 0^Total N will vary due to missing responses in some categories.*Contingency table analysis not done due to small numbers in many cells.TABLE 6:^DEMOGRAPHIC CHARACTERISTICS OF ATTENDERS WHODID AND DID NOT PERCEIVE THEMSELVES AS PARTICIPANTSP2: Responded asNonparticipants^(12)P1: Responded asParticipants^(51) SignificanceAge:^(years) 77 74.3 0.0475Gender:^M  11,1 ^ 1239 NSEducation:^< HS HS > HS165122313 NSResidence:^Apar. SFD Other____  ^4^08 34133 NSLive:^Alone Spouse Rel./Fr. Other ^9 ^ ^2  ^1 ^026214_0 NS4.2.1 Age and Place of ResidenceAge and place of residence were strongly associated (p = .0004, x2 = 32.276),with apartment dwellers being significantly older than residents of single familydwellings.4.2.2 Age and ParticipationAge showed a significant association with participation, using the MannWhitney U test. The average age of respondents was 76.3 years, with 2 individualsnot responding. Participants, with a mean age of 74.8 years (n = 62), weresignificantly younger than nonparticipants, who had a mean age of 77.1 years (n =116) at the p = .0293 level (Z = -2.179).Attenders who did not perceive themselves as participants (P2) were also olderthan those who answered the surveys as participants (P1). The mean age for the P2group (n=12) was 77.0 years, compared to 74.3 years for the P1 group (n=50). At a Z-score of -1.981, the p value was .0475.Age was not significantly associated with the various program components.The mean age for attenders of the Blood Pressure Monitoring component (BPM) was74.7 years, for the Exercise component (E) 74.8 years and for the Neck/ShoulderMassage (N/SM) 75.5 years.4.2.3 Gender and Place of ResidenceTwenty-two males and 88 females lived in apartments, 27 males and 39females resided in single family dwellings and 3 females checked the "other" category.Respondents in the "other" category specified townhouse (2) and condo (1). Gendershowed a significant association with place of residence (x2 = 10.221, p = .006), withfemales more likely to reside in apartments, while males had a greater chance ofresiding in a single family dwelling.-75-4.2.4 Gender and ParticipationFifty respondents were male, representing 27.8% of the sample, and 130(72.2%) were female. Overall, gender was not significantly associated withparticipation, with 26% of the males and 38.5% of the females indicating they wereparticipants. Females were significantly more likely to attend the E component (2males and 35 females reported attending, x 2 = 6.992, p = .0082). Gender approachedsignificance for the BPM component (p = .0522), with more male participants thanwould be expected. Gender was a nonsignificant factor for attendance in the N/SMcomponent.There was no significant difference between P1 (answered as participants) andP2 (answered as nonparticipants, but attended E component) with regards to gender.There were 12 males and 39 females in P1, while P2 consisted of 1 male and 11fern ales.4.2.5 Education and Place of ResidenceOf the 171 respondents who provided information about educational status, 46(26.9%) had less that high school, 74 (43.3%) had completed high school and 51(29.8%) had at least some education above high school. Level of education was notsignificantly associated with place of residence.4.2.6 Education and ParticipationThere was no significant association between education and participation in theHealth Drop-In. Of the respondents who had less than high school education, 28.3%were participants. Of those who had high school completion or equivalent, 39.2%were participants, while 35.3% of those with education past the high school levelindicated that they were participants. There were also no significant associationsbetween education and attendance at various components of the program, or betweenP1 and P2.-76-4.2.7 Living Arrangement and Place of ResidenceThe majority of the sample (99, or 55%) lived alone, with 66 (36.7%) living witha spouse, 13 (7.2%) living with a relative or friend and 2 with some other arrangement.There was a strong association between living arrangement and place of residence(x2 = 47.979, p = .0001), with respondents who lived alone more likely to live inapartment buildings or "other", while those who lived with spouses or relatives/friendsmore likely to reside in a single family dwelling than would be expected.4.2.8 Living Arrangement and ParticipationLiving arrangement was not significantly associated with participation in theHealth Drop-In. Of the respondents who lived alone, 35.4% were participants, while34.8% of respondents living with a spouse were participants. Of the 13 respondentswho lived with a relative or friend, 38.5% were participants, and neither of the 2respondents with a different arrangement indicated that they were participants.It was not possible to determine whether respondents with some livingarrangements were more or less likely to attend the various program components dueto small numbers in some contingency table cells. Likewise, living arrangement couldnot be compared for the P1 and P2 groups.4.3 KNOWLEDGE OF THE HEALTH DROP-INOf 179 respondents who provided information about their knowledge of theHealth Drop-In, 125 (69.8%) were aware of the program and 54 (30.2%) did not knowabout it. Half (50.4%) of the 125 respondents who knew about the Health Drop-In,therefore, were participants.4.3.1 Knowledge and Previous Attendance Among NonparticipantsAmong the 116 nonparticipants who provided this information, 55.9% (38) ofapartment residents were aware of the program compared to 56.3% (27) of the singlefamily resident respondents. Twenty-one of the male (55.7%) and 44 of the female(56.8%) nonparticipants knew about it. Those who lived alone were more likely to be-77-aware of it (41 respondents, or 64.1%) than those who lived with a spouse (19respondents, or 45.2%) or those who lived with a relative/friend (3 respondents or37.5%). However, there were no significant associations between knowledge andgender, place of residence or living arrangement.Among nonparticipants, 14 (12.1%) stated that they had previously attended theHealth Drop-In. Two of these past participants were male, 12 were female. Ten livedalone, 2 with a spouse, 1 with a relative/friend, with 1 choosing "other". Ten of thesepast participants were apartment residents, the other 4 lived in single family dwellings.None of these variables was significantly associated with being a past participant.4.3.2 Source of KnowledgeOf 55 participants who provided a source of knowledge about the Health Drop-In, 20 (36.4%) had read about it, 18 (32.7%) had heard about it from a friend, 3 (5.5%)had been told by their doctor and 1 had been informed by a public health nurse. Tworespondents chose two sources: one had read about it and heard from a friend, andthe other had read about it and heard about it from their doctor. Thirteen (23.6%) hadheard about it from other sources; 8 of these specified membership in the senior centreor long term involvement in the community centre. One specified through volunteeringat West Main Health Unit, 1 through living in the area, 1 through the newspaper, 1found out about it on the bus, and 1 did not specify the source.Sixty-five of the nonparticipants, or 57%, responded that they knew about theHealth Drop-In. Although not requested, 27 of these individuals provided a source ofknowledge: nearly half (13) had read about it, 4 had found out from a friend, 1 wasinformed by a Public Health Nurse, 2 by a doctor, and 7 checked "other".Therefore, 33 of the the 82 respondents who knew about the program, andprovided the source of their knowledge, had read about it. Twenty-two had found outfrom a friend, 2 heard about it from a public health nurse, 5 were told by their doctorand 20 indicated an alternate source.-78-Differences in source of information between participants and nonparticipantswere not significant, but it is interesting that nonparticipants were more likely to haveread about it (13 of 27 respondents, or 48%) than participants (20 or 55 respondents,or 36.4%). Participants, on the other hand, had more frequently found out about it froma friend (18 participants, or 32.7% compared to 4 nonparticipants, or 14.8%) or from anurse or doctor (4 participants, or 7.3% compared to 3 nonparticipants, or 1.1%).Gender, educational level, living arrangement, place of residence and age werenot significantly associated with source of knowledge.4.4 LENGTH OF PARTICIPATIONOf 72 respondents who had attended the Health Drop-In (participants and pastparticipants) and who provided length of participation information, 10 (16.1%) hadattended once, 8 (12.9%) had attended more that once but less than 3 months, 7(11.3%) had participated for at least 3 months but less than one year and the majority(37, or 59.7%) had participated for more than one year. Gender, level of education,living arrangement and place of residence were not significantly different forrespondents who had attended for different lengths of time.Of the participants, 3 responded that they had attended once (but filled out thesurveys as regular attenders as previously discussed in Section 4.1), 3 had attendedmore than once but less than 3 months, 6 had attended for 3 months to one year and36 had attended for more than one year. Once again, gender, place of residence andliving arrangements were all non-significant with regard to length of attendance.There was no significant difference between lengths of attendance for respondentswho reported attending the three program components (BPM, E or N/SM).Past participants were more likely to have attended once (7, or 50%) or morethan once but less than 3 months (5, or 35.7%). One past participant had attended for3 months to one year, and one had attended for more than one year. Length ofattendance for past participants was not significantly associated with gender, living-79-arrangement or place of residence.Of 5 past participants who provided information about program componentsattended, 3 had attended the BPM component, 2 the E component and 1 the N/SMcomponent.Information about knowledge of and past attendance patterns amongnonparticipants is summarised in Table 7.TABLE 7:^KNOWLEDGE OF HEALTH DROP-IN AND PASTATTENDANCE AMONG NONPARTICIPANTSA. KNOWLEDGE (n = 116) B.^PAST PARTICIPATIONYES^65^(56.0%) ND 51^(44.0%) YES^14^(12.1%)By Place of Residence:Apartment^38^(55.9%)* SFD  ^27^(66.3%)* By Place of Residence:Apartment SFD 104By Gender:Male^21^(55.7%)* Female^44^(56.8%)* By Gender:Male Female212* Percentages refer to the proportionof all nonparticipant respondentswith that characteristic.Program Components AttendedBPM ^3E   2N /S M^ 14.5 PROGRAM COMPONENTS4.5.1 Past and Present ParticipantsThere were 40 respondents who indicated present or past participation in theBlood Pressure Monitoring (BPM) component, with a mean age of 74.7 years. Thirty-seven respondents, with an average age of 74.8 years, reported past or presentparticipation in the Exercise (E) component, and 23 respondents indicated past orpresent participation in the Neck and Shoulder Massage (N/SM) component, with amean age of 75.5 years.Place of residence was significantly associated with participation in the E-80-component, with 27 of the 109 apartment dwellers reporting past or presentinvolvement in E, compared to 8 of 66 single family residence respondents and 2 or 3"other" respondents (x 2 = 7.896, p = .0193). However, this variable was non-significant for BPM and N/SM components. Twenty-seven apartment dwellers, 11single family dwelling residents and 2 of those with other arrangements reported pastor present involvement in the BPM component. Seventeen of the apartment dwellers,5 single family residents and 1 with another arrangement reported past or presentN/SM participation.Age was not significantly associated with present or past participation in any ofthe three components.Of 49 male respondents, 11 indicated present or past involvement in the BPMcomponent, compared to 30 of 129 females. In the N/SM component, 7 males and 17females reported participation. Gender was not associated with participation in eitherof these components. However, gender was significantly associated with participationin the E component, with 2 of 49 male respondents, and 35 of 130 femalerespondents, indicating present or past involvement (x 2 with continuity correction =9.973, p = .0016).Educational attainment was not associated with participation in any of thecomponents. Of 46 individuals who had less than high school education, 9 attended(or had attended in the past) the BPM, 10 the E and 8 the N/SM components. Therewere 73 respondents with the equivalent of high school education; 16 indicatedpresent or past participation in the BPM, 17 the E and 11 the N/SM components. Of 51respondents who had received education past the high school level, 14 attended theBPM, 8 the E and 4 the N/SM components.Living arrangement was also a non-significant variable with respect toparticipation in any of the components. Of 97 respondents who reported living alone,-81-20 had past or present involvement in the BPM, 25 the E and 13 the N/SMcomponents. Sixty-six individuals lived with a spouse, and 19 of these were, or hadbeen participants in the BPM, 8 in the E and 9 in the N/SM. Two of the 13 respondentswho lived with a relative or friend reported present or past participation in the BPM, 4in E and 2 in N/SM. None of the respondents who reported another livingarrangement were involved in any of the program components.4.5.2 Current ParticipantsArea of participation was given, or could be determined, for 62 of theparticipants. Thirty-nine individuals responded that they were participants of the BPMcomponent. There were 35 respondents who stated that they were participants of theE program (including those who perceived themselves as nonparticipants of theHealth Drop-In) and 23 stated that they participated in the N/SM. Sixty percent (37respondents) attended only one of the three components. 4 (6.5%) attended BPM andE, 10 (16.1%) attended BPM and N/SM, 2 (3.2%) attended E and N/SM, and 9 (14.5%)attended all 3 components. Table 8 provides a summary or patterns of participation.TABLE 8:^ATTENDANCE PATTERNS OF PROGRAM PARTICIPANTSA. LENGTH OF PARTICIPATIONMORE THAN 1 YEAR 36^(75%)D. FREQUENCY OF ATTENDANCEWEEKLY 34^(66.7%)3 MOS. TO 1 YEAR 6^(12.5%) REGULARLY 9^(17.6%)> ONCE AND <3 MOS. 6^(12.5%)1 5IRREGULARLY 8^(15.7%)NO RESPONSE NO RESPONSE 1 2B. PROGRAM COMPONENTSBLOOD PRES. MONITORING 39E. SOURCE OF KNOWLEDGEREAD ABOUT IT 20^(36.4%)18^(32.7%)EXERCISE 3 5 HEARD FROM A FRIENDNECK/SHOULDER MASS. 23 DOCTOR 3 (5.5%i1NO RESPONSE 1 PUBLIC HEALTH NURSEOTHER 13^(23.6%)C. NUMBER OF COMPONENTSONE 37^(59.7%)NO RESPONSE 8TWO 16^(25.8%)THREE 9^(14.5%)-82-Of the individuals who had been participating for more than one year in theHealth Drop-In (n=36), 27 attended the BPM, 16 the E and 17 the N/SM components.Of those who had been attending for less than one year (n=12), 6 were BPMparticipants, 7 E participants and 3 N/SM attenders. Length of attendance was notassociated with participation in any of the components.4.6 FREQUENCY OF ATTENDANCEThe majority of participants who gave frequency of attendance information (34respondents, 66.7%) visited the Health Drop-In weekly. Nine respondents (17.6%)stated that they were regular attenders and 8 (15.7%) indicated that they wereirregular attenders. Twelve individuals did not provide this information, including 9 ofthe exercise attenders who perceived themselves as nonparticipants of the HealthDrop-In.By place of residence, 25 of the 34 weekly attenders lived in apartments, 5 insingle family dwellings and 3 in another type of abode. Three of the 9 regularattenders lived in apartments, the other 6 in single family dwellings. Five of the eightirregular attenders were apartment residents, the other 3 lived in single familydwellings. Single family dwelling residents are more likely to attend regularly whileapartment dwellers are more likely to be weekly attenders. This tendency wasmarginally significant at the x2 = 10.501 level, with p = .0328.Gender, educational attainment and age showed no significant associationswith frequency of attendance. There was, however, a perceptible tendency for lessregular participation with increased age, with a mean age for weekly attenders of 72.8years, compared to 75.0 years for regular attenders and 80.7 years for irregularattenders.Living arrangement was significantly associated with frequency of attendance.Of the 34 weekly attenders, 20 lived alone, 11 with a spouse and 3 with a relative orfriend. Of 9 regular attenders, 1 each lived alone and with a relative/friend, while 7-83-lived with a spouse. Seven of the 8 irregular attenders lived alone, the other lived witha spouse. There was some tendency toward either weekly or irregular participationamong those who lived alone, and regular participation for those who resided with aspouse (x2 = 11.081, p = .0257).There were no significant associations between frequency of attendance andany of the program components. Among the 34 weekly attenders, there were 22 whoparticipated in the BPM, 20 of the E participants and 14 of the N/SM attenders.Included in the 9 regular attender respondents were 9 BPM, 2 E and 5 N/SMparticipants. The 8 irregular attenders included 6 BPM, 2 E and 3 N/SM participants.4.7 ISAI SCORESScores on the ISAI scales were first analysed to determine whether there weredifferences between respondents based upon place of residence. Comparisons werethen made between participants and nonparticipants, and between attenders of thethree program components. Comparisons between participants who responded asnonparticipants and those who answered as participants are also provided.Nonparametric tests (Mann Whitney U for two groups and Kruskall-Wallis for morethan two groups) were utilised to determine statistical differences between groups asmean values were highly skewed to the right (large numbers of high scores andrelatively few low scores). Table 9 summarises information about mean ISAI scoresand place of residence. Differences between participants and nonparticipants aresummarised in Table 10. Profiles of ISAI means for participants of the various programcomponents are included in Table 11, while Table 12 shows ISAI scale scores forparticipants based upon whether they responded as participants (P1) or asnonparticipants (P2).TABLE 9:^MEAN ISAI SCORES BY PLACE OF RESIDENCEEconomicResourcesEmotionalBalancePhysicalHealthTrustingOthersMobility27.9107CognitiveStatus_24.4 106SocialSupport29.6 106APARTMENTS (N)29.3 ^10725.5 ^10422.4 ^10630.7 ^103SFD (N)29.4 64 ^ 25.8 6522.4 6530.4 6527.7 6425.2 6529.8 64OTHER (N)31 322 323.3 330.7329.7 323.7329.3 3TABLE 10:^MEAN ISAI SCORES FOR PARTICIPANTS AND NONPARTICIPANTSSCALETOTAL SAMPLE PARTICIPANTS NONPARTICIPANTSSignificanceRange Mean N Mean^N Mean^NEcon. Res. 12-32 29.4 175 30.1^61 29^114 NSEmot. Bal. 9-32 25.5 173 24.4^61 26.1^112 NSPh. Health 8-32^22.4 175 24.4^61^21.3^11430.8^112p = .0019NSTrust.^Oth 16-32;^30.5 172 30.1^60Mobility 11-32'^27.8 175 28 ,^ 61 27.2 ,^ 114 p = .0318Cog. Stat. 8-32^24.6 175 24.2 61 24.9 114 NSSoc. Sup. 15-32 ,^ 29.7 174 29.3 60 29.9 114 NSTABLE 11: ISAI SCORES FOR PROGRAM COMPONENT ATTENDERSPROGRAMB P M E N /S MN 38 35 22SCALE: Mean^Sig. 1Mean^1^Sig., Mean;^Sig.Econ. Res. 30.11 29.71 29.91Emot. Bal. 24.2; 24.3; 23.8;Ph. Health 23.41^0.0292 261^0.0088 23.11Trust.^Oth 30.1; 30.21 30.4;Mobility 28.8; 29.3! 27.5 0.0318Cog. Stat. 24.41 24.11 23.6Soc. Sup. 29.4; 28.911-85-TABLE 12: MEAN ISAI SCORES OF ATTENDERS WHO DID ANDDID NOT PERCEIVE THEMSELVES AS PARTICIPANTSP2: Responded asNonparticipants^(12)P1: Responded asParticipants^(51)Economic Res. 29.6 30.2Emotional Bal. 24.4 24.4Phys. Health 26.3 24Trusting^Oth. 29.3 30.3Mobility 29.7 28.8Cognitive^Stat. 23.8 24.4Social Support 27.5 Economic Resources (ER)The standard mean of 50 is achieved at a raw score of 27 on the ER scale(Morris et al., 1991). Usable scores were obtained from 175 respondents in this study.Scores ranged from 12 to 32, with a mean score of 29.4. This translates to astandardised score of approximately 54, indicating that the Kerrisdale seniors in thissample believe that they are relatively well off financially.Because apartment residents and single family dwelling residents wereindependent samples, scores were compared by place of residence, using theKruskal-Wallis test for 3 groups, to determine whether financial status was associatedwith place of residence. The differences between mean ER scores for single familydwellers (29.4, n = 64), apartment residents (29.3, n = 107) and other (31.0, n = 3)were not significant.Among participants, there were 61 usable scores, with a mean raw score of30.1. Of 114 scores from nonparticipants, the mean raw score was 29.0. Thisdifference was not significant. There was also no significant difference between thosewho perceived themselves as participants (raw score of 30.2, n = 49) and participantswho responded as nonparticipants (raw score of 29.6, n = 12). No significantdifferences in ER were noted between attenders of the three program components.-86-The raw score for BPM participants was 30.1, for E the mean raw score was 29.7 andfor N/SM it was Emotional Balance (EB)A raw score of 25 on the EB scale translates to the standard mean score of 50(Morris et al., 1991). This sample (n = 173) had a range of scores from 9 to 32, with amean of 25.5. This converts to a standardised score of 52.Once again, mean scores among apartment residents (25.5, n = 105), singlefamily residents (25.8, n = 65) and others (22.0, n = 3) were not significantly different,using the Kruskal- Wallis test.The mean EB score for the participants was 24.4 (n = 61), and fornonparticipants 26.1 (n = 112). EB was not a significant variable with respect toparticipation, using Mann Whitney U test. The means for the P1 and the P2 groupswere equivalent to each other and to the participant mean, at 24.4. EB was not asignificant factor with regards to participation in the different program components: theBPM participants had a mean of 24.2, the E participants 24.3 and the N/SMparticipants a mean of Physical Health (PH)On the PH scale, a raw score of 21 is equivalent to the standard mean of 50(Morris et al., 1991). This sample's mean score (n = 175) was 22.4, which converts toa standardised score of 52. Scores ranged from 8 to 32.Single family dwelling residents (n = 65) and apartment dwellers (n = 106) bothhad a mean of 22.4 on the PH scale, while those with another type of residents (n = 3)had a mean score of 23.3. This difference was not significant.PH mean scores were higher than the standard mean of 50 for both participants(24.4, n = 61) and nonparticipants (21.3, n = 114), but differences in PH scores weresignificantly associated with participation (Z corrected for ties = -3.112, p = .0019).PH was a significant factor in participation for two of the program components.-87-The mean score of BPM participants was 23.4, which was significantly lower than therest of the participants (Z corrected for ties = -2.181, p = .0292). Alternatively, the meanscore for E participants on the PH scale was 26.0, which was significantly higher thanthe rest of the participant group (Z corrected for ties = -2.618, p = .0088). PH scoreswere not significantly associated with N/SM participation. The mean for this group was23.1.The PH mean score for the P2 group (n = 12) was in line with the mean scorefor all E participants, at 26.3. This was not significantly different from the P1 group (n =51), at Trusting Others (TO)A raw score of 30 on the TO scale converts to the standard mean of 50 (Morriset al., 1991). The mean for the 172 valid Trusting Others (TO) scale scores in this studywas 30.5 (or a standardised score of 52.5). The range of scores was from 16 to 32.Those who lived in apartments (n = 103) and those who chose "other" as aplace of residence (n = 3) had a slightly higher mean (30.7) than those who lived insingle family residences (30.4, n = 65). This difference was not significant.Scores on the TO scales were not significantly different for participants (30.1, n= 60) and nonparticipants (30.8, n = 112). Likewise, level of trust was not a significantfactor in participation in the various program components. The mean score for BPMparticipants was 30.1, for E participants 30.2 and for N/SM participants 30.4. The P2group, which responded as nonparticipants, had a mean score of 29.3, compared tothe P1 mean score of 30.3. This was also non-significant.4.7.5 Mobility (MO)To achieve a standardised score equal to the standard mean of 50 on theMobility (MO) scale, a raw score of 27 must be attained (Morris et al., 1991). For the175 scores computed on the MO scale in this study, ranging from 11 to 32, the meanwas 27.8 (standardised score of 52).-88-Apartment residents (n = 107) had a mean score of 27.9, single family residents(n = 64) a mean of 27.7 and others (n = 3) a mean of 29.7. This was not a significantdifference.Greater mobility was marginally associated, statistically, with participation(score of 28, n = 61) in the Health Drop-In. The actual clinical difference, however,was minimal, with nonparticipants having a lower MO mean of 27.2 (n = 114). TheMann Whitney U Z value corrected for ties was -3.112 (p = .0316).There were no significant differences in MO score means between participantsof the BPM component (28.8) and other component participants, or for E participants(29.3). However, the N/SM participants, with a mean score of 27.5, were marginallysignificantly lower than other participants (Z corrected for ties = -2.147, p = .0318).Degree of mobility did not show up as significantly associated with response as aparticipant or nonparticipant among the participants; P2 had a mean score of 29.7 andP1 a mean score of Cognitive Status (CS)A raw score of 23 on the CS scale corresponds to the standard mean of 50(Morris et al., 1991). In this study, on the CS scale (n = 175) the range of scores wasfrom 8 to 32, with a mean of 24.6 (standardised score of approximately 53).For the 106 apartment residents, the mean score was 24.4. The 65 singlefamily residents had a mean of 25.2 and the rest (n = 3) had a mean of 23.7. This wasnot significant.Cognitive status was not significantly associated with participation in the HealthDrop-In. The mean for participants (n = 61) was 24.2 and for nonparticipants (n = 114)24.9. This variable was also not significantly associated with participation among thethree components, with BPM participants having a mean of 24.4, E participants amean of 24.1 and N/SM participants a mean of 23.6. Likewise, there was nosignificant difference between the P1 mean of 24.4 and the P2 mean of 23.8.-89-4.7.7 Social Support (SS)A raw score of 29 on the Social Support (SS) scale is equivalent to thestandard mean score of 50 (Morris et al., 1991). The individual scores on the SS scalewere highly skewed, with a range from 15 to 32 with a mean of 29.7, converting to astandardised score of 52.SS scores were not associated with place of residence. Apartment residents (n= 106) had a mean score of 29.6, single family residents (n = 64) a mean of 29.8, andthe others (n = 3) a mean of 29.3.Level of social support was not significantly associated with participation in theHealth Drop-In. Participants (n = 60) had a mean score of 29.3, compared tononparticipants' (n = 114) mean score of 29.9. Social support was also not asignificant factor in participation among the various components, with BPM participantshaving a mean of 29.4, E participants a mean of 28.9 and N/SM participants a mean of28.6. Social support did not show up as significantly associated with response as aparticipant or nonparticipant among the participant group, with P2 (n = 12) having aSS mean of 27.5 compared to the P1 (n = 51) mean of COMMUNITY INVOLVEMENTInvolvement in community activities was requested in terms of whetherrespondents attended each activity or led/volunteered for it. Due to small numbers,especially in the leader/volunteer cells, the attend and leader/volunteer cells werecombined to determine total involvement for contingency table analysis. Table 13summarises community involvement in terms of demographic variables. Table 14provides a summary of community involvement for participants and nonparticipants.TABLE 13:^PERCENTAGE OF COMMUNITY INVOLVEMENT AMONG RESPONDENTSBY DEMOGRAPHIC CHARACTERISTICSTotal^NResidence , (%) Apart.^SFD  10ther 110 ^6 6^1^3Gender (%) M^L^F 50^1 130Ag_emean yearsLiving  Arrange.^(%) Alone ipouse1Rel/Fr99^I^66^;^1 3Education ,(%)< HS ^HS  1  >HS 46 1^74^51Kerrisdale^S.C. Sig.18.2^12.1: ^66.7 0.0461_ 20.8 0.0375.2,19± 1L^15.21 ^7.7 19.6 ! ^13.51  19.6 ,1Church Sig.33.61^28.8 ,133.3_ 26_, - 33.8 77.6 0.0222.2_;^33.3 ,,130.8 43.5^271 33.31,,Church Grps. 21.81^15.21^33.3 121^22.3 76.3 17.21^24.2;^15.4 28.3^17.6;^17.6Cultural Orbs_Sig.^10.91^9.11r^1^,21^13.1 72.70.0411.1 1^10.6:^0,-16.5^9.51^13.7,- -1,r ,Symphony, etc Sig.35.5_ 27.3 33.3 201^36.9 0.0574.9130.31 ^36.4130.8 ,'t ^,^28 ^271^47.10.05, ,Recreat.^Act. Sig.21.81+ ^30.31^0 7I,^,' 124: ^24.6 73.7 0.0121.2!^33.3 4--I,,7.7 15.2;^24.3;^37.30.04I ,^,Other Sen. Cen. 12.71^4.5;^0 4^11.5 75.9,12.11^4.5;^15.4 4.3;^10.8;^13.7, ,,,,Commun. Cen.,7.21^16.71 8^11.5 76.7 9;^12.1;^15.4 131^8.1;^13.71 , ,Others 20.9;^28.8;^33.3 60;^10 75.9 26.3;^24.21^7.7 19.6;^31.1;^19.6TABLE 14:^COMMUNITY ACTIVITY PATTERNS BY PARTICIPATIONParticipantsn = 63Nonparticipantsn = 117SignificanceKerrisdale Senior Centre 16 (25.4%) 14 (12.0%) 0.036West Main Health Unit 4 (6.3%) 0 (0%)Church 19 (30.2%) 38 (32.5%)Church Groups, Organisations 15 (23.8%) 20 (17.1%)Cultural^Centres,^Organisat. 9 (14.3%) 9 (7.7%)Symphony, Opera, etc. 27 (42.9%) 31 (26.5%) 0.0251Recreational^Activities 21 (33.3%) 23 (19.7%)Other Senior Centres 10 (15.9%) 7 (6.0%)Community Centre 6 (9.5%) 13 (11.1%)Others 14 (22.2%) 29 (24.8%)4.8.1 Other Activities at Kerrisdale Senior CentreOf the 180 respondents, 30 (16.7%) reported that they were involved with otheractivities at Kerrisdale Senior Centre. There was a marginally significant difference inattendance by place of residence, with 20 of 110 apartment residents (18.2%), 8 of 66single family dwelling residents (12.1%) and 2 of 3 others reporting involvement (x2 =6.533, p = .0381). However, the largest difference was with those reporting a place ofresidence other than apartment or single family dwelling, where numbers were verylow, which may represent a confounder.Gender was also marginally associated with other Kerrisdale Senior Centreinvolvement, with 3 males (6%) compared to 27 females (20.8%) reportinginvolvement (x2 corrected for continuity = 4.658, p = .0309). Age was not significantlyassociated with other Kerrisdale Senior Centre involvement, with an average age ofrespondents reporting such involvement of 75.2 years. Nineteen of 99 (18.2%)respondents who lived alone were involved, compared to 10 of 66 (15.2%) ofrespondents who lived with a spouse and 1 of 13 (7.7%) who lived with a relative orfriend. Nine of 46 (19.6%) individuals with less than high school education wereinvolved, compared to 10 of 74 (13.5%) respondents who completed high school and10 of 51 (19.6%) with education beyond high school.Other involvement with Kerrisdale Senior Centre was weakly associated withparticipation in the Health Drop-In (x 2 with continuity correction = 4.396, p = .036).Sixteen of the 63 participants reported such involvement, which is 53.3% of allrespondents reporting other Kerrisdale Senior Centre involvement. There was nosignificant difference between those participants who answered as participants (P1)versus those who answered as nonparticipants (P2).4.8.2 West Main Health UnitFour respondents reported involvement with West Main Health Unit activities.-92-Two were apartment residents, 1 resided in a single family dwelling and 1 reportedanother arrangement. All 4 respondents who indicated involvement were female.Three had post secondary education, while the fourth had completed high school. Allactivity related to the West Main Health Unit was reported by participants.4.8.3 ChurchFifty-seven individuals (31.7%) reported involvement with church. Place ofresidence was not significantly associated with church involvement, with 37 apartmentresidents (33.6%), 19 single family dwelling residents (28.8%) and 1 individual fromanother place of residence (33.3%) reporting such involvement. Increased age was,however, associated with involvement. The average age of those reporting churchattendance/leading/volunteering was 77.6 years, which was significant at the p =.0216 level on the Mann Whitney U test (Z corrected for ties = -2.297).Gender was not significantly associated with such involvement, although agreater proportion of females (44, or 33.8%) than males (13, or 26%) were involved. Agreater proportion of respondents with less than high school (20, or 43.5%) wereinvolved with church compared to high school graduates (20, or 27%) or respondentswith education past the high school level (17, or 33.3%), but this difference was non-significant. Of the 99 respondents who lived alone, 22 (22.2%) reported churchinvolvement, compared to 22 of 66 (33.3%) who lived with spouses, 4 of 13 (30.8%)with a relative or friend and 2 of 2 who had some other living arrangement.Church involvement was not significantly associated with Health Drop-Inparticipation, with 19 of the respondents who were involved with church, or 33.3%,reporting participation in the Health Drop-In. There was no difference between the P1(14 of 51) and P2 (5 of 12) groups, either.4.8.4 Church Groups and OrganisationsThirty-five of the respondents (19.4%) indicated that they were involved withchurch groups or organisations. Place of residence was not significantly associated-93-with involvement: Twenty-four apartment dwellers (21.8%), 10 single family residents(15.2%) and 1 individual from another type of dwelling (33.3%) were included amongthese respondents. The average age of respondents indicating involvement was 76.3years. Although considerably more females were involved (29, or 33.2% were femalecompared to 6, or 12% male), this difference was not significant. Neither wereeducation or living arrangement significantly associated with involvement. Seventeen(17.2%) of those living alone, 16 (24.2%) of those living with a spouse and 2 (15.4%)respondents living with a relative or friend stated that they were involved with churchgroups. Thirteen (28.3%) respondents with less than high school education, 13(17.6%) with high school and 9 (17.6%) with post secondary education were involved.Although more participants reported involvement (15, or 42.9% of allrespondents with church involvement) the difference was not significant. Likewise,there was no significant difference between P1 (13 respondents) and P2 (2respondents).4.8.5 Cultural Centres or OrganisationsEighteen respondents (10%) indicated involvement in cultural centres ororganisations. Place of residence was not significantly associated with involvement:10.9% (12) of apartment residents and 9.1% (6) of single family residents respondedthat they were involved with cultural organisations or centres.Younger age was weakly associated with involvement in cultural centres ororganisations. The mean age of such respondents was 72.7 years (Z corrected for ties= -2.002, p = .0453).Females were more likely to be involved than males, with 17 women (15%)indicating involvement compared to 1 of the 50 males. Those respondents witheducational status beyond high school (7, or 13.7%) were also represented in greaterproportion than those with high school (7, or 9.5%) or those with less than high school(3, or 6.5%). These tendencies did not show up as significant, however. Eleven of-94-respondents who were involved lived alone (11.1% of respondents living alone) and 7(10.6%) lived with a spouse.Involvement in a cultural centre or organisation was not significantly associatedwith participation in the Health Drop-In, although 9 of the respondents who reportedsuch activity, or 50%, were participants. Seven respondents in the P1 group and 2 inthe P2 group were involved.4.8.6 Symphony, Opera, etc.The most frequently checked activity by respondents was "symphony, opera, etc.".The involvement of the 58 respondents indicating attendance at these activitiesrepresents 32.2% of the sample. Place of residence was not significantly associatedwith involvement, with 39 (35.5%) apartment residents, 18 (27.3%) single familyresidents and 1 (33.3%) "other" respondent indicating involvement.The average age of attenders in this category was 74.9 years. Females weremarginally more likely to attend these activities (48, or 36.9%) than males (10, or20.0%; x2 with continuity correction = 3.992, p = .0457). Respondents with postsecondary education (24, or 47.1%) were also more likely to report attendance atthese functions than those with high school (20, or 27.0%) or less than high school(13, or 28.3%). This was marginally significant at the p = .0455 level (x 2 = 6.18).Living arrangement was not significantly associated with symphony, opera, etc.attendance. 30.3% of respondents who lived alone (30 respondents), 36.4% of thoseliving with a spouse (24 respondents) and 30.8% of those living with a friend orrelative (4) reported attendance or leading/volunteering.Symphony, opera, etc. involvement was positively associated with Health Drop-In participation (x2 with continuity correction = 4.298, p = .0251). Twenty-seven of therespondents indicating involvement, or 46.5%, were participants. This variable wasnot significantly associated with P1 (21 of 51 respondents) and P2 (6 of 12).-95-4.8.7 Recreational ActivitiesThere were 44 respondents (24.4%) who indicated involvement in recreationalactivities. Place of residence was not associated with involvement in recreationalactivities, with 24 respondents from apartments (21.8%) and 20 single family dwellingresidents (30.3%) responding positively.However, younger respondents were significantly more likely to report suchinvolvement, at a mean age of 73.7 years (Z corrected for ties = -2.652, p = .008).Males and females showed equal involvement, with 12 of the males (24.0%) and 32of the females (24.6%) reporting involvement. Living arrangement was also a non-significant variable, with 21 (21.2%) of the respondents who lived alone, 22 (33.3%) ofthose living with a spouse, and 1 of those living with a friend or relative indicatingrecreational involvement.Higher educational status was significantly associated with involvement inrecreational activity (x2 = 6.281, p = .0433). Nineteen of the 51 respondents with morethan high school education (37.3%), 18 of 74 with high school education (24.3%) and7 of 46 with less than high school education (15.2%) responded that they wereinvolved in recreational activities.Although 21 of the respondents indicating recreational activity involvement, or47.7%, were Health Drop-In participants, this difference did not show up as significant.One third of each of the P1 (17 respondents) and P2 (4 respondents) reportedinvolvement in recreational activities.4.8.8 Other Senior CentresA total of 17 respondents (9.4%) reported involvement of some kind in othersenior centres. Although a greater proportion of apartment dwellers (12.7%, or 14respondents) indicated involvement than single family residents (4.5%, or 3respondents) or others (no respondents), this difference was not significant.The mean age for respondents involved in other senior centres was 75.9 years.-96-Four percent of the males in this sample (2 respondents) and 11.5 % of the females(15 respondents) indicated involvement. Living arrangement and education were notsignificantly associated with attendance at other senior centres. Twelve (12.1%) ofthose who lived alone, 3 (4.5%) of those living with a spouse, and 2 (15.4%) of thoseliving with a relative or friend indicated involvement. Two respondents with less thanhigh school education (4.3%), 8 who finished high school (10.8%) and 7 with postsecondary education (13.7%) were involved.Although respondents who were involved with other senior centres were morelikely to be Health Drop-In participants, the difference did not show up as significant.Ten of the respondents who attended other senior centres, or 58.8%, wereparticipants. There was no significant difference in proportion of involvement betweenP1 (7 respondents) and P2 (3 respondents).4.8.9 Community CentreThe proportion of respondents who reported attending the community centrewas 10.6% (19 respondents). Single family residents (11, or 16.7%) were more likelythan apartment dwellers (8, or 7.3%) or others (0) to report attendance, but this wasnot significant.Fifteen females (11.5%) and 4 males (8.0%) attended the community centre.These respondents had a mean age of 76.7 years. Nine of them lived alone (9.1% ofall respondents who lived alone), 8 were single family residents (12.1% of singlefamily residents) and 2 lived with a relative or friend (15.4% of all who lived with afriend/relative). Of those with less than high school education, 6 (13%) attended thecommunity centre. Six of those with high school (8.1 %) and 7 with post secondaryeducation (13.7%) reported attending. None of these variables were significant.Attendance at the community centre was not associated with participation in theHealth Drop-In. Six of the respondents who attended the community centre, or 31.6%,were participants. Four respondents in the P1 category and 2 in the P2 category-97-reported community centre involvement.4.8.10 OtherForty-three of the respondents (23.9%) reported involvement in some othercommunity activities. A list of the activities specified is included in Appendix D.Place of residence was not significantly associated with involvement in othercommunity activities. Twenty-three of the apartment residents (20.9%), 19 of the singlefamily residents (28.8%) and 1 of the 3 respondents from another place of residenceindicated such involvement.The mean age of respondents with involvement in other community activitieswas 75.9 years. This was the only category where males were represented in greaterproportion than females: 13 males (26.0%) utilised the "other" category compared to30 (23.1 %) females. This difference, however, was non-significant. Neither wereeducational attainment or living arrangement significantly associated with involvementin other community activities. Ten (21.7%) of those respondents with less than highschool education, 23 (31.1%) of the high school graduates or equivalent and 9(17.6%) of respondents with more than high school education listed other communityactivities. Twenty-six (26.3%) respondents who lived alone, 16 (24.2%) of those livingwith a spouse and 1 of the 3 respondents with some other arrangement was involvedin some other community activity.Reported involvement in other community activities besides those listed was notsignificantly associated with participation. Fourteen (32.6%%) of those indicating suchinvolvement were participants. Eleven of the 51 P1 respondents and 3 of the 12 P2respondents used the "other" community activities category.4.9 PATTERNS OF ACTIVITYThe final question of the survey, which asked for hours per week spent on each-98-of seven activities, seemed to pose the most difficulty for respondents. Only 154 of the180 respondents provided usable data for this question. The majority of the rest didnot provide any answers. Several provided descriptive information (ie. the types ofhobbies, exercise or community activities engaged in, or comments such as "too little"or "too much" for sleep or household activities). One respondent furnished thecomment that she could not divide up her time in this way, and another commentedthat her time expenditures varied.A few individuals provided information in hours per day; these were multipliedby 7 to attain weekly expenditures. In some cases, respondents provided a range ofhours (eg. 2-4 hours), in which case the middle value was chosen (ie. 3 hours).As individual results did not approximate a normal distribution, analysis ofresults was carried out for categorical data with the Kruskal Wallis test (for more than 2groups) or the Mann Whitney U (for 2 groups). Age was compared using theSpearman Correlation Coefficient test. Mean weekly hours for the seven activities aresummarised by demographic grouping in Table 15, and differences betweenparticipants and nonparticipants are provided in Table 16. The mean reported weeklyhours of activity for attenders of the three program components are summarised inTable 17.TABLE 15:^MEAN WEEKLY HOURS OF ACTIVITY BY DEMOGRAPHIC GROUPHous. Act. Exercise Hobbies Visiting Comm. Act Gardening SleepResid: Apar (89) 12.5 4.4 5.8 5.6 1.2 1 50.5SFD (63) 13.8 5.4 9.1 5.9 2.7 6.3 49.6Other (3) 7.3 4 6.7 2.7 6.3 1 43.7Sig. 0.0354 0.0001Age: j with age* I with age* j with age*Sig. 0.0072 0.0004 0.0334Gender:^M (41) 8.9 5.7 8.3 5.1 0.7 4.1 50.7F^(113) 14.4 4.7 6.9 5.9 2.4 2.7 49.7Sig. 0.0415Live:^Alone (83) 12.7 4.4 5.9 6.1 1.9 2 48.9Spouse (59) 13.4 5.5 9 5.6 2 4.8 50.8Rel/Fr.^(11) 13 4.7 6.5 3.1 1.4 2.8 52.6Sig. 0.0043Education:^< HS 11.9 2.8 8.6 4.8 2.6 2.9 52.9HS 15.1 5.7 6.5 6.2 2 3.4 49.7>HS 1 1 5.5 6.8 6 1.4 3.1 47.7Sig. 0.0065* 1 denotes an inverse relationship between variablesTABLE 16:^MEAN REPORTED WEEKLY HOURS OF ACTIVITY BY PARTICIPATIONACTIVITYParticipants Nonparticipants Total Sample154 Significancen = 56 n = 98 n =Household^Activities 11.3 13.7 12.9Exercise 5.6 4.3 4.8Hobbies 5.2 8.2 7.1Visit Friends and Relatives 5.7 5.6 5.6Community^Activities 2.8 1.4 1.9 0.0058Gardening 2.1 3.7 3.1Sleep 43.1 49.4 47TABLE 17:^MEAN WEEKLY HOURS^OF ACTIVITY BY PROGRAM COMPONENT^n = 39B P M E N/SMSig. n = 35 Siq. n = 22 Sig.Household^Activity 12.7 10.5 13.6Exercise 5.3 5.9 4.2Hobbies 5.1 4.6 6.6Visiting Friends and Relatives 4.6 5.5 6.4Community^Activities 2.9 2.2 3.8Gardening 3 0.0275 1.5 2.6Sleep 49.8 46.3 50.6-100-4.9.1 Total Hours of ActivityPlace of residence was significantly associated with hours of activity reported,with single family residents (n = 63, mean of 94.8 hours) reporting more hours thanapartment dwellers (n = 89, mean of 89.0 hours) or others (n = 3, mean of 71.7 hours).The Kruskal Wallis H value corrected for ties was 12.125, p = .0023.Living arrangement was also significantly associated with hours of reportedactivity. As would be expected given the results based upon place of residence,respondents living with a spouse (n = 59) reported the most hours, at a mean of 94.3hours. Respondents living with a relative or friend reported the next highest hours at81.7 (n = 11), followed by respondents living alone (n = 83, mean of 77.3 hours). Onerespondent who fit none of these categories reported 52.0 hours of activity per week.Living arrangement was significant at the p = .0247 level (H corrected for ties = 9.377).Older respondents were significantly more likely to report less hours (Zcorrected for ties = -2.498, p = .0125). However, there were no significant differencesin hours reported by gender or educational level. Males (n = 41) reported 81.7 hoursper week on average, and females (n = 113) reported an average of 84.8 hours perweek. The 40 respondents with less than high school education reported a mean of82.6 hours per week, compared to 88.0 hours for those with high school (n = 63) and81.8 hours for those with post secondary education (n = 47).Hours of activity reported per week was not significantly associated with HealthDrop-In participation. Participants (n = 56) reported a mean of 76.3 hours, whilenonparticipants (n = 98) had a mean of 88.3 hours. P1 participants (n = 44) had amean of 79.0 hours and P2 (n=12) participants a mean of 66.7 hours. Total reportedhours of activity was also not a significant factor among program components attendedby participants. The BPM participants (n = 33) reported a mean of 76.7 hours ofactivity per week, compared to 75.2 hours (n = 33) for E participants and 86.3 hoursfor N/SM participants (n = 19).-101-4.9.2 Household ActivityThe mean weekly hours for household activity for the 154 respondents withusable data was 12.9, with a range of 0 to 64 hours. Hours spent on householdactivity were not significantly associated with place of residence. For the 89 apartmentresidents, the mean number of hours per week spent on household activity was 12.5.For residents of single family dwellings (n = 62), the mean was 13.8 hours and for the3 respondents who fitted neither of these categories the mean was 7.3 hours.There was no significant correlation of reported hours of household activityaccording to age. However, males reported marginally less hours of householdactivity per week (8.9 hours, n = 41) than females (14.4 hour, n = 113). The MannWhitney U Z score was -2.038, with a p value of .0415.Those who lived alone had a mean of 12.7 hours of household activity perweek, respondents who lived with a spouse reported 13.4 hours on average,respondents who resided with a relative or a friend reported 13.0 hours, and the 1respondent who did not fit any of these categories reported 3.0 hours of householdactivity. Respondents who had less than high school education spent an average of11.9 hours on household activity, compared to 15.1 hours for high school graduatesand 11.0 hours for those with educational levels past high school. Neither livingarrangement nor educational status was significantly associated with reported hoursper week of household activity.Hours spent on household activity per week were not significantly associatedwith participation in the Health Drop-In. Participants (n = 56) spent a mean of 11.5hours compared to 13.7 hours for nonparticipants (n = 98). The P1 participantsreported 12.3 hours on average, while the P2 group reported 8.8 hours. The P2group's lower average was not significant, but was in line with the tendency for other Eparticipants, who reported lower (but not significantly lower) hours per week ofhousehold activity than the other components, at 10.5 hours. The BPM participants-102-had a mean of 12.7 hours, and the N/SM participants a mean of 13.6 hours.4.9.3 ExerciseFor hours per week of exercise, the range among the 154 respondents was 0 to28 hours, with a mean of 4.8 hours. Place of residence was not significantlyassociated with exercise hours. Apartment residents averaged 4.4 hours per week,compared to single family dwelling residents at 5.4 hours and others at 4.0 hours.Gender was not a significant variable with respect to hours of exercise perweek. Males reported a mean of 5.7 hours per week, while females reported 4.7 hoursper week on average. Living arrangement was also not significantly associated withhours of exercise. Respondents who lived alone reported a mean of 4.4 hours perweek, compared to 5.5 hours for respondents living with a spouse, 4.7 hours for thoseliving with a relative or friend, and 0 hours for the 1 individual not fitting thesecategories.Educational status was, however, associated with hours of exercise per weekreported by respondents. Respondents who had finished high school (n = 63, mean of5.7 hours per week) or received education past the high school level (n = 47, 5.5 hoursper week) reported significantly higher hours than those with less than high schooleducation (n = 40, mean of 2.8 hours per week). The Kruskal Wallis H value correctedfor ties was 10.086, which was significant at the .0065 level. Age was also negativelycorrelated with hours of exercise per week, with older respondents reportingsignificantly lower hours of exercise per week (Z corrected for ties = -2.69, p = .0072).Although participants had a higher mean for hours of exercise per week (5.6hours compared to 4.3 hours for nonparticipants), this difference was not significant.There was no significant difference between P1 (5.6 hours) and P2 (5.4 hours), norwere there significant associations between hours of reported exercise per week andthe various program components. E participants reported a mean of 5.9 hours perweek, BPM participants 5.3 hours per week and N/SM participants 4.2 hours per week.-103-4.9.4 HobbiesThe range of reported hours by the 154 respondents for hobbies was 0 to 50and the mean was 7.1 hours. Single family dwelling residents reported more hoursper week on average, at 9.1 hours. This was not, however, significantly different fromthe reported hours by apartment residents (5.8 hours) or others (6.7 hours).Males reported spending more time on hobbies (8.3 hours) than females (6.9hours), but the difference was not significant. There was no significant correlationbetween age and reported hours spent on hobbies. Neither was living arrangementsignificantly associated with time spent on hobbies, although individuals who livedwith a spouse reported more hours (9.0) than those living alone (5.9) or with a relativeor friend (6.5). Educational status was also not associated with time spent on hobbies.Respondents with less than high school education had the highest mean, at 8.6 hours,followed by those with post secondary education (6.8 hours) and those who finishedhigh school (6.5 hours).Mean hours per week spent on hobbies was not significantly associated withHealth Drop-In participation, although nonparticipants had a higher mean (8.2 hours)than participants (5.1 hours). P1 reported a mean of 5.5 hours, while P2 had a meanof 4.0 hours per week. N/SM participants had the highest mean of reported hours perweek among the three program components, at 6.6 hours, followed by BPM at 5.1hours and E at 4.6 hours. However, reported hours spent on hobbies per week werenot significantly associated with participation in any of the program components,among the participants.4.9.5 Visiting Friends and RelativesIn the category of "visiting friends or relatives", the range of values was 0 to 50hours with a mean of 5.6 hours. Place of residence was not associated with reportedhours spent on visiting by this respondent sample. Apartment dwellers reported amean of 5.6 hours per week, compared to 5.9 hours per week for single family-104-residents and 2.7 hour per week for the 3 in the "other" category.Age was not significantly correlated with reported hours per week of visiting.Likewise, gender was not significantly associated with time spent visiting, with malesreporting a mean of 5.1 hours and females a mean of 5.9 hours. Respondents wholived alone reported 6.1 hours per week on average, compared to 5.6 hour forindividuals living with a spouse and 3.1 hours for those living with a relative or friend.The respondent who reported another living arrangement had 0 hours per week in thevisiting category. Respondents with high school education had a mean of 6.2 hoursper week of visiting, compared to 4.8 hours for those with less than high school and6.0 hours for those with more than high school.Hours per week spent visiting family and friends were not significantlyassociated with participation in the Health Drop-In. Participants reported 5.8 hours perweek on average, while nonparticipants had a mean of 5.6 hours. P1 participantsreported 5.5 hours, and P2 participants 6.8 hours. Hours of visiting per week werealso not significantly associated with participation in any of the components among theparticipants. The mean for BPM participants was 4.6 hours, for E participants 5.5hours and for N/SM participants 6.4 hours.4.9.6 Community ActivitiesRespondents spent an average of 1.9 hours per week on community activities,with a range of 0 to 35 hours. There was a weak association between place ofresidence and reported time spent on community activities (H corrected for ties = 6.68,p = .0354). Apartment dwellers spent reported a mean of 1.2 hours per week (n = 89)compared to single family residents with a mean of 2.7 hours (n = 63) and others at 6.3hours (n = 3).There was a strong correlation between age and hours spent on communityactivities, with older respondents reporting less time per week (Z corrected for ties =-3.547, p = .0004). Males reported a mean of 0.7 hours per week, while females-105-reported a mean of 2.4 hours per week. This difference was not significant. Livingarrangement and educational status were also non-significant. Respondents wholived alone reported 1.9 hours per week, those who lived with a spouse reported 2.0hours and those who lived with a relative or friend reported 1.4 hours. Individuals withless than high school education had a mean of 2.6 hours, while those with high schoolreported 2.0 hours and respondents with post secondary education 1.4 hours perweek.As would be expected from the higher proportions of participants involved invarious community activities reported in the previous section, time spent on communityactivities was significantly associated with participation in the Health Drop-In (Zcorrected for ties = -2.758, p = .0058). Participants reported a mean of 2.9 hours perweek of time spent on community activities, while nonparticipants reported a mean of1.4 hours per week. Participants in P1 (3.0 hours) were not significantly different fromparticipants in P2 (2.5 hours). Hours of community activity were not significantlyassociated, among participants, with attendance in any of the components. BPMparticipants had a mean of 2.9 hours, compared to E participants at 2.2 hours andN/SM participants at 3.8 hours.4.9.7 GardeningGardening accounted for a mean of 3.1 hours per week among respondents,with a range of 0 to 42 hours. Several individuals commented that this was a seasonalactivity. Single family residents reported significantly higher (H corrected for ties =40.138, p = .0001) hours per week spent on gardening (6.3 hours) than apartmentresidents (1.0 hours) or others (1.0 hours).Greater age was correlated with fewer reported hours spent on gardening (Zcorrected for ties = -2.127, p = .0334). In addition, respondents living with a spousereported significantly more hours (4.8) than those living alone (2.0), with a relative orfriend (2.8) or the 1 "other" (0 hours) (H corrected for ties = 12.986, p = .0043).-106-Education was not significantly associated with gardening hours. Respondents withless than high school reported 2.9 hours on average, while those with high school orequivalent reported 3.4 hours and those with post secondary education reported 3.1hours.There was no association between hours spent on gardening and Health Drop-In participation. Participants reported spending a mean of 2.1 hours per week ongardening, while nonparticipants reported a mean of 3.7 hours. P1, at 2.3 hours, wasnot significantly different than P2, at 1.1 hours per week.However, BPM participants reported spending significantly more hours perweek on gardening (3.0 hours) than non-BPM participants (Z corrected for ties =-2.204, p = .0275). Participants of the E component reported 1.5 hours and those inthe N/SM component reported a mean of 2.6 hours.4.9.8 SleepThe reported range of hours per week of sleep among respondents was 0 to 90,with an average of 47 hours. There was no significant association between reportedhours of sleep and place of residence. Apartment dwellers had a mean of 50.5 hours,single family residents a mean of 49.6 hours and the 3 who resided in another type ofdwelling reported a mean of 43.7 hours.Age was not correlated with reported hours of sleep. Nor was gendersignificantly associated with time spent on sleep. Males reported a mean of 50.7hours per week, while females reported 49.7 hours. Living arrangement was not asignificant variable: respondents who lived alone reported 48.9 hours on average,those living with a spouse reported 50.8 hours, respondents living with a relative orfriend had a mean of 52.6 hours, and the 1 individual with another arrangementreported 49 hours of sleep per week. Educational status was also a non-significantvariable with respect to hours of sleep per week. Respondents with less than highschool reported 52.9 hours per week, those with high school reported 49.7 hours and-107-those with post secondary education reported a mean of 47.7 hours per week.Reported hours of sleep per week were not significantly associated with HealthDrop-In attendance. Participants had a mean of 47.0 hours per week compared to51.5 hours per week reported sleep, on average, by nonparticipants. Participants inP1 reported a mean of 49.8 hours, while P2 participants reported 38.0 hours.Reported hours of sleep per week did not vary significantly among the variousprogram components. BPM participants had a mean of 49.8 hours, E participants amean of 46.3 hours and N/SM participants a mean of 50.6 hours.4.10 REASONS FOR ATTENDING THE HEALTH DROP-INFifty-five respondents provided reasons for their participation in the HealthDrop-In. All 8 of the respondents who did not provide reasons were exerciseparticipants who perceived themselves as nonparticipants (P2). The 8 individuals whodid not provide reasons were deleted in calculation of averages.There were 18 possible categories of reasons, plus an "Other" category.Respondents gave an average of 5.2 reasons, with a range of 0 to 15. Of the 55respondents, 51 (92.7%) chose at least one predisposing reason, 26 (47.2%) pickedat least one enabling reason and 37 (67.3%) selected at least one reinforcing reason.On average, individuals chose 2.5 (49%) predisposing reasons, 1.0 (19%)enabling reason and 1.6 (31%) reinforcing reasons. This is, however, misleading, asthere were different numbers of choices in each category (6 predisposing, 4 enablingand 8 reinforcing). Therefore, each reason was weighted (predisposing = 1.5X,enabling = 2X and reinforcing = X) so that all would have had an equal probability ofbeing chosen if the three of the factors were equally important to participants. (Forexample, "Health is important to me" was chosen by 40 respondents and is counted asa predisposing reason, so 40 * 1.5 = 60). Weighting each reason in a similar fashionled to a weighted total number of reasons chosen of 406. Dividing the sum of the-108-weighted numbers for the six predisposing reasons by 406 yielded a weightedpercentage for the predisposing factor. Once the weighting calculations werecompleted, enabling reasons were relatively more frequently chosen than reinforcingreasons, with reasons classified under the predisposing factor having a 52% chanceof being chosen by the participants, reasons classified under the enabling factorhaving a 27% chance and reasons classified under the reinforcing factor having a21% chance.Another method of determining the relative importance placed upon each of thefactors is to determine how frequently it is chosen by respondents relative to totalpossible chances to select it. There were 6 predisposing reasons. If all 55respondents chose each predisposing reason, there would have been 330predisposing choices in total. Respondents actually chose 140 predisposing reasonsin total. Therefore, respondents chose 42.4% of all possible predisposing choicesavailable to them.The 4 enabling reasons provided an opportunity for the 55 respondents tochoose 220 reasons. Respondents actually chose 55 enabling reasons in total, whichis equal to 25.0% of all possible enabling choices available to them.The 8 reinforcing reasons provided 440 total choices in this category to the 55respondents. There were 86 reinforcing reasons chosen it total, representing 19.5% ofall possible reinforcing choices available to them. This represents a similar pattern tothat which was obtained with unweighted percentages.Looking at the individual responses, the most commonly chosen reason forattendance (40 responses) was "Health is important to me". This was followed by"Want to stay active" (28 responses), "Need the exercise" (24), "Centre is close andeasy to get to" (22), "I'm interested in learning how to keep healthy" (21) and "I want tocontinue living at home" (21) responses. Five of the six most frequently chosenresponses were classified under the predisposing factor. The only exception was the-109-response, "Centre is close and easy to get to", an enabling response.The response, "Enjoy exercising" was chosen by 19 respondents, "Toparticipate with a group" by 16 and "My friends and /or spouse go" by 15. "To makefriends" and "I'm good at exercise" had an equal number of responses at 14. The firstfour responses were all classified as reinforcing factors, while "I'm good at exercise"was classified as enabling, but could also be a reinforcing factor.The lowest number of responses were for the enabling reasons "I go to otheractivities so I'm there" (10), "Have lots of time" (9), "Break in routine" (7), "Relief fromboredom" (7); the reinforcing reasons "My doctor told me to go" (6), "To be accepted byothers (2) and the predisposing reason "To participate/volunteer in community work"(7). These results are summarised in Table 18.Five individuals provided comments in the "Others" section. Two of thesecomments were to clarify role (stating that the respondents were RNs who volunteeredtheir time to take blood pressures). One person stated that "keeping busy is a greathelp". She was also a centre volunteer and commented that she worked in the kitchenand took care of the plants. A fourth individual, also a Health Drop-In volunteer,provided details of her volunteer work in this section, which included being a memberof the Kerrisdale Senior Centre Operating Committee, reporting to the A.S.K. (Arbutus-Shaughnessy-Kerrisdale Friendship Association) and working each Wednesdaymorning at the Centre (Health Drop-In). The final respondent's comment was, "To rubshoulders with others and learn how others feel about the world around them.""My doctor told me to go" was classified as a reinforcing reason, but could alsobe considered a predisposing one. Therefore, a contingency table analysis was tohave been performed to determine whether individuals who chose this response alsotended to choose other predisposing reasons. However, due to the low numbers ofindividuals choosing this response, it was not possible to determine which of the twofactors it was most appropriately classified under.-110-Full data is presented for each of the three factors (predisposing, enabling andreinforcing), and summarised in Table 19. Only significant differences are reportedhere for individual reasons falling into each of these categories. Where the number ofrespondents choosing any reason was less than 5, results are not reported assignificant. Tables presenting complete information by reason are included inAppendix E.TABLE 18:^REASONS GIVEN FOR ATTENDANCEREASON Number Choosing *Weighted % *c% Total Resp.PREDISPOSING:^(# choosing @ least 1) 5 1 52 42.4• Health is important to me 40• Want to stay active 2 8• Need the exercise 2 4• Interested in learning how to keep healthy 21• Want to continue living at home 20• ParticipateNolunteer in community work 7ENABLING:^(# choosing @ least 1) 2 6 27 25• Centre is close and easy to get to 22• I'm good at exercise 1 4• I go to other activities so I'm there 1 0• Have lots of time 9REINFORCING:^(# choosing @ least 1) 3 7 21 19.5 • Enjoy exercising 19• To participate with a group 1 6• My friends and/or spouse go 1 5• To make friends 1 4• Relief from boredom 7• Break in routine 7• My doctor told me to go 6• To be accepted by others 2N = 55* Weighted % refers to percentage of each of the factors (predisposing, enabling and reinforcing) respondentschose, taking into account the differing numbers of reasons under each factor (ie. enabling = 2X,predisposing = 1.5X, reinforcing = X). % of total responses refers to the proportion of responses under eachfactor chosen, compared to the total if all 55 respondents had chosen each reason under that factor.TABLE 19: CHARACTERISTICS OF ATTENDERS CHOOSING ATLEAST ONE PREDISPOSING, ENABLING OR REINFORCING REASO NVARIABLE PREDISPOSING (51) ENABLING (26) REINFORCING (37)Value Sig. Value Sig. Value Sig.Age: (mean years) 74.4 75.3 74.6ISAI Scales:^ER 30.2 30.7 29.8ES 24.3 25.3 24.3PH 23.8 24.8 24.6TO 30.8 30.6 30MO 28.9 29.6 28.6CS 24 23.8 24.6SS 29.5 29.4 29.1Activity^Hrs:House. Act. 11.9 10.1 10.4Exercise 5.1 0.0331 6 6Hobbies 4.7 0.0469 4.9 5.1Visiting 6 3.8 4.7Commun. Act. 2.9 1.8 2.3Gardening 2.2 0.0403 2 1.9Sleep 48.6 49.2 47.3Total Reported Hrs 76.7 0.0441 74.5 Predisposing ReasonsThe average age of the 51 respondents choosing at least one predisposingreason was 74.4 years. Place of residence was non-significant, with 35 living inapartments, 13 in single family dwellings and 3 with some other arrangement.Eleven males and 40 females were in this category. Thirteen had less thanhigh school education, while 22 had finished high school and 14 had post secondaryeducation. Twenty-six lived alone, 21 with a spouse and 4 with a relative or friend.None of these demographic variables were significantly associated with choosing apredisposing reason. Educational status was, however, significantly different for thepredisposing reason, "Interested in learning how to keep healthy," with lessrepresentation from respondents with postsecondary education (3) and morerepresentation from those with less than high school (9). Nine respondents with high-112-school or equivalent education also chose this reason (p = .0278).The mean score on the ER scale was 30.2, on the EB scale 24.3, on the PHscale 23.8, on TO 30.8, on MO 28.9, on CS 24.0 and on the SS scale 29.5.Differences in mean scores on the ISAI scales were not significantly associated withchoice of at least one predisposing reason. Mean scores on the TO scales weresignificantly higher for two of the predisposing reasons: "Health is important to me,"(30.7, p = .0498) and "I want to continue living at home" (31.2, p = .0091).Fourteen of the 51 respondents were involved with other activities at theKerrisdale Senior Centre, 15 attended church, 13 belonged to church groups ororganisations and 18 attended cultural centres or organisations. Twenty-threeattended the symphony, opera, etc., 18 were involved with recreational activities, 7were members of other senior centres, 5 attended the community centre and 11indicated some other form of community involvement. For the group choosing at leastone predisposing reason, community involvement was not significant. However, otherinvolvement in Kerrisdale Senior Centre was positively associated with choice of thereason, "To participate/volunteer in community work" (6 respondents, p = .0011), andinvolvement in other senior centres was weakly associated with choice of "I want tocontinue living at home" (6 respondents, p = .0394). In general, proportions ofrespondents indicating involvement in the various community activities were equal toor above the mean for all participant respondents.Respondents choosing at least one predisposing reason reported marginallyless hours of exercise (5.1 hours, p = .0331) and hobbies (4.7 hours, p = .0469) perweek and marginally more time per week of gardening (2.2 hours, p = .0403) andmore total hours (76.7 hours, p = .0441) than the rest of the participants. Mean hoursspent on household activities (11.9), visiting (6.0), community activity (2.9) and sleep(48.6) were not significantly different from respondents who did not choose at leastone predisposing reason. Reported hours of community activity were significantly-113-higher (7.7 hours) for respondents choosing the reason, "To participate/volunteer incommunity work" (p = .0011).Choice of at least one predisposing reason was not significantly associated withattendance in the BPM component (35 respondents), E component (25 respondents)or the N/SM component (21 respondents). Choice of, "I want to continue living athome" was positively associated with attendance in the N/SM component of theprogram (p = .0111). Choice of "Need the exercise" was negatively associated withparticipation in the BPM component (p = .0085) and positively associated withparticipation in the E component (p = .0001) of the Health Drop-In.When the age and ISAI scale scores of respondents who chose predisposingreasons were divided according to the number of predisposing reason chosen, nosignificant patterns of responses were found. This may, however, be a power issue, asinspection of the data shows a tendency for decreasing age with number ofpredisposing reasons chosen, along with decreasing EB, MO, CS and SS scores.There is also a tendency for increasing TO scores. The ER and PH scales do not showa pattern, but there is a remarkable variation in scores on the PH scale. Table 20summarises mean age and ISAI scores by number of reasons chosen.TABLE 20:^PARTICIPANT AGE AND ISAI  RESULTS BY NO. OF PREDISPOSING REASONS CHOSENVARIABLES: NUMBER OF PREDISPOSING REASONS CHOSEN:by means 0 (N . 3) 1 - 3 (N . 32) 4 - 6 (N = 17)Age 75.3 74.6 73.9Economic Res. 28.3 30.7 29.2Emotional Bal. 29 24.5 24Physical Health 30.3 23.2 25Trusting^Others 29.7 30.2 30.9Mobility 30 28.9 28.8Cognitive Status 29 24.2 23.5Social Support 30.7 29.7 29-114-4.10.2 Enabling ReasonsPlace of residence was nonsignificant for the 26 respondents choosing at leastone enabling reason. Twenty-one resided in apartments, 4 in single family dwellingsand 1 in another type of residence.The mean age of individuals choosing at least one enabling reason was 75.3years. Gender, level of educational attainment and living arrangement were allnonsignificant, as were mean scores on the ISAI scales. The ER mean score was30.7; EB, 25.3; PH, 24.8; TO, 30.6; MO, 29.6; CS, 23.8 and SS, 29.4. All of thesemeans except CS were above the means for the participant sample overall.Hours of reported activity were also not significantly different for respondentschoosing at least one enabling reason. The mean reported hours for householdactivity was 10.1 hours, for exercise, 4.9 hours; for visiting, 3.8 hours; communityactivity, 1.8 hours; gardening, 2.0 hours and sleep, 49.2 hours.Ten of these respondents reported involvement with other activities atKerrisdale Senior Centre, 9 with church, 6 with church groups or organisations and 4with cultural centres or organisations. Twelve respondents in this category reportedattending the symphony, opera, etc. Nine were involved in recreational activities, 7 inother senior centres and 4 attended the community centre. Five indicated othercommunity involvement. Of the reported community involvement, there was amarginally higher proportion of individuals choosing at least one enabling reason whoattended other senior centres (p = .0373). Choice of "I go to other activities so I'mthere" was positively associated with attendance at other Kerrisdale Senior Centreactivities (p = .0002) and involvement with other senior centres (p = .0025).Sixteen of the respondents in this category were involved in the BPMcomponent, 17 in the E component and 12 in the N/SM component. None of thesewere significantly different from respondents who did not choose at least one enablingreason. Two of the individual reasons, however, ("I'm good at exercise" and "Centre is-115-close and easy to get to") were positively associated with attendance in the Ecomponent (p = .0052 and .0127, respectively).When mean age and scores on the ISAI scales were calculated for respondentsaccording to the number of enabling reasons that were chosen, there were nosignificant patterns using simple regression. There was a tendency for increased age,along with higher values on the ER scale with more enabling reasons chosen. Resultson the other scales are inconclusive; nevertheless, there is a great deal of variation inmean values depending upon number of enabling reasons chosen. Results aredisplayed in Table 21.TABLE 21:^PARTICIPANT AGE AND ISAI RESULTS BY NO. OF ENABLING REASONS CHOSENVARIABLES: NUMBER OF ENABLING REASONS CHOSEN:by means 0 (N = 27) 1^(n =^11) z 2 (n = 15)Age 73.5 75.6 75.1Economic Res. 29.5 30.3 30.9Emotional Bal. 23.9 26.5 24.5Physical Health 23.6 22.4 26.7Trusting^Others 30.3 30.7 30.5Mobility 28.3 30.3 29.1Cognitive Status 24.6 23.5 24.1Social Support 29.7 30 Reinforcing ReasonsPlace of residence was not significantly associated with choice of at least onereinforcing reason (n = 37).Gender, age, living arrangement and education were also not associated withchoice of at least one reinforcing reason. There were some demographic differencesrelated to choice of some of the reasons under the reinforcing category, however.Gender was significantly associated with choice of "To participate with a group", withnone of the males choosing this reason (p = .0218). Living arrangement was-116-significantly associated with choice of three reasons: "Relief from boredom" (morerespondents living with friends or relatives choosing, p = .0004); "To participate with agroup" (more respondents living alone or with friends or relatives choosing, p = .0089)and "To make friends" (more respondents living with friends or relatives choosing, p =.0013).ISAI scale scores were not significantly associated with choice of at least onereinforcing reason. Mean scores for these 37 respondents were: ER, 29.8; EB, 24.3;PH, 24.6; TO, 30.0; MO, 28.6; CS, 24.6 and SS, 29.1. Choice of "To participate with agroup" was associated with a higher mean on the MO scale (29.9, p = .0221). Tworeasons were associated with lower means on the SS scale: "Relief from boredom"(26.4, p = .029) and "To make friends" (27.7, p = .0239).Reported hours of activity per week were not significant for respondentschoosing at least one reinforcing reason. The means for this group were: householdactivities, 10.4 hours; exercise, 6.0 hours; hobbies, 5.1 hours; visiting, 4.7 hours;community activities, 2.3 hours; gardening, 1.9 hours and sleep, 47.3 hours. Choice ofthe reason "My friends and/or spouse go" was associated with reporting significantlyfewer hours of community activity (0.9 hours, .0213).Proportions of respondents attending other community activities were notsignificantly associated with choice of at least one reinforcing reason. Twelve of the37 individuals attended other activities at Kerrisdale Senior Centre, 12 attendedchurch, 10 were involved in church groups or organisations and 6 in cultural centres ororganisations. Fifteen attended the symphony, etc., and 13 were involved inrecreational activities. Seven attended other senior centres, 5 went to the communitycentre and 5 indicated involvement in another community activity. Choice of "Toparticipate with a group" was marginally associated with involvement in other activitiesat Kerrisdale Senior Centre (8 respondents, p = .0366).Choice of at least one reinforcing reason was marginally associated with-117-attendance at the E component of the Health Drop-In (22 respondents, p = .0433) butnot the BPM (27 respondents) or N/SM (16 respondents) components. Choice of"Enjoy exercising" as a reason for participating was positively associated withparticipation in the E component (p = .0006), as was choice of "To participate with agroup" (p = .037).When mean age and mean scores on the ISAI scales were compared forrespondents choosing different numbers of reinforcing reasons, no significant patternswere found using simple regression tests. On inspection of the data, the mean ER, EB,MO and SS scale scores tend to decrease with the number of reinforcing reasonschosen. Results are shown in Table 22.TABLE 22: PARTICIPANT AGE AND ISAI RESULTS BY NO. OF REINF. REASONSVARIABLES: by meansNUMBER OF  REINFORCING0 (N = 16)REASONS CHOSEN: 1 - 3 (n = 29) 4 - 6 (n = 7)Age 74 74.8 73.6Economic Res. 30.7 30 28.9Emotional Bal. 25.3 24.3 24.3Physical Health 23.3 24.7 24.4Trusting^Others 31.5 29.8 30.6Mobility 29.6 28.8 27.9Cognitive Status 23.6 25.2 21.7Social Support 30.5 29.7 26.94.11 REASONS FOR NOT ATTENDING THE HEALTH DROP-INOf the 117 nonparticipant respondents, 107 provided reasons for not attending.Four of the ten who did not provide reasons checked that they did not know about theHealth Drop-In (which is also one of the reasons for not attending). The 107respondents who completed this section provided a mean of 2.26 reasons for non-attendance, with a range of 1 to 9. Of the 107 individuals who provided reasons, 50%(55) chose at least one predisposing reason, 83% (90) chose at least one enablingreason and 32% (33) chose at least one reinforcing reason. On average, respondents-118-chose 0.67 predisposing reasons (29.7%), 0.93 enabling reasons (41.2%) and 0.25reinforcing reasons (13.7%). There was a greater proportion of "other" reasons amongnonparticipants than participants. Altogether, there were 40 "other" responses, whichis a mean of 0.37 per respondent, accounting for the other 16% of the responses.Using the same weighting method that was performed for attendance reasons,the predisposing factor (6 items) had a weighting of 1.1667X, enabling (7 items) aweighting of X and reinforcing (5 items) a weighting of 1.4X. Recalculation of numbersof responses for each item using the appropriate weighting provided a total weightednumber of reasons of 230.9 ("other" reasons were excluded from this calculation). Thetotal weighted predisposing reasons were 81.67, which, when divided by the 107,came to a weighted average of 0.76 predisposing reasons (35.4%). For the enablingfactor, the total weighted reasons were 105.83, for a weighted average of 1.0 (45.8%).The total weighted reinforcing reasons were 43.4, with a weighted average of 0.41reasons, or 18.8%.If all 107 respondents chose each of the 6 predisposing reasons, a total of 642predisposing responses would have been recorded. In fact, 72 predisposingresponses were chosen in total, or 11.2% of all possible predisposing responses. Forthe 7 enabling reasons, there was a total of 749 possible responses for the 107respondents. There were 99 actual enabling responses. Therefore, 13.2% of allpossible enabling responses were actually chosen by the respondents. For the 5reinforcing reasons, the total possible responses would have been 535 for the 107respondents. Thirty-one reinforcing reasons, or 5.8% of the total possible reinforcingresponses were chosen.The most common reason given for lack of attendance at the Health Drop Inwas "Don't know about it". Forty-one individuals chose this response. Other commonresponses were "Not interested" (27), "Don't need it" (25), "Not enough time" (22),"Don't know anyone there" (19), "Would not enjoy it" (14) and "Can't do exercise" (14).-119-Table 20 shows a complete breakdown of numbers of respondents choosing eachreason."My doctor said not to" was classified as a reinforcing reason, but could also beconsidered an enabling reason related to "poor health". Therefore, responses were tobe compared to determine whether individuals who chose "My doctor says not to" alsochose "Poor health". "My doctor says not to" was the only reason which was notchosen by any respondents, eliminating the need for this analysis.In the following section, complete profiles of respondents choosing at least onepredisposing reason, at least one enabling reason and at least one reinforcing reasonwill be given. These are summarised in Table 23. Results for specific reasonsclassified under each of these categories will only be discussed where they weresignificant. A complete summary of information by reason chosen is provided inAppendix F.4.11.1 Predisposing ReasonsPlace of residence was not significantly associated with choice of at least onepredisposing reason. Of the 50 respondents in this category, half lived in apartmentsand the other half in single family dwellings.The average age of these respondents was 76.4 years. Nineteen were maleand 31 female. Nine had less than high school education, 23 had completed highschool and 17 had post secondary education. Twenty-one lived alone, 24 with aspouse and 5 with a relative or friend. These variables were not significantlyassociated with choice of a predisposing reason. Gender was marginally associatedwith choice of "Would not enjoy it" (males choosing more frequently, p = .038).TABLE 23:^REASON GIVEN FOR NONATTENDANCEREASONS RESPONDENTSCHOOSING*WEIGHTED % *% OF TOTALPOSSIBLEPREDISPOSING:^(# choosing @ least 1) 50 35.4 1 1 .2• Not interested 27• Don't need it 2 5• Would not enjoy it 1 4• Don't like to leave home 3• Won't do me any good 2• Health is not important to me 1ENABLING:^(# choosing @ least 1) 7 0 45.8 13.2 • Don't know about it 41• Not enough time 2 2• Can't do exercise 1 4• Poor health 9• No transportation 8• Can't afford it 4• Too far to centre 1REINFORCING:^(# choosing @ least 1) 23 18.8 5 . 8• Don't know anyone there 1 9• Friends/spouse don't go 7• Don't feel welcome there 3• Friends/Spouse think its silly 2• Doctor says not to 0OTHER: 40n = 107* Weighted % refers to percentage of each of the factors (predisposing, enabling and reinforcing)respondents chose, taking into account the differing number of reasons classified under each factor.% of total possible refers to the proportion of responses chosen under each factor, compared to thetotal if all 107 respondents had chosen each reason under that factor.-121-TABLE 24: CHARACTERISTICS OF NONATTENDERS CHOOSING ATLEAST ONE PREDISPOSING, ENABLING OR REINFORCING REASONVARIABLE PREDISPOSINGValue(50)Sig.ENABLINGValue(70)Sig.REINFORCINGValue(23)Sig.Age: (mean years) 76.4 77.2 80.7 0.0081ISAI Scales:^ER 29.7 28.7 28.2EB 27.3 0.0222 25.7 24.8PH 23.2 0.01 20.9 20.1TO 30.9 30.8 30.2MO 28.2 26.3 0.0328 24 0.0095CS 25.9 0.0314 25.1 24.1SS 30.6 29.8 27.5 0.0042Activity^Hrs:House. Act. 13.3 11.4 10.3Exercise 5.4 4.1 3.2Hobbies 7.5 10.1 9.2Visiting 5.8 5 4.3Commun. Act. 0.6 0.0185 1.4 0 0.0146Gardening 4.5 4.4 1.3 0.0045Sleep 53.1 49.1 0.01 48.1Total Reported Hrs 89.6 84.9 73.7 0.0386Respondents choosing at least one predisposing reason had significantlyhigher means on 3 of the ISAI scales: EB (27.3, p = .0222), PH (23.2, p = .01) and CS(25.9, p = .0314). Means on the other four scales were all above the totalnonparticipant means, although not significantly. The mean for the ER scale was 29.7,for TO the mean was 30.9, for MO it was 28.2 and for SS the mean was 30.6. Choiceof the reason "Not interested" was associated with higher means of the EB (27.8, p -.0447) and SS (30.9, p = .0214) scales. Choice of "Don't need it" was associated withhigher means on the PH (25.6, p = .0001), TO (31.4, p = .0415) and MO (29.8, p =.015) scales.Reported hours of activity were nonsignificant for respondents choosing at leastone predisposing reason, with the exception of community activities, which wassignificantly lower for this group (p = .0185). Mean values were: household activities,-122-13.3 hours; exercise, 5.4 hours; hobbies, 7.5 hours; visiting, 5.8 hours; communityactivities, 0.6 hours; gardening, 4.5 hours and sleep, 53.1 hours.Likewise, participation in other community activities was not associated withchoice of at least one predisposing reason. Five respondents attended other activitiesat Kerrisdale Senior Centre, 14 attended church and 10 belonged to church groups ororganisations. Four belonged to cultural centres or organisations, 15 attended thesymphony, etc., 12 participated in recreational activities, 1 went to other senior centres,and 5 attended the community centre. Thirteen respondents indicated involvement inother community activities.Twenty-nine of the 50 respondents choosing at least one predisposing reasonindicated that they knew about the Kerrisdale Health Drop-In.When mean age and mean scores on ISAI scales were calculated forrespondents by the number of predisposing reasons chosen, a pattern of increasingage and increasingly high scores with number of reasons was evident. Using simpleregressions, significant patterns were found for PH (higher scores with increasingnumber of reasons chosen, F = 10.917, p = .0013) and for EB (higher scores withincreasing number of reasons chosen, F = 4.812, p = .0305). Table 25 summarisesinformation by number of predisposing reasons chosen.TABLE 25: NONPARTICIPANT AGE AND ISAI RESULTS BY NO. OF PRED. REASONSVARIABLES:by meansNUMBER OF0 (N = 60)PREDISPOSING1 (n = 35)REASONS2 (n = 9)CHOSEN:3 (n = 5) 4 (n = 1) Sig.Age 77.2 75.5 78.6 78.4 79 NSEconomic Resources 28.5 29.8 29.6 31.8 16 NSEmotional Balance 25.1 27.1 28.6 26.4 29 0.0305Physical Health 20 22.3 24.7 25.8 29 0.0013Trusting^Others 30.6 30.6 31.7 32 29 NSMobility 26.5 28.2 27.8 28.8 28 NSCognitive Status 24.1 25.9 27.2 22.2 31 NSSocial Support 29.4 31.1 29.9 31.4 15 NS-123-4.11.2 Enabling ReasonsPlace of residence was not significantly associated with choice of at least oneenabling reason, with 39 respondents residing in apartment buildings compared to 31in single family dwellings. Choice of the reason "Not enough time" was, however,positively associated with living in single family dwellings (15 of the 22 respondentschoosing this reason lived in single family dwellings, p = .0148).Age, gender, educational level and living arrangement were all nonsignificantwith respect to choice of at least one enabling reason. The average age of the 70individuals in this category was 77.2 years. Twenty-one respondents were male and49 female. Thirty-eight lived alone, 24 with a spouse, 7 with a friend or relative andone had some other arrangement. Twenty-two had finished high school, while 21 hadless than high school and 23 had received education past the high school level. Meanage was significantly higher with respect to choice of three specific reasons: "Notransportation" (86.4 years, p = .0019); "Poor health" (82.1 years, p = .0315) and "Can'tdo exercise" (82.0 years, p = .008). Choice of "Poor health" as a reason was alsoassociated with living alone or with a relative/friend, rather than a spouse (p = .005).Scores on the ISAI for the 70 respondents who chose at least one enablingreason were at or below the sample means for all scales except CS, but only MO wassignificantly lower (p = .0328). Mean scale scores were: ER, 28.7; EB, 25.7; PH, 20.9;TO, 30.8; MO, 26.3; CS, 25.1 and SS, 29.8. With regard to specific reasons, "Notenough time" showed a pattern similar to the predisposing reasons, with all meanscores above the nonparticipant group means, two of them significantly above themeans (PH, 24.7, p = .0074 and CS, 26.5, p = .0427). Choice of the reason "Notransportation" was associated with significantly lower scores on the MO scale, at 21.5(p = .0039). Choice of "Can't afford it" was, as would be expected, associated withlower scores on the ER scale, at 15.0 (p = .0027). There was also a significantassociation with lower scores on the TO scale (29.3, p = .0199). Choice of "Poor-124-health" was associated with lower scores on the EB (mean of 23.9, p = .0109), PH(mean of 12.9, p = .0001), MO (mean score of 18.1, p = .0001) and CS (mean of 21.7,p = .0465) scales. The reason "Can't do exercise" was also significantly associatedwith lower scores of four of the scales; PH (17.2, p = .0084), MO (21.4, p = .0001), CS(22.2, p = .0338) and SS (28.4, p = .04).For respondents choosing at least one enabling reason, reported hours ofsleep were significantly lower than the nonparticipant group mean at 49.1 hours perweek (p = .01). Means for the other areas of activity were: household activity, 11.4hours; exercise, 4.1 hours; hobbies, 10.1 hours; visiting, 5.0 hours; communityactivities, 1.4 hours and gardening, 4.4 hours. Respondents choosing "Not enoughtime" as a reason reported spending significantly more hours on gardening, at 7.1hours (p = .0225). Choice of "Poor health" as a reason was associated with fewerreported hours on household activities (4.2 hours, p = .0253) and exercise (1.0 hours,p = .022). Choice of "Can't do exercise" was also associated with fewer reportedhours on household activity (6.5 hours, p = .0111) and on sleep (45.2 hours, p =.0174).Proportions of respondents involved in community activities were not significantwith regard to choice of at least on enabling reason. Six respondents were involved inother activities at Kerrisdale Senior Centre, 27 attended church, 13 participated inchurch groups or organisations and 7 in cultural centres or organisations. Twenty-onerespondents attended the symphony,etc., 12 participated in recreational activities, 5attended other senior centres, 8 went to the community centre and 18 indicatedinvolvement in other community activities. Choice of "Don't know about it" as a reasonfor nonattendance was marginally associated with less involvement in other KerrisdaleSenior Centre activities (only 1 of 43 respondents choosing this reason was involvedin other Kerrisdale Senior Centre activities, p = .0302).A lack of knowledge about the Health Drop-In was significantly associated with-125-choice of at least one enabling reason (p = .0001), but the inclusion of "Don't knowabout it" as a reason under the enabling category may account for this result, as a full43 of the 70 respondents chose the reason "Don't know about it".When mean age and mean scores on ISAI scales were calculated by number ofenabling reasons chosen, a pattern of increasing age and decreasing scale scoreswith more reasons chosen was observed. Using simple regression, PH and MOscores showed a significant pattern of decreasing scale scores with increasingnumber of enabling reasons chosen. (PH had an F-value of 5.746 with p = .0183; MOhad an F-value of 26.809, p = .0001). Values are shown in Table 26.TABLE 26: NONPARTICIPANT AGE AND ISAI RESULTS BY NO. OF ENAB. REASONSVARIABLES:by meansNUMBER OF0 (N = 39)ENABLING REASONS1 (n = 48)CHOSEN:2 (n = 16) 3 (n = 5) 4 (n = 1) Sig.Age 76.1 75.7 80.1 81.81 79 NSEconomic Resources 29.5 29.4 27.1 27 30 NSEmotional Balance 26.8 26.2 25.3 23.3 21 NSPhysical Health 22.7 21.7 19.6 16.8 17 0.0183Trusting^Others 30.7 30.8 30.6 31.8 29 NSMobility 29 28.2 22.4 20.5 22 0.0001Cognitive Status 25.1 25.3 26.6 19.8 11 NSSocial Support 30.2 30.1 28.8 30 29 NS4.11.3 Reinforcing ReasonsPlace of residence was not a significant variable with respect to choice of atleast one reinforcing reason. Fourteen of the 23 respondents resided in apartments,and the remainder were single family dwelling residents.The age of respondents choosing at least one reinforcing reason wassignificantly higher than the rest of the nonparticipant group at 80.7 years (p = .0081).No other demographic characteristics were significant with respect to choice of at leastone reinforcing reason. Nine of the respondents were male, and 25 were female.-126-Thirteen lived alone, 6 with a spouse, 3 with a relative or friend and 1 had anotherarrangement. Seven each had less than high school and high school equivalenteducation, while 8 had post secondary education. With regard to specific reasons,choice of "Don't know any one there" was associated with being older (mean of 81years, p = .015).Mean scores on all ISAI scales were below the means for the totalnonparticipant group: ER, 28.2; EB, 24.8; PH, 20.1; TO, 30.2; MO, 24.0; CS, 24.1 andSS, 27.5. The MO scale (p = .0095) and the SS scale (p = .0042) were significantlylower. Lower scores on the EB (mean of 24.3, p = .039), MO (mean of 23.3, p = .0081)and SS (mean of 26.4, p = .0001) scales were associated with choice of the reason"Don't know any one there".Means for reported hours of activity were below the total nonparticipant groupmeans on all activities except hobbies. Significantly lower reported values were foundfor community activities (mean of 0 hours, p = .0146) and gardening (mean of 1.3hours, p = .0045). Mean reported hours for other activities were: household activities,10.3 hours; exercise, 3.2 hours; hobbies, 9.2 hours; visiting, 4.3 hours and sleep, 48.1hours. Lower reported hours were also significant for choice of the reason "Don'tknow any one there" for exercise (2.5 hours, p = .0347), community activities (0 hours,p = .0332) and gardening (1.5 hours, p = .0165).For respondents choosing at least one reinforcing reason, no involvement wasreported with cultural centres and organisations or other senior centres. Threeindividuals indicated that they attended other activities at Kerrisdale Senior Centre, 9attended church, 4 belonged to church groups or organisations, 3 went to thesymphony, etc., 3 attended recreational activities and 3 attended the communitycentre. Six respondents indicated involvement in other community activities.Knowledge of the Health Drop-In was not significantly associated with choice ofat least one reinforcing reason.-127-When mean age and mean scores on the ISAI scales were subjected to simpleregression techniques by number of reinforcing reasons chosen, two scales showedsignificant patterns. Scores on the MO scale decreased with increasing number ofreinforcing reasons chosen (F-value = 4.529, p = .0357), and a similar pattern wasfound for the SS scale (F-value = 10.159, p = .0019). Mean values by number ofreinforcing reasons chosen are summarised in Table 27.TABLE 27: NONPARTICIPANT AGE AND ISAI RESULTS BY NO. OF REIN. REASONSVARIABLES:by meansNUMBER OF REINFORCING0 (N = 86) 1^(n = 16)REASONS CHOSEN:2 (n = 6) 3 (n = 1) Sig.Age 75.8 82.6 76.3 76 NSEconomic Resources 29.2 27.6 29 32 NSEmotional Balance 26.4 24 26 29 NSPhysical Health 21.9 18.4 23.2 27 NSTrusting^Others 30.9 30.5 29.3 32 NSMobility 28.1 22.4 27.7 27 0.0357Cognitive Status 25.3 23.1 25.8 27 NSSocial Support 30.6 27.3 27.7 29 0.00194.12 SUMMARY OF RESULTS4.12.1 Hypothesis 1a) Scores in the normal range (less than one standard deviation above orbelow the mean) on the ISAI ER, EB, TO, MO, CS PH and SS scales will beassociated with participation in the Health Drop-In.b) Gender will be associated with participation (participants will be more likelyto be female).c) Participation in other Kerrisdale Senior Centre activities will be associatedwith participation in the Health Drop-In, as will involvement in other communityactivities, but generally as an attender rather than a leader or volunteer.d) Choice of predisposing and reinforcing reasons will be most commonamong participants.Mean ISAI scores were, as predicted, all within one standard deviation above orbelow the standard mean of 50. Mean standardised scores ranged fromapproximately 49.5 on the EB scale to 55 on the ER scale. (In retrospect, given therelatively wealthy condition of residents of the Kerrisdale area, it might have beenexpected that all mean results would be at or above the standard mean of 50. It wasalso noted, however, that elderly people, especially on fixed incomes, might havefinancial difficulties even in an area which is relatively well off economically.)A higher percentage of female respondents were participants (38.5%) thanmale respondents (26.0%). Contrary to what was hypothesised, however, thisdifference was not significant, but this may be due to lack of power. Within theexercise component of the program there were significantly more women than in theother components (94.3% female).As hypothesised, the participants of the Health Drop-In were significantly moreinvolved with other activities at the Kerrisdale Senior Centre. They were also morelikely to attend the symphony, opera, etc. It was not possible to utilise the "attend"-129-versus "lead/volunteer" categories in statistical analysis due to small numbers in someof the cells.It was hypothesised that the most common reasons chosen for attending wouldbe predisposing and reinforcing. This was only partially supported. Predisposingreasons were, by far, the most commonly chosen, but enabling reasons accounted forthe next most common choice.4.12.2 Hypothesis 2a) Higher mean scores than those of the participants on the ER, MO and SSscales will be associated with one subset of nonparticipants.b) PH, TO EB and CS mean scale scores will be equivalent to participantscores.c) Gender, (more females) and educational level (greater than high school) willbe associated with this subset of nonparticipants.d) Active involvement in other community activities, especially in a leader ofvolunteer (as opposed to attender) capacity, and reporting of more hours of activity perweek will be associated with this group.e) Past participation in the Health Drop-In may be associated with this subset ofnonparticipants.f) Reasons for nonattendance such as lack of interest, lack of need, lack ofenjoyment of or lack of time for the activity will more often be chosen by this subset ofnonparticipants.As three of the four reasons (all except lack of time) hypothesised to be mostcommonly chosen by this higher functioning group of nonparticipants were classifiedas predisposing, the group of nonparticipants who chose at least one predisposingreason was analysed as the possible group of high functioning nonparticipants.Contrary to what was anticipated, means on the ER, MO and SS scales wereapproximately equal for the participants and the nonparticipants choosing at least one-130-predisposing reason. This subset of the nonparticipants had a higher mean score onthe EB scale (27.3 compared to 24.4 for participants). Mean scores were also slightlyhigher for this subset of nonparticipants on the TO, CS and SS scales, but wereslightly lower on the ER, PH and MO scales.The subset of nonparticipants who chose at least one predisposing reason wasnot more likely to be female (62% of nonparticipants choosing at least onepredisposing reason were female compared to 68.4% of nonparticipants overall). Norwas this subset significantly more likely to have education past the high school level,although the difference was in the specified direction (30% of participants versus 34%of nonparticipants choosing at least one predisposing reason had education past thehigh school level). This subset was not more involved in community activities thanthe participants, with the exception of the "other" category (26%, compared to 22% forparticipants). In fact, their mean for reported hours of community activity wassignificantly lower than that of nonparticipants who did not choose at least onepredisposing reason for nonattendance, and, at 0.6 hours per week, far below theparticipant mean of 2.9 hours per week.The subset choosing at least one predisposing reason did, however, reportmore hours of activity per week than the participant group. Their mean for totalreported hours was 12 hours higher than the participants'. The increase was taken upby more hours of household activity (mean was 1.8 hours higher), hobbies (mean was1.7 hours higher), gardening (mean was 2.4 hours higher) and sleep (mean was 6.1hours higher).Therefore, the second hypothesis, that there would be a subset ofnonparticipants who was higher functioning than participants, as evidenced by highermeans scores on some ISAI scales and more community participation, was notsupported. The differing patterns of activity, together with the differing score profiles,suggests that the participants and the nonparticipants who chose at least one-131-predisposing reason are both high functioning groups, but in different ways.Participants reported spending more time in structured community activity, whereas,nonparticipants choosing at least one predisposing reason spent more time withactivties in the home.This subset of nonparticipants was, however, higher functioning than the rest ofthe nonparticipant group. Mean scores for the subset of nonparticipants who chose atleast one predisposing reason were higher than the total nonparticipant group meanson all ISAI scales, and significantly higher on the EB, PH and CS scales. At a mean of5.4 hours of exercise per week, they were higher than the total nonparticipant groupmean of 4.3 hours. In addition, this subset had a higher mean (4.5 hours) than thegroup mean (3.7 hours) for gardening (also an energetic activity). Their mean hours ofcommunity activity were significantly lower than the nonparticipant group mean,however.4.12.3 Hypothesis 3a) Lower scores on all the ISAI scales will be associated with a second subsetof nonparticipants.b) Gender and educational attainment will be associated with this subset, withgreater representation of males and higher proportions of individuals with less thanhigh school education being represented.c) Lower levels of community involvement and lower attendance at otherKerrisdale Senior Centre activities will be associated with this subset ofnonparticipants.d) Reasons for nonattendance will more likely be enabling factors, and will beassociated to results in the ISAI (for example, lack of knowledge associated with lowerscores in SS, TO and/or CS scales; too far to the centre, no transportation or poorhealth associated with low scores on MO and PH scales; can't afford it associated withlow scores on ER scale; don't like to leave home, don't feel welcome there associated-132-with low scores on the TO scale).For the other two groups in the nonparticipant sample (those who chose at leastone enabling reason and those who chose at least one reinforcing reason) means onthe ISAI scales were, with two exceptions, below the means for the total group ofnonparticipants (the exceptions were both with the "enabling" group and included TO,which was equal to the total group mean, and CS which was higher than the groupmean). Three values were significantly lower; MO for the "enabling" group; MO andSS for the "reinforcing" group. More would probably have shown up as significantlylower, if the reason "Not enough time" had not been included as an enabling reason.This reason was chosen by a large number of nonparticipants (22) and showed apattern more similar to that of the respondents who chose at least one predisposingreason than to those who chose at least one enabling reason (all ISAI means wereabove the total nonparticipant group means, with 2 of them significantly higher, PHand CS).Nonparticipants who chose at least one enabling reason had mean scoreslower than the participants' on the ER, PH and MO scales, and higher than theparticipants' on the EB, TO, CS and SS scales. Nonparticipants who chose at leastone reinforcing reason had lower mean scores than the participant group means onthe ER, PH, MO and SS scales, and equal or higher means on the EB, TO and CSscales. A look at specific reasons in these two categories supports a tendency towardlower, rather than higher, mean ISAI scale scores among these groups: four meanscores are significantly lower for the reasons, "Can't do exercise" and "Poor health";three are significantly lower for "Don't know anyone there"; two for "Can't afford it" andone for "No transportation." The only enabling or reinforcing reason which hadsignificantly higher mean scores was "Not enough time", with significantly highermeans on the PH and CS scales.The hypothesis that individual reasons would be associated with corresponding-133-scales was also supported. Significantly lower scores were present on the SS, EBand TO scale for individuals choosing, "Don't know anyone there." Significantly lowerscores were present on the PH and MO (as well as the CS and SS scales) forrespondents choosing the reason "Can't do exercise". Significantly lower scores werepresent on the PH and MO scales (as well as the EB and CS scales) for respondentschoosing the reason, "Poor health". Respondents choosing the reason "Can't afford it"had significantly lower means on the ER and TO scales. And respondents choosingthe reason "No transportation" had a significantly lower mean on the MO scale.There was not a significantly higher representation of males in the subsets ofrespondents choosing at least one enabling reason (30% male) or at least onereinforcing reason (26.5%). The proportion of males in these groups was higher thanthe proportion among participants (20.6%) but lower than the subset of nonparticipantschoosing at least one predisposing reason (38%).The predicted tendency for nonparticipant respondents in the lower functioninggroup to have less than high school education, while not significant, is present. In thesubsets choosing at least one enabling reason and at least one reinforcing reason,31.8% of respondents had less than high school education. Among participants, only21.7% had less than high school education. Among the group of nonparticipantschoosing at least one predisposing reason, 18.4% had less than high schooleducation.Age was not explicitly addressed in the hypotheses, but was, nevertheless, asignificant factor. Participants were significantly younger than nonparticipants.Nonparticipant respondents choosing at least one reinforcing reason (a group whichhad lower means on all ISAI scales than the nonparticipant group means) weresignificantly older than the rest of the nonparticipants. Nonparticipants choosing atleast one enabling reason had a mean age almost equivalent to the overallnonparticipant group mean (77.2 years compared to 77.1 years), but this is-134-complicated by the presence of "Not enough time", which is classified as an enablingreason although it exhibited a pattern more similar to the predisposing reasons. Thefour enabling reasons which show mean ISAI scale scores significantly lower than thenonparticipant group means ("No transportation", "Poor health", "Can't do exercise"and "Don't know about it") all show mean respondent ages above 80 years (86.4years, 81.3 years, 82.1 years and 82.0 years respectively).The reason "Don't know about it", chosen by 43 of the nonparticipants, had ISAIscale means approximately equal to the nonparticipant group means. The onlysignificant difference between this group and the total nonparticipant group was in theproportion who were involved in other Kerrisdale Senior Centre activities, which wassignificantly lower (only one of the 43 respondents). It was not, as hypothesisedassociated with low scores on the SS, CS or TO scales. The mean age ofnonparticipant respondents choosing this reason, at 75.8 years, is closer to theparticipant mean age of 74.8 years than the nonparticipant age of 77.1 years. It ispossible that lack of knowledge is not an enabling reason, but represents a fourthfactor which precedes the predisposing factor.The third hypothesis, that there would be a lower functioning subset ofnonparticipants as evidenced by lower ISAI means, who would choose reasons fornonattendance which would also represent barriers to their attendance, seems to besupported. Specific reasons chosen showed close agreement with significantly lowerscores on the associated health measure. For example, where physical health andmobility scores were low, which might target the individual as a good candidate for thehealth education program, poor health and an inability to do exercise were frequentreasons given for not attending. Where social support system scores were low, whichmight indicate that the individual would benefit from the supportive environmentprovided by the program, one frequently chosen reason for not attending was notknowing anyone there. Transportation was a reason provided for individuals with poor-135-mobility even within the 5-block radius canvassed in this study; lack of mobility mightbe an indicator for attendance at the program.TABLE 28:^MEAN AGE AND ISAI SCORES FOR PARTICIPANTS ANDFOR NONPARTICIPANTS BY REASONPARTICIPANTS NONPARTICIPANTSCHOOSING AT LEAST1 PRED. REASONNONPARTICIPANTSCHOOSING AT LEAST1 ENAB. REASONNONPARTICIPANTSCHOOSING AT LEAST1 REINF. REASONAge (years) 74.8 76.4 77.2 80.7Economic Res. 30.1 29.7 28.7 28.2Emotional Bal. 24.4 27.3 25.7 24.8Physical Health 24.4 23.2 20.9 20.1Trusting^Others 30.1 30.9 30.8 30.2Mobility 29 28.2 26.3 24Cognitive^Stat. 24.2 25.9 25.1 24.1Social Support 29.3 30.6 29.8 27.5CHAPTER 5DISCUSSION AND CONCLUSIONS5.0 LIMITATIONSThis was a cross-sectional survey at one point in time, and therefore, it is notpossible to determine whether significant factors such as higher mean scores of thePhysical Health and Mobility scales preceded the health education program (ie. arepredictive of use), or developed concurrently with or as a result of the program.The sample was limited to a small geographic area in Vancouver and it isexpected that there was systematic bias due to variables associated with thatparticular area. Since the study was carried out in the Kerrisdale area, which housesa relatively more wealthy and well educated population than other areas ofVancouver, the study may have systematically under-represented elderly individualswho do not have adequate income, are less educated and do not have transportationand social support resources. As income and education have been positivelyassociated with measures of health (Health and Welfare Canada, 1989), the studymay also have under-represented elderly individuals with health problems whichmight target them as ideal participants of the health education program.These same individuals, however, may have had a wider array of options, thusallowing this study to assess factors directly related to program participation due to thegreater array of optional activities open to respondents.The study was limited to comparison of the characteristics of participants andnonparticipants of one health education program. There may have been systematicbias due to factors unique to the program (structure of the program, connection of theprogram with the Senior Centre which operates it, characteristics of leaders andinstructors, etc.) which limit the generalisability of the results. The literature reviewedincluded characteristics of senior centre attenders and of health education attenders.The results of this study seem to be characteristic of both profiles of attenders-137-(increased measures of health, younger age, greater proportions of females), whichprovides some assurance of the generalisability of the study.In addition, the study was limited to a survey of community residents within afive-block radius of the Senior Centre out of which the health education programoperates. This was done to control for distance as a study variable. As such, theproportion of nonparticipants who do not attend due to mobility and transportationproblems may have been under-represented. There may also have been systematicdifferences in characteristics of Kerrisdale residents according to where within theneighbourhood they reside (eg. income, education, etc.). Even within this 5-blockradius, however, some respondents reported having transportation and monetarydifficulties, suggesting that these are problems for some elderly people which must beaddressed in programming.The sampling technique drew study participants from three sources: (1)participants of the health education program of interest, (2) apartment residents fromthe surrounding area and (3) single family dwelling residents from the surroundingarea. Place of residence was a potentially confounding variable, as no explicit attemptwas made to ensure that the proportions of seniors canvassed from each of thesesources was equivalent to proportions residing in the area, although a greater numberof apartment residents was targeted, in keeping with the large number of apartmentresidents in the area. Place of residence was included in the analysis, so that areaswhere it is a potential confounder were identified, and can be taken into considerationin interpreting results.There may be systematic differences among the potential study participantswho did not choose to participate. Such differences might have related to variablesthat were of interest in the study, such as reasons for attending or not attending theprogram (eg. lack of interest or lack of time), multi-dimensional measures of health (eg.cognitive status, level of trust) or levels of community involvement and time spent on-138-ADL (eg. were too busy to participate).The multi-dimensional measure of health utilised (Iowa Self-AssessmentInventory or ISAI) was developed and tested in the United States (Morris, Buckwalter,Cleary, Gilmer, Hatz and Studer, 1991). One question had to be changed to suit aCanadian population (number 50, "I use food stamps"). There may be otherdifferences between Canadian and American elderly people which systematically biasthe results of this study (eg. interpretation of questions, etc.). Piloting of thequestionnaire with a group of Kerrisdale seniors, however, provided assurance thatsuch bias did not exist: the questions were found to be understandable and deemedrelevant by the pilot respondents.The second part of the survey, which asked questions specific to participation inthe health education program and community involvement in the Kerrisdale area, wasdeveloped for this study and did not undergo rigorous testing of reliability and validity.Through scrutiny by experts (thesis committee members, health professionals andresearchers working with the senior population in Kerrisdale) and piloting on a smallgroup of Kerrisdale seniors, face and content validity were established. Associationsbetween the Kerrisdale-specific questions and the ISAI (ER mean associated with thereason "Can't afford it"; MO mean associated with the reason "Lack of transportation",for example) also support the validity of the second part of the questionnaire. Theremay, however, have been unidentified problems with the content and/or reliability ofthe questionnaire.Sample size calculations were performed based upon the data available fromthe ISAI standardisation (Morris, Andrews, Gilmer, Buckwalter, Cleary, Boutelle andHatz, 1991). Due to small numbers in some cells on the second half of thequestionnaire, however, sample size may not have been adequate to provide thepower necessary to capture differences among the study participants. For this reason,significant test values were reported at the p < .05 level, in spite of the large number of-139-tests run, in order to establish trends in community participation and patterns of ADL.Results which are marginally significant should be interpreted with caution, as some ofthem may have been significant due to chance.5.1 DISCUSSION RE METHODOLOGY ISSUES5.1.1 Return RatesLow return rates were observed among the Kerrisdale Senior Centre sample,and among portions of the apartment sample. This was contrary to what wasanticipated when the methodology was formulated; it was expected that the KerrisdaleSenior Centre sample, which consisted mainly of participants, would have the highestreturn rates due to committment to the program.The single family dwelling sample, which had the highest return rate, wasdifferent from the other two in that each residence that took part in the survey wasrevisited on a specified day for pick up of the completed survey. This was also true forsome of the apartment residents, and return rate was high in these cases as well.The apartment sample, however, had the option of returning their questionnaireto the manager, while the Kerrisdale sample had the option of returning theirquestionnaire to the Kerrisdale Senior Centre office.For the Kerrisdale Senior Centre sample, arrangements could have been madeto have all participants complete the survey at once and return it the same day.Refusal rates would probably have increased in this case. Many individuals at theSenior Centre were only willing to participate if they could take the survey home, asthey had other plans following their visit to the Centre. Therefore, even though returnrate might have been higher, the information gained would have been equally (orperhaps more) compromised by higher refusal rates. A benefit of this method wouldhave been the ability to provide assistance to participants who had difficultyinterpreting questions. For example, the problem with self-perception as a participantor nonparticipant of the Health Drop-In among exercise attenders might have been-140-avoided.Another option would have been to allow Kerrisdale Senior Centre participantsto take the survey home, as was done, but to record an address and arrange for pick-up of the survey at the participants' resident one or two days later. This would havemade the methodology for the Kerrisdale Senior Centre sample more similar to that ofthe single family resident sample, where return rates were the highest, while allowingthe same opportunity to choose an opportune time and place to complete thequestionnaire. There might also have been a higher refusal rate, but this could havebeen at least partially circumvented by explaining to participants that the recording ofaddresses would not compromise their anonymity.5.1.2 ISAI Trusting Others ScaleAs was noted in the Results section, many respondents seemed to object tosome of the items on the TO scale, particularly item 4 (People secretly say bad thingsabout me), item 11 (Friends are disloyal to me behind my back), item 18 (I believe I ambeing plotted against) and item 53 (I see things when others do not). This led to thelowest number of usable scores of any of the of the ISAI scales (172). It is not possibleto discern a particular reason for this discomfort level, as respondents were not askedfor any evaluative comments about the survey. It is possible, however, that some ofthe possible respondents among the Kerrisdale Senior Centre and apartment sampleswho did not return their questionnaires were also offended by these items. Perhapschanging the wording on the ISAI Directions (page 3 of the survey) to provide anexplanation of the scales would have partially avoided this problem. It may be that,where multi-dimensional measures of health are utilised with a general population,this is a general issue which must be considered.5.2 DISCUSSION OF RESULTS5.2.1 Place of Residence as a Possible ConfounderApartment residents were significantly more likely to be female and live alone,-141-and were significantly older, than residents of single family dwellings. Because placeof residence was an independent variable, this has the potential to confound theresults which compare participants to nonparticipants.With significantly more apartment dwellers in the participant group (if no truedifferences existed between participants and nonparticipants) the participants shouldhave been older. In fact, participants were significantly younger than nonparticipants,which is the opposite trend from what would be expected if there were no truedifferences between participants and nonparticipants. Even though there were moreapartment residents within the participant group, gender and living arrangement werenot significantly different. Mean scores on the ISAI scales were not significantlydifferent by place of residence, so it is unlikely that significantly different resultsbetween participants and nonparticipants on the scales are due to sampling bias orplace of residence.5.2.2 Demographic Characteristics and ParticipationParticipants of the Health Drop-In were significantly younger thannonparticipants. This finding is similar to findings from other community-basedprograms for seniors (Lalonde and Fallcreek, 1985; Schneider et. al., 1985), butcontrary to what is stated in the literature about target populations (Schneider et. al.,1985; Green and Gottlieb, 1989). At a mean of 74.8 years, however, the program isreaching people near or in the specified target age range (over 70 year, Anderson,1982; over 75 years, Schneider et. al., 1985).Level of education was not significantly different for participants andnonparticipants. The participant group consisted of more individuals with high schoolcompletion (48.3% compared to 40.5% for nonparticipants) and fewer individuals withless than high school (21.7% compared to 29.7% for nonparticipants). The proportionof respondents with more than high school education was equivalent between thegroups. Excluding the proportion of the study sample with post-secondary education,-142-the trend for more education among participants is consistent with the literaturereviewed (Schneider et. al., 1985; Benson et. al., 1989; Cox and Monk, 1989). Theliterature suggests, however, that programs should be targeted at individuals who areless educated (Schneider et. al., 1985; Green and Gottlieb, 1989) as less education iscorrelated with poorer health (Health and Welfare Canada, 1989).Although gender was not significantly associated with overall Health Drop-Inparticipation in this study, there were proportionately less males in the participantgroup (20.6%) than the nonparticipant group (31.6%). There also were somedifferences between program components, with males significantly less likely to attendthe exercise components (only 6.0% were male), but more likely (although notsignificantly) to attend the blood pressure monitoring component (26.9% were male).The proportion of participant males in this study was less than the 27% averagefound in five Vancouver seniors' wellness programs by Calsaferri (1990). Thenonparticipant proportion of males to females is closer than the participants' to thosefound in the two census tracts which most closely approximate the area canvassed inthis study (30.2% of individuals 55 years of age and older were male in census tract022, while 33.7% were male in census tract 009; Statistics Canada, 1987).The proportions of respondents living alone, with a spouse or with a relative orfriend were nearly identical for participants and nonparticipants in this study. Over halfof both groups lived alone and just over one third lived with a spouse. The similarity inproportions is compatible with the finding of Cox and Monk (1989), although theproportion living alone is lower (70% in the quoted study). Schneider et. al. (1985)found that a higher proportion of participants than nonparticipants lived alone, andBenson et. al. (1989) found that there were fewer married individuals in theirparticipant sample. The findings are also compatible with Statistics Canada (1987)census data, which determined that in 1986 approximately half the individuals 65years and older in this area lived alone, and between 41 and 44% were family-143-persons (ie. lived with a spouse).The demographic characteristics of this sample are similar to the findings inother studies of seniors' community-based health education programs. The programis not reaching older and less educated individuals who live alone, although these arevariables which the literature suggests should be targeted with such programs.5.2.3 Measures of HealthAs previously stated in the Limitations section, significant differences inmeasures of health cannot be attributed directly to the Health Drop-In due to the cross-sectional nature of the study.Nevertheless, the participants had significantly higher mean scores on self-rated measures of physical health and mobility. Perhaps this should be expected for agroup of respondents which is significantly younger, but self-rated scores on scales ofemotional balance, level of trust, cognitive status and social support were virtually thesame as, or lower than, those for nonparticipants. The participants were not, therefore,a younger group of elderly people who scored higher on all measures of health thanthe older nonparticipants; the two groups had distinctly different profiles.Schneider et. al. (1985) found, at base-line, that their participant group wasmore healthy, but also found that they were more likely to expect an interesting future,which would imply higher levels of mental health. The opposite trend was found byBenson et. al. (1989) (participants scored significantly lower on health measures, wereless satisfied with their lives and had less social interaction). Slivinske and Kosberg(1984) also found participants to be less healthy at base-line.Cox and Monk (1989) found in cross-sectional study that health educationattenders who had participated for at least 12 months had significantly lower ratings ofself-perceived health than the nonparticipants. This is in contrast to other studieswhich have found that measures of both physical and mental health increased forparticipants (Slivinske and Kosberg, 1984; Lalonde and Fallcreek, 1985).-144-The significantly higher scores in mobility found in this study are consistent withthe results found regarding participants of senior centres (Hanssen et. al., 1978; Krout,1983b; Krout et. al., 1988). But these studies also found less depression (Hanssen et.al., 1978) and more social interaction (Hanssen et. al., 1978; Krout, 1983b; Krout et.al., 1990), differing from the results of this study.Schneider et. al. (1985) suggested that programs should be directed atindividuals with low SES who have mobility problems. Low income (Green andGottlieb, 1989; Health and Welfare Canada, 1989), depression (Anderson, 1982) andsocial support (Anderson, 1982; Green and Gottlieb, 1989), because of theircorrelations with poorer health, are also considered to be characteristics which shouldbe targeted. The present study found a significant group of nonparticipants who fellinto the targeted categories (respondents choosing at least one enabling or reinforcingreasons for nonattendance). This group was often significantly older, and scoredlower (often significantly) on all of the above characteristics (mobility, economicresources, emotional balance and social support). The Health Drop-In did not seem tobe effectively reaching this group of nonparticipants. Moreover, the reasons that theseindividuals indicated for nonattendance were the same reasons why they should betargeted. A lower score on economic resources, for instance, was associated withchoice of "Can't afford it" as a reason for not participating. Lower scores on mobility,physical health and emotional balance scales, as well as increased age, wereassociated with choice of "Poor health" as a reason for lack of attendance. Lowermobility and physical health scores, along with increased age, were associated withchoice of "Can't do exercise". Finally, lower social support, cognitive status andemotional balance scores, along with increased age, were associated with choice of"Don't know any one there" as a reason for not attending.A second subset of nonparticipants, mainly those choosing at least onepredisposing reason, had higher scores on all measures of health and were younger.-145-These nonparticipants did not have the characteristics of older age, low economicstatus, poorer scores on the emotional balance scale, poorer health or less socialsupport which would target them as candidates for the health education program.They may have self-screened themselves, as the most common reasons given fornonattendance were "Don't need it" and "Not interested". Respondents choosing theenabling reason "Not enough time" had a similar profile to those choosingpredisposing ones (younger age and higher scores on the ISAI scales).The higher participant scores in measures of physical health and mobility foundin this study are similar to the findings among participants on post-tests in some of thelarge community-based studies. The results differ, however, in the finding ofequivalent or lower scores on measures of emotional health and social support.In addition, one group of nonparticipants was identified which has the targetcharacteristics identified in the literature for community-based health educationprograms. These respondents identified, as reasons for nonattendance, the samecharacteristics which identify them as part of the target population. A second group ofnonparticipants, which was similar to participants in age and measures of health, didnot have the target characteristics of health education programs identified in theliterature and, thus, may have self-screened themselves for other reasons not capturedin this study.5.2.4 Community ActivityA significantly higher proportion of participants than nonparticipants wereinvolved in other Kerrisdale Senior Centre activities, and a significantly higherproportion attended the symphony and similar events. Although not showing up assignificant (perhaps due to lack of power), there were also higher proportions ofparticipants involved in West Main Health Unit activities, church groups andorganisations, cultural centres and organisations, recreational activities and other-146-senior centres. Proportions of participants were not higher for attendance at church,the community centre and "other" activities.There is support in the literature for the increased involvement of participants insenior centres (Schneider et. al., 1985). Schneider and colleagues found the oppositetrend for church attendance, however, with more participants reporting churchinvolvement. Hanssen et. al. (1978) found, for senior centre participants, attendanceat more out-of-home activities and less passive activities than for nonparticipants.In this study, church attenders were found to be significantly older thatnonattenders, while those attending recreational activities or cultural organisationswere significantly younger. This may have influenced the community activity profile forparticipants and nonparticipants of the Health Drop-In, as participants were alsosignificantly younger. Educational attainment may also have had an influence, as43.5% of the church attenders had less than high school education, although amongthe total sample approximately 27% had less than high school. Recreational activitiesand symphony, etc. were significantly more attended by those with post-secondaryeducation.The tendency for less community involvement by nonparticipants thanparticipants in activities other than the Health Drop-In suggests that other communitysources are not providing nonparticipants with support in place of the Health Drop-In.5.2.5 Reported Hours of ActivityThe activity to which the largest amount of time was attributed, apart from sleep,was household activities, which tends to bear out the findings of Singleton et. al (1986)who found that the majority of time would be allotted to home-based activities. Gender(females reporting significantly more hours of household activity), age (olderindividuals reporting significantly fewer hours of household activity, exercise,community activity and gardening) and educational status (respondents with less thanhigh school education reporting significantly less hours of exercise) were related to-147-activity profiles. Gender differences were also found by Singleton et. al. (1986), anddifferences in gender, age and educational status were found by Spakes (1979).Residents of single family dwellings reported significantly more hours of householdactivity and gardening than apartment residents, as would be expected.Nonparticipants reported, on average, more total hours of activity per week thanparticipants. The areas which had higher mean scores were household activities,hobbies, gardening and sleep. The nonparticipant group had lower means for hoursof exercise, visiting friends and relatives and community activities (only communityactivities showed a significant difference). These results may be associated with thefinding of significantly lower ratings of physical health and mobility amongnonparticipants.The significantly lower mean for hours of community activity is consistent withthe trend found in reported community involvement for these respondents (withparticipants proportionately more involved in most community activities thannonparticipants).The lower mean for exercise hours (5.6 hours for participants versus 4.3 hoursfor nonparticipants) is interesting in light of the correlation between lack of regularexercise and declines in functional status (Mor et. al., 1989) and is consistent with thelower proportion of nonparticipants who are involved in recreational activities.Nonparticipant respondents choosing at least one enabling reason or at least onereinforcing reason had means lower than the nonparticipant group means for exercise(4.1 and 3.2 hours, respectively), community activity (1.4 hours and 0 hours,respectively) and visiting (5.0 hours and 4.3 hours respectively). This would suggestthat the subset of nonparticipants who belong to the target population for communityhealth education programs are not receiving physical activity and social support fromsources other than the Health Drop-In. These two groups (nonparticipant respondentschoosing at least one enabling reason and nonparticipant respondents choosing at-148-least one reinforcing reason) had higher means, however for hobbies ( 10.1 hours and9.2 hours, respectively) than either the nonparticipant group as a whole (8.2 hours) orthe participant group (5.2 hours). It would have been interesting to get more specificinformation of the nature of the hobbies; are they active or passive in nature?McKinnon's (1982) finding, with a sample of 1,398 Canadian elders, that the majorityof leisure activity time was spent on watching television would suggest that, should thisinformation have been requested, passive activities would have been more prevalent.5.2.6 Reasons for Attendance or NonattendanceThere was no literature found providing reasons for attendance at a community-based health education program. In this study, the reasons classified as predisposingwere most frequently chosen. These reasons were, with only 1 exception ("Toparticipate/volunteer in community work") related to health and functional status("Health is important to me", "Want to stay active", "Need the exercise", "Interested inlearning how to keep healthy" and "Want to continue living at home"). The threeoverall most frequently chosen reasons were among these.The frequently noted reasons in studies of senior centre participants reflectsome of the reasons chosen in this study. For example, "something to do" was chosenby 50% of participants in the Krout (1983b) study and "desire for outside interest by43% in the Trela and Simmons (1971) study. These reasons are similar to "Break inroutine", "Relief from boredom" and "Participate/volunteer in community work", chosenby 7 respondents each (12.7%) in this study. The reason "need for companionship,chosen by 12% of the Trela and Simmons (1971) sample, is similar to "To makefriends", chosen by 14 respondents (25.5%) in this study.Hanssen et. al. (1978) found senior centre participants to enjoy structuredcommunity activity more than nonparticipants. Sixteen respondents in this study(29%) chose the reason "To participate with a group". Among exercise participants,this percentage was significantly higher, at 34.3%.-149-It is also interesting that over one third of participants in the Trela and Simmons(1971) study and 25% in the Krout (1983b) study gave, as their reason for joining asenior centre, invitation or persuasion of friends, volunteer recruiter, etc. A higherpercentage of participants of the Health Drop-In in this study had gained theirknowledge of the program from a friend, whereas nonparticipants of the Health Drop-In were less likely to have heard about it from a friend and more likely to have readabout it. It may be that the personal contact is a significant force in determiningwhether individuals attend any community function, including a health educationprogram. If true, this would support the need for an outreach component to healthprograms, as suggested by Stuen (1985).The most frequently chosen reasons for not attending the Health Drop-In werevery similar to documented reasons for not attending senior centres. Excluding, for themoment, the reason "Don't know about it", the most frequently chosen reasons in thisstudy were "Not interested", "Don't need it", "Can't do exercise" and "Not enoughtime". Similar, or equivalent reasons found in studies of senior centre nonattenderswere "competing activities and interests" (Trela and Simmons, 1971); "too busy"(Krout, 1983b); "lack of time" (Ralston and Griggs, 1985); "lack of interest" (Krout,1983b; Ralston and Griggs, 1985) and "no need" (Krout, 1983b). "Would not enjoy it",chosen by 14 respondents in this study, was similar to "ambivalence towardorganisational activities" (Trela and Simmons, 1971) and "lack of interesting activities"(Ralston and Griggs, 1985). Ralston and Griggs (1985) did not find any enabling orreinforcing reasons which would also constitute barriers to attendance, but "lack oftransportation" was cited by 2% of the Krout (1983b) sample and 10% of nonattendersin the study by Trela and Simmons (1971). Eight respondents in this study (7.3%)cited "No transportation" as a reason for not attending the Health Drop-In, even withina 5-block radius of the Senior Centre. "Poor health" was cited as a reason by 8.6% inthe Trela and Simmons (1971) study, and was chosen by 9 (8.1%) of respondents in-150-this study.Among nonparticipant respondents in this study, choice of the reasons "Notinterested", Don't need it" and "Not enough time" were most predictive of high meanson all ISAI scales, with significantly higher mean scores than the rest of thenonparticipants on several. Respondents choosing "No transportation", "Can't affordit", "Poor health", "Can't do exercise", "Don't know anyone there" and "Don't like toleave home" had significantly lower means on one or more of the ISAI scales andwere older. Choice of "Poor health", "Can't do exercise" and "Don't know anyonethere" were most predictive of lower mean scores on ISAI scales and increased age.Forty-one respondents gave, as a reason for not attending, "Don't know aboutit". This represents 37.3% of the nonparticipant respondents who provided reasons fornot attending, and tends to bear out the finding of Krout (1983) that there is a low rateof knowledge among the public of services they can access.5.3 IMPLICATIONS OF RESULTSGreen and Kreuter's (1991) framework, which provides three categories ofmotivational factors for an individual's participation in a health education program, wasused to group reasons for attendance and reasons for nonattendance at the HealthDrop-In. The different patterns that seem to have been obtained, in terms of age andISAI score profiles, for respondents choosing at least one predisposing, enabling orreinforcing reason, provide partial support for Green and Kreuter's (1991) model.Some reasons, however, might have been more effectively grouped. Forexample, "Not enough time" was classified as an enabling reason. The logic followedinvolved consideration of time as a tangible resource; if an individual did not haveenough time they would not be able to participate. It could also be true, however, thatthe individual perceived that he/she did not have enough time to attend because theHealth Drop-In was relatively low on the priority list of activities they wished toparticipate in. If this were the case, "Not enough time" should properly be classified as-151-a predisposing reason. The trend for higher than average mean scores on the ISAIscales found in this study might indicate that more individuals who chose "Not enoughtime" had a predisposing, rather than an enabling, motivation for doing so.Nearly 40% of the nonparticipant respondents stated that they did not attendbecause they did not know about the program. If the area which was surveyed hadbeen widened to include more than a 5-block radius around the Senior Centre, thispercentage would most likely have been higher. In addition, it is conceivable thatknowledge of community resources is higher than the norm for Vancouver in a moreeducated area such as Kerrisdale (Statistics Canada, 1987). Given that contact withindividuals (friends, recruiters, etc.) is a commonly chosen reason for attending seniorcentres (Krout, 1983b; Trela and Simmons, 1971), to increase knowledge of theHealth Drop-In and similar programs in a manner which will translate into increasedattendance, a word of mouth campaign may be one plausible strategy.The 7.3% of respondents who chose "No transportation" as a reason for notattending is notable, particularly since surveys were only distributed in a 5-blockradius around the senior centre. The percentage is between the 2% found by Krout(1983b) and 10% found by Trela and Simmons (1971) as a reason for nonattendanceat senior centres. If surveys had been distributed throughout the area which is servedby the Kerrisdale Senior Centre, this percentage would possibly have been higher.The subset of respondents who chose lack of transportation as a reason wassignificantly older and significantly less mobile than the rest of the nonparticipants.According to the literature reviewed, therefore, they probably represent one group ofindividuals who should be targeted by community health education programs(Schneider et. al., 1985; Jacks, 1975; Krout, 1983). Perhaps the use of volunteerescorts or drivers could be employed by a program such as the Health Drop-In toreach these individuals.As was pointed out by the Kerrisdale Senior Centre Program Coordinator, the-152-costs involved in participating in the Health Drop-In are not prohibitive ($4 or 7 for amembership, depending on age, and $9 for 12 weeks of exercise), but 3.6% ofindividuals still chose "Can't afford it" as a reason for nonattendance. It may be thatthe respondents who chose this reason were not informed about the low cost of theprogram. It may also be that, on a fixed income, even the small amount of moneyrequired is more than some individuals can afford. There is a difference between networth and cash flow, and in 1986, one of the census tracts in this area had a medianannual income which was lower than the city's median annual income (StatisticsCanada, 1987). In addition, if one were to look at other similar programs in lessaffluent areas of Vancouver, cost would probably be an even greater deterrent toattendance. An advertised policy of not charging individuals who are unable to paymight allow these individuals to participate.Approximately 9% of nonparticipant respondents chose "Poor health" as areason for not attending and 12.7% chose "Can't do exercise". Choice of thesereasons may be partially due to lack of awareness of the various programcomponents, as "Poor health" is actually a possible reason for attending the BloodPressure Monitoring component, and neither the Blood Pressure Monitoringcomponent nor the Neck/Shoulder Massage involve exercise. Advertising the differentarray of program components might draw some of these individuals into the program.The respondents who chose these two responses may also be unaware of theGentle Fit exercise program, which caters to less mobile participants. Any strategy topromote the Gentle Fit program would require expansion of the exercise program,however, due to the current full enrollment.Another possibility is that respondents who selected these two responses havemisperceptions about the value of exercise for elderly individuals, or are unaware ofthe ability of older people to participate in exercise (Health and Welfare Canada, 1989;Larson, 1988). If this is the case, it implies the need for a more far-reaching-153-educational campaign (for example, through the media) which attempts to reverse thebeliefs that exercise is not good or necessary for older people (Health and WelfareCanada, 1989), and explains the benefits and content of a regular exercise program.The response "Don't know anyone there", chosen by 17.3% of thenonparticipant group, and accompanied by a significantly higher mean age along withsignificantly lower scores on scales for emotional balance, mobility and social support,suggests that some of these respondents may be socially isolated and depressed.This seems to be a group of individuals who would benefit from a program such as theHealth Drop-In, but would require a personal contact to allow them to attend. In onestudy, over one-third of participants in a senior centre attributed their attendance to useof volunteer recruiters (Trela and Simmons, 1971), which suggests that a similarstrategy employed in the case of the Health Drop-In might help to provide a link forindividuals who find themselves unable to participate because they don't know anyone there.This study found that there is a group of nonparticipants who might benefit fromthe Health Drop-In but have barriers to participating such as lack of mobility, lack oftransportation, limited income, poor health or no contact with individuals in the Drop-In.These characteristics also target them as potential participants of the program (Jacks,1975; Krout, 1983; Schneider et. al, 1985). The results of this study support the view ofStuen (1985), that to reach these respondents, and others with the samecharacteristics, would involve direct outreach in some fashion.5.4 SUGGESTIONS FOR FURTHER RESEARCHThe trends documented in this study suggest that further research to documentthe characteristics of elderly health education participants over wider regions would bejustified. Such research would be helpful in understanding what factors are predictiveof senior health education program participation. It could determine whether or not the-154-trends identified in this study are generalisable beyond the Kerrisdale area, and wouldprovide the increased power necessary to illuminate further additional variables whichdid not provide significant results in this study, such as areas of communityparticipation. An understanding of the factors which result in the decision to participatein health education programming would be useful at the policy and program planninglevel in determining strategies for recruitment and priorising funding alternatives. Itwould also be useful for program evaluation, by providing information which is usefulin determining whether or not the target population is being effectively reached, and ifnot, how to modify the program to achieve this objective.In addition to the information requested in this study, any similar research mightwish to ask respondents to specify what forms of exercise and hobbies they engage in.A question which determined knowledge and practice of various health behaviors (eg.knowledge of exercise and nutritional benefits, nutritional habits, smoking behavior,alcohol intake, etc.) would also be useful to correlate with participation behavior. Suchquestions might be seen as threatening by some respondents in a general population,however, and decrease response rate.The trends noted in this study among respondents choosing predisposing,enabling and reinforcing reasons for attendance and nonattendance suggest thatlarger studies building upon this information could be helpful in understanding thesethree factors. Particularly among the nonparticipants, there were identifiabledifferences in trends between these factors. A study with more respondents, whichgenerated larger numbers for each of the reasons, would allow the reasons to besubjected to differential analysis. This would help to determine which of the reasonswere best classified in each of the predisposing, enabling and reinforcing categories(eg. would determine whether "Not enough time" was more consistent withpredisposing or enabling factors) and would also identify any factors which did notbelong with any of these categories (eg. might help to determine whether "Don't know-155-about it" is a fourth factor which precedes the predisposing factors). A more completeunderstanding of the reasons why people do and do not attend programs could begleaned from such research. This would, in turn, aid in determining recruitment andintervention strategies for the target population.Table 29 provides a comparison of possible characteristics of participants andnonparticipants who, according to the literature, do or do not belong to the targetpopulation for health education programs. Such characteristics include age, lowersocio-economic status, chronic health conditions and social isolation. The summary inTable 29 is based upon the findings of this study, and includes both the significantassociations and nonsignificant tendencies that were noted. As was noted in theResults section, the mean is not necessarily an appropriate analytical measure fordata from this study, as the distribution of scores for respondents was highly skewed insome instances. Nevertheless, it provides a means of summarising the patterns noted.The characteristics listed in the tables are not definitive, but could prove to be helpfulin generating hypotheses for future research.Research aimed at determining peoples' reasons for attendance andnonattendance, and classifying these reasons using the predisposing-enabling-reinforcing framework, need not be limited to seniors' health education programs. Itwould be interesting to see whether the reasons which were most common in thisstudy are also common with other age groups and other types of health programming.For example, it would be reasonable to expect that "Don't know anyone there" wouldbe a barrier to attendance for younger as well as older people, and for many types ofcommunity-based activities. Green and Kreuter's (1991) framework is likelygeneralisable well beyond health education and promotion programming, andpotentially generalisable to organisational behavior in settings other than health (eg.social and leisure settings, employment settings, etc.).-156-TABLE 29: Comparison of Selected Characteristics Between: Column A: Participants and NonparticipantsBelonging to the Target Group^ (P v. NPTG) Column B: Participants and Nonparticipants Not Belonging to theTarget Group (P v. NPTG) Column C: Nonparticipants Belonging and not Belonging to the Target GroupVariable (Means)Column AP v. NPTGColumn BP v. NPTGColumn CNPTG v. NPTGAge (years) Younger Younger OlderEmot. Bal. (ISAI score) Similar* Lower LowerPh. Health (ISAI score) Higher Higher LowerMobility (ISAI score) Higher Higher LowerCog. Status (ISAI score) Similar* Lower LowerSocial Support (ISAI score) Higher Lower LowerHome-based Activity(self-reported hours) Fewer Fewer FewerExercise (self-report hrs.) More Similar* FewerCommunityActivity (self-report hrs.) More More Fewer* Similar: no significant difference between the means of the two groups^Target Populations: as defined in the literature (higher age, lower SES, social isolation, chronic healthproblems)-157-REFERENCESAgostino, J.N. (1985). Subjective Rating of Health Among the Elderly Activities,Adaptation and Aging, 6(4): 53-62.Anderson, W.F. (1982). Is Health Education for the Middle-Aged and Elderly a Wasteof Time? In Wells, T. (ed.). Aging  and Health Promotion Aspen Systems Corporation.Rockville, Maryland. 101-110.Barr, G. (1983). Educators Play Crucial Role In Promoting Healthy Aging. HealthEducation, 21(3): 2-7.Bennett, R., Killeffer, E.H.P. (1989). Concluding Remarks and Suggested Issues forFurther Discussion.  ^andDc ^Therapy in Geriatrics, 8(1/2): 147-58.Benson, L., Nelson, E.C., Napps, S.E., Roberts, E., Kane-Williams, E., Salisbury, Z.(1989). Evaluation of the Staying Healthy After Fifty Educational Program: Impact onCourse Participants. Health Education Quarterly 16(4): 485-508.Bergner, M. (1985). Measurement of Health Status. MadicaLCare, 23(5): 696-704.Bergner, M., Kaplan, R.M., Ware, J.E. (1987). Evaluating Health Measures.Commentary: Measuring Overall Health: An Evaluation of Three ImportantApproaches. Journal of Chronic Disease, 40(Suppl. 1): 23S-26S.Bergner, M., Bobbitt, R.A., Carter, W.B., Gilson, B.S. (1981). The Sickness ImpactProfile: Development and Final Revision of a Health Status Measure. Medical Care, 19(8): 787-805.Blakenbaker, R.G. and Moore, A.T. (1983). Health Promotion and Disease Prevention:The Second Public Health Revolution. In Landry, F. (ed.). Health Risk Estimation, Risk Reduction and Health Promotion. Canadian Public Health Association. Ottawa.Boshier, R. (1991). Psychometric Properties of the Alternative Form of the EducationParticipation Scale. Adult Education Quarterly,41(3): 150-67.Boshier, R., Riddell, G. (1978). Education Participation Scale Factor Structure forOlder Adults. Adult Education 28(3): 165-75.Breslow, L. (1983). The Risk Factor Approach to Chronic Disease Prevention-- SomeUpdating. In Landry, F. (ed)) Health Risk Estimation, Risk Reduction and Health Promotion. Canadian Public Health Association. Ottawa.British Columbia Minister of Health and Minister Responsible for Seniors (1989).- -^- - • :^•^II • - 1-• - II^•^leVictoria, B.C.-158-British Columbia Ministry of Health, Division of Community and Family Health (1989).Healthy Communities' She Process  Ministry of Health. Victoria. 27 pp.British Columbia Research (1986) Community Based Planning for Seniors: A Review of the Literature, Vol 1  B.C. Research. Vancouver.British Columbia Royal Commission on Health Care and Costs (1991). Part B:Organisation, Public Policies for Health. Closer to Home. Crown Publications, Inc.Victoria, B.C. pp B-2-B-8.Butler, R.N., Gellman, J.S., Oberlander, D.L., Schindler, L. (1979-80). Self-Care, Self-Help, and the Elderly. I. •l.. • it a• -10 • . n-a •• ia - a 1(1):95-119.Butler, R.N. (1975). Psychiatry and the Elderly: An Overview AmericanJournaLotPsychiatry, 132(9): 893-900.Calsaferri, K. (1990). Communityliealth Promotion ProgramalorSeniorsi_Program_Eocus_ancLContributingfactors to Composition. Unpublished M.Sc. Thesis.University of British Columbia. 208 p.Canadian Nurses Association (1989). HealthCareSiefonmforSeniors, CanadianNurses Association. Ottawa.Caplan, R., Holland, R. (1990). Rethinking Health Education Theory. Health_Education JournaL49(1): 10-12.Cox, C., Monk, A. (1989). Measuring the Effectiveness of a Health Education Programfor Older Adults. Educational Gerontology 15: 9-23.Cox, C., Monk, A. (1990). Integrating the Frail and Well Elderly: The Experience ofSenior Centers. Journal of Gerontological Social Work, 15:  131-47.Dorevitch, M.I., Cossar, R.M., Bailey, F.J., Bisset, T., Lewis, S.J., Wise, L.A.,MacLennan, W.J. (1992). The Accuracy of Self and Informant Ratings of PhysicalFunctional Capacity in the Elderly Jo_umaLoLainicaLEpidemiology,  45(7): 791-98.Downie, R.S., Fyfe, C. Tannahill, A. (1990). Health Promotion Models and Values Oxford University Press. Oxford.Duffy, M.E., MacDonald, E. (1990). Determinants of Functional Health of OlderPersons. The Gerontologist, 30(4): 503-9.Duncan, D.F., Gold, R.S. (1986, May/June). Reflections: Health Promotion-- What IsIt?. Health Values 10(3):  47-48.-159-Eng, K., Emlet, C.A. (1990). SRx: A Regional Approach to Geriatric MedicationEducation. The Gerontologist 30(3): 408-10.Ernst, M., Ernst, N.S. (1984). Functional Capacity. In Mangen, D.J. and Peterson, W.A.(eds.). Research Instruments in Social Gerontology, Vol 3' Health, Program Evaluationand_Demografft_ University of Minnesota. 9-84.Estes, C.L., Fox, S., Mahoney, C.W. (1986). Health Care and Social Policy: HealthPromotion and the Elderly. In Wellness and Health Promotion for the Elderly.Dychtwald, K. (editor), Aspen Systems Corp., Rockville, Maryland. 55-70.Eustis, N.N., Patten, S.K. (1984). The Effectiveness of Long-term Care. In Mangen,D.J. and Peterson, W.A. (eds.). Research Instruments in Social Gerontology, Vol 3 .Health_ Program Evaluation and Demography_ University of Minnesota. 217-316.Fallcreek, S.J., Allen, B.P., Halls, D.M. (1986) Health_Promotion_and_AgingL_A_National Directory of Selected Programs. Institute for Gerontological Research andEducation. New Mexico State University.Fallcreek,S., Warner-Reitz, A., Mettler, M. (1986). Designing Health PromotionPrograms for Elders. In WellnessancttlealthPromotion_forlheElderty_ Dychtwald, K.(editor), Aspen Systems Corp., Rockville, Maryland. 219-234.Feinstein, A.R., Josephy, M.S., Wells, C.K. (1986). Scientific and Clinical Problems inIndexes of Functional Disability. Annal_sof internal Medicin_e_195: 413-420.Feldman, D.S., Hofmann, R., Gagnon, J., Simpson, J. (1987). Statview II . Statview SE +__Grap_hica. Abacus Concepts, Inc. Berkeley, Ca.Fillenbaum, G.G., Smyer, M.A. (1981). The Development, Validity, and Reliability ofthe OARS Multidimensional Functional Assessment Questionnaire. Journal of Gerontology 36(4): 428-34.Fitch, V.L., Slivinske, L.R. (1988 Winter). Maximizing Effects of Wellness Programs forthe Elderly. Healtirand Social Work 61-67.French, J. (1990). Boundaries and Horizons, The Role of Health Education WithinHealth Promotion. Health_Education_dournal.,_49(1): 7-9.Fries, J.F., Green, L.W., Levine, S. (1989, March 4). Health Promotion and theCompression of Morbidity The Lancet 481-83.Gilbert, S.B. (1986, May/June). Health Promotion for Older Americans. Health Values_1Q(3): 38-46.-160-Gilmer, J.S., Cleary, T.A., Lu, D.F., Morris, W.W., Buckwalter, K.C., Andrews, P.Boutelle, S., Hatz, D.L. (1991). The Factor Structure of the Iowa Self AssessmentInventory. EducationaLand Psychological Measurement, 51(2): 365-75.Green, L.W. (1974). Toward Cost-Benefit Evaluations of Health Education: SomeConcepts, Methods and Examples Health Education Monographs, 2 (Suppl 1): 34-64.Green, L.W. (1978). Determining the Impact and Effectivenenss of Health Educationas it Relates to Federal Policy. liealth_Echication_Monographs,_6 (Suppl 1): 28-66.Green, L.W. (1992). Unpublished personal communications. University of BritishColumbia. Vancouver.Green, L.W., Kreuter, M.W., Deeds, S.G., Partridge, K.B. (1980) Health Education Planning:ADiagnastic_Appmach.. Mayfield Publishing Co. Palo Alto, Ca.Green, L.W., Gottlieb, N.H. (1989). Health Promotion for the Aging Population:Approaches to Extending Life Expectancy. In Andreopoulos, S., Hogness, J.R. (eds.).Health Ca -^•^Churchill Livingstone. New York. 139-57.Green, L.W., Kreuter, M.W. (1991) HealthPromotion_Plannincy_AmEducationaLand_Environmental Approach  Mayfield Publishing Co. Mountain View, Ca. 2nd Ed.Gurland, B., Kuriansky, J., Sharpe, L., Simon, R., Stiller, P. Birkett, P. (1977-78). TheComprehensive Assessment and Referral Evaluation (CARE)-- Rationale,Development and Reliability InternationallournaLotAging and Human Development_8(1): 9-42.Hanssen, A.M., Meima, N.J., Buckspan, L.M., Henderson, B.E., Helbig, T.L., Zarit, S.H.(1978). Correlates of Senior Center Participation. The Gerontologis8(2): 193-199.Health And Welfare Canada (1989) The Active Health Report on Seniors.  Minister ofSupply and Services. Ottawa.Health and Welfare Canada (1987) The Active Health Report: Perspectives on Canacta'slieattliPromationaurvey_19_85_ Minister of Supply and Services Canada.Ottawa.Institute for Health Care Facilities of the Future (1988). A ging: Future Health Care Delivery. Institute for Health Care Facilities of the Future. Ottawa.Jacks, M. (1975 April). Changing Perspectives Under the Older Americans Act. InOsterbind, C.C. (ed.) SociaLGoals_aocialProgramsancLtheAging_ Published for theCenter for Gerontological Studies and Programs. University Presses of Florida.Gainesville. 9-17.-161-Jackson, C., Altman, D.G., Howard-Pitney, B., Farquhar, J.W. (1989). EvaluatingCommunity-Level Health Promotion and Disease Prevention Interventions. InBraverman, M.T. (ed.) Evaluating_health Promotion Programs.Jossey-Bass, Inc. SanFransisco. 19-32.Janz, N.K., Becker, M.H. (1984 Spring). The Health Belief Model: A Decade Later.Health Education Quarterly 11(1): 1-47.Jette, A.M. (1980). Health Status Indicators: Their Utility in Chronic-DiseaseEvaluation Research. Journal of Chronic Disease, 33* 567-79.Kane, R.A., Kane, R.L. (1981). Assessing the Elderly._ The Rand Corp. Lexington,Massachusetts.Kane, R.L. (1987). Commentary: Functional Assessment Questionnaire for GeriatricPatients--Or The Clinical Swiss Army Knife. Journal of Chronic Disease, 40(Suppl. 1):95S-98S.Kearney, P., Plax, T.G., Lentz, P.S. (1985). Participation in Community Organisationsand Socioeconomic Status as Determinants of Seniors' Life Satisfaction Activities, Adaptation and Aging(4): 31-37.Kemper, D. (1986). The Healthwise Program: GROWING YOUNGER. In Wellness and Health Promotion for the Elderly_ Dychtwald, K. (editor), Aspen Systems Corp.,Rockville, Maryland. 263-274.Kerrisdale Senior Centre Policy Manual. (1987). Kerrisdale Senior Centre.Vancouver.Kerrisdale Senior Centre Promotional Pamphlet. (1991). Kerrisdale Senior Centre.Vancouver.Kimble, C.S., Longe, M.E. (1989). Health Promotion Programs for Older Adults' A Planningand_Mana.gement_Guide_ American Hospital Publishing, Inc., Chicago.Krout, J.A. (1983). Knowledge and Use of Services by the Elderly: A Critical Reviewof the Literature International Journal of Aging and Human Development 17(3): 153-67.Krout, J.A. (1983b). Correlates of Senior Center Utilization. Research on Aging, 5(3):339-52.Krout, J.A. (1988). The Frequency, Duration and Stability of Senior CenterAttendance. Journal of Gerontological Social Work, 13(1/2): 3-19.Krout, J.A., Cutler, S.J., Coward, R.T. (1990). Correlates of Senior CenterParticipation: A National Analysis. The Gerontologist, 30(1): 72-79.-162-Labonte, R.N., Penfold, P.S. (1981). liealthPromation_Philosophy: From Victim-BlaminVoSaciaLliesponsibility_ Western Region Office, Health PromotionDirectorate, Health and Welfare Canada.Lalonde, B.I.D., Fallcreek, S.J. (1985). Outcome Effectiveness of the WallingfordWellness Project: A Model Health Promotion Program for the Elderly. Journal of Gerontological Social Work 9(1): 49-64.Lalonde, B., Hooyman, N., Blumhagen, J. (1988). Long-Term Outcome Effectivenessof a Health Promotion Program for the Elderly: The Wallingford Wellness Project.dounaaLotGerontologicaLSociaLWork_13(1/2): 95-112.Lalonde, M. (1974). A New Perspective on the Health of Canadians. National Healthand Welfare, Canada. Ottawa.Landry, F. (ed.) (1982). Health Risk Estimation, Risk Reduction and Health Promotion Canadian Public Health Association. Ottawa.Larson, E.B. (1988). Health Promotion and Disease Prevention in the Older Adult.Geriatrics, 43(Suppl.): 31-39.Lawton, M.P., Moss, M., Fulcomer, M., Kleban, M.H. (1982). A Research and ServiceOriented Multilevel Assessment Instrument JournaLotGerontolow_37(1): 91-99.Leanse, J. (1986). The Senior Center as a Wellness Center. In Wellnessandl-lealth_Promotion for the Elderly  Dychtwald, K. (ed.), Aspen Systems Corp., Rockville,Maryland. 105-118.Lederman, S., Farrar, M. (1986). The Wisdom Project of the American Red Cross inGreater New York: A Blueprint for a Community-Based Health Care and HealthEducation Program. In Wellness and Health Promotion for the Elderly Dychtwald, K.(editor), Aspen Systems Corp., Rockville, Maryland. 247-262.Maiman, L.A., Becker, M.H. (1974). The Health Belief Model: Origins and Correlatesin Psychological Theory. Health Education Monographs, 2(4): 336-53.Marshall, V.W. (1980). Aging in Canadal_Social Perspectives Fitzhenry andWhiteside, Ltd. Don Mills, Ont.Martin, S. (1986)Development Vancouver Health Department. Vancouver.Martin, S., Robertson, A., Altman, J. (1988 February). A Framework for Health Promotion: Older Adults. Vancouver Health Department. Vancouver. 11 p.• • I • I^1 • I .^ • 11 11 1-163-MacRae, P.G. (1986). The Effects of Physical Activity on the Physiological andPsychological Health of the Older Adult: Implications for Education. In Peterson, D.A.,Thornton, J.E., Birren, J.E. (eds.) Education_and Aging. Prentice-Hall, Inc., EnglewoodCliffs, New Jersey. 205-230.McGinnis, J.M. (1982). Future Directions of Health Promotion. In Taylor, R.B., Ureda,J.R., Denham, J.W. Health_Prond_ClinicaLApplications Appleton-Century-Crofts. East Norwalk, Conn.McLeroy, K.R., Bibeau, D., Steckler, A., Glanz, K. (1988). An Ecological Perspective onHealth Promotion Programs. Health Education Quarterly, 15(4): 351-377.McKinnon, A.L. (1992). Time Use for Self Care, Productivity, and Leisure AmongElderly Canadians. Canadian Journal of OccupationaLTherapy,-5_9(2): 102-110.Meeks, S., Johnson, S. (1988). Health Promotion at a Senior Center: Designing theIdeal Program. dournaLofGerontologicaLSocial_Work_13(1/2): 21-35.Minkler, M., Pasick, R.J. (1986). Health Promotion and the Elderly: A CriticalPerspective on the Past and Future. In Wellness_and_HealthPromotioniortheEldedy._Dychtwald, K. (editor), Aspen Systems Corp., Rockville, Maryland. 39-54.Mitic, W. (1985 February/March). Nutrition Education for Older Adults: Implementationof a Nutrition Instruction Program. Health Education 15* 7-9.Moneyham, L.L., Rider, M.S. (1988 September/October). Nursing's Role In PromotingHealth For Elders In The Community. Geriatric Nursing: 282-3.Mor, V., Murphy, J., Masterson-Allen, S., Willey, C., Razmpour, A., Jackson, M.E.,Greer, D., Katz, S. (1989). Risk of Functional Decline Among Well Elders. Journal of Clinical Epidemiology, 42(9): 895-904.Morris, J.N., Sherwood, S., Mor, V. (1984). An Assessment Tool for Use in IdentifyingFunctionally Vulnerable Persons in the Community. The-Gerontologist,_24(4): 373-79.Morris, W.W., Boutelle, S. (1985). Multidimensional Functional Assessment in TwoModes. The Gerontologist, 25(6): 638-43.Morris, W.W., Buckwalter, K.C. (1988). Functional Assessment of the Elderly: TheIowa Self-Assessment Inventory. In Waltz, C.F., Strickland, O.L. (eds.). Measurement of Nursing Outcomes, Vol. 1., Measuring Client Outcomes Springer Publishing Co.New York. 328-51.Morris, W.W., Buckwalter, K.C., Cleary, T.A., Gilmer, J.S., Hatz, D.L., Studer, M. (1989).Issues Related to the Validation of the Iowa Self-Assessment Inventory. Educational_and Psychological Measurement, 49: 853-61.-164-Morris, W.W., Buckwalter, K.C., Cleary, T.A., Gilmer, J.S., Hatz, D.L., Studer, M. (1990).Refinement of the Iowa Self-Assessment Inventory. TheGerontalocItst30(2): 243-8.Morris, W.W., Andrews, P.H Gilmer, J.S., Buckwalter, K.C., Cleary, T.A., Boutelle, S.,Hatz, D.L. (1991). 1- • I . - 11-1 -11 it ►l -1 - © WoodrowW. Morris and Kathleen C. Buckwalter, 1985. University of Iowa.Mullen, K.D. (1986). Wellness: The Missing Concept in Health PromotionProgramming for Adults. Health Values 19(3): 34-37.Neugarten, B.L., Havighurst, R.J., Tobin, S.S. (1961). The Measurement of LifeSatisfaction dournaLotGerontology_16: 134-43.O'Neill, K., Reid, G. (1991). Perceived Barriers to Physical Activity by Older Adults.Canadian_dournaLoLPublic±lealti:482: 392-6.Osterbind, C.C. (ed.) (1975) Social Goals, Social Programs and the Aging._ UniversityPresses of Florida. Gainesville.Pannill, F.C. (1991). A Patient-Completed Screening Instrument for FunctionalDisability in the Elderly. TheAmericarblournal of Medicine,  90: 320-27.Parkerson, G.R., Gehlbach, S.H., Wagner, E.H., James, S.A., Clapp, N.E., Muhlbaier,L.H. (1981). The Duke-UNC Health Profile: An Adult Health Status Instrument forPrimary Care. Medical Care, 19(8): 806-828.Pearlman, R.A. (1987). Development of a Functional Assessment Questionnaire forGeriatric Patients: The Comprehensive Older Persons' Evaluation (COPE). Journal of Chronic Disease, 40(Suppl. 1): 85S-94S.Peterson, D.A. (1986). Aging and Higher Education. In Peterson, D.A., Thornton, J.E.,Birren, J.E. (eds.) Educatioaand_Aging,_ Prentice-Hall, Inc. Englewood Cliffs, NewJersey. 30-61.Peterson, D.A. (1990). A History of the Education of Older Learners. In Sherron, R.H.and Lumsden, D.B. (eds.) Introduction to Educ  tional Gerontology. 3rd Edition.Hemisphere Publishing Corp. New York.Pfeiffer, E. (1975). Multidimensional Functional Assessment: The Oars MethodologyCenter for the Study of Aging and Human Development. Durham, N.C. 134 p.Pirie, G. (1991-2). Personal communications and interviews regarding the operationof the Kerrisdale Senior Centre. Kerrisdale Senior Centre. Vancouver.Radecki, S.E., Cowell, W.G. (1990). Health Promotion for Elderly Patients. Family Medicine, 22(4): 299-302.-165-Rakowski, W. (1986). Research Issues in Health Promotion Programs for the Elderly.In Wellness and Health Promotion for the Elderly_ Dychtwald, K. (editor), AspenSystems Corp., Rockville, Maryland. 313-326.Ralston, P.A., Griggs, M.B. (1985 Fall). Factors Affecting Utilization of Senior Centers:Race, Sex, and Socioeconomic Differences. ilaurnaLoLGerolitologicaLSociaLWork,_9(1): 99-111.Read, J.L., Quinn, R.J., Hoefer, M.A. (1987). Measuring Overall Health: An Evaluationof Three Important Approaches. Journal of Chronic Disease,A0(Suppl.  1): 7S-21S.Rimer, B., Keintz, M.K., Glassman, B., Kinman, J.L. (1986). Health Education for OlderPersons: Lessons from Research and Program Evaluations. In Ward. W.B., Salisbury,Z.T., Zapka, J.G., Kar, S.B. (eds). AdvaiacesimlieattliEduciationand  Promotion  Vol. 1,Part B. Jai Press. Greenwich, Conn. 369-96.Rosenstock, I.M. (1974). Historical Origins of the Health Belief Model HealthEducatio_n_Monographs 2(4): 328-35.Rubinson,L., Alles, W.F. (1984). Introduction and Editors Remarks to Chapter 3, HealthPromotion. Health_Ed cations Foundations for the FutuLe Times Mirror/MosbyCollege Publishing. St. Louis, Missouri. p. 63, 87.)Rubinson, L., Alles, W.F. (1984). Editors Remarks. In Lussier, R. History of HealthEducation.^ ' • ns for the Future_ Times Mirror/MosbyCollege Publishing. St. Louis, Missouri. p. 27-29.Schneider, M.J., Chapman, D.D., Voth, D.E. (1985). Senior Center Participation: ATwo-Stage Approach to Impact Evaluation. The Gerontologist 25(2): 194-200.Shirreffs, J.A. (1984). The Nature and Meaning of Health Education. In Rubinson, L.,Alles, W.F. (eds.)±lealttEchicationlationsforthehiture_ Times Mirror/MosbyCollege Publishing. St. Louis, Missouri. p.35-53.Singleton, J.F., Mitic, W., Farquharson, J. (1986). Activity Profile of Retired Individuals.Activities, Adaptation and Aging, 9(1): 17-24.Simmons,J.J., Roberts, E., Nelson, E.C. (1986). The Dartmouth Self-Care for SeniorCitizens Program: Tools, Strategies, and Methods. In Wellness and Health Promotion for the Elderly Dychtwald, K. (editor), Aspen Systems Corp., Rockville, Maryland. 235-246.Simmons, J.J., Nelson, E.C., Roberts, E., Salisbury, Z.T., Kane-Williams, E., Benson, L.(1989 Winter). A Health Promotion Program: Staying Healthy After Fifty Health Education Quarterly 16(4): 461-72.-166-Slivinske, L.R., Kosberg, J.I. (1984). Assessing the Effect of a Personal HealthManagement System Within Retirement Communities: A Preliminary Investigation.TheGerontologist,24(3): 280-85.Spakes, P.R. (1979). Family, Friendship and Community Interaction as Related to LifeSatisfaction of the Elderly. douniaLotGerontologicaLSocia.LWork_1(4): 279-93.Spector, W.D., Katz, S., Murphy, J.B., Fulton, J.P. (1987). The HierarchicalRelationship Between Activities of Daily Living and Instrumental Activities of DailyLiving. Journal of Chronic Disease, 40(6): 481-89.Stahl, S.M. (1984). Health. In Mangen, D.J. and Peterson, W.A. (eds.). Research_Instruments in Social Gerontology, Vol. 3: Health, Program___Evaluation_andDemography University of Minnesota. 85-116.Statistics Canada (1987). Census Tract Profiles: Vancouver Part 1  Minister ofSupply and Services Canada. Ottawa.Statistics Canada (1987) Census Tract_Profiles_:Vancouver Part 2. Minister of Supplyand Services Canada. Ottawa.Stewart, A.L., Hays, R.D., Ware, J.E. (1988). The MOS Short-form General HealthSurvey. Medical Care. 26(7): 724-35.Stuen, C. (1985). Outreach to the Elderly: Community Based Services. Journal of_Gerontological Social Work 10: 85-95.Taylor, R.B., Ureda, J.R., Denham, J.W. (1982). Health_fromotort:_Prthciples_and_ClinicaLANDlications Appletom-Century-Crofts. East Norwalk, Conn.Taylor, V. (1990). Health Education - A Theoretical Mapping. Health Education__Journal 49(1): 13-14.Thornton, J.E. (1986). Life Span Learning and Education: A Conceptual Progressionin the Life Course. In Peterson, D.A., Thornton, J.E., Birren, J.E. (eds.) Education -nd Aging. Prentice-Hall, Inc.. Englewood Cliffs, New Jersey. 62-92.Tones, K. (1990). Why Theorise? Ideology in Health Education Health Education Journal, 49 (1): 2-6.Toseland, R.W., Decker, J., Bliesner, J. (1979 Spring). A Community OutreachProgram for Socially Isolated Older Persons Journal of Gerontological Social Work 1(3): 211-224.Toseland, R., Sykes, J. (1977). Senior Citizens Center Participation and OtherCorrelates of Life Satisfaction. TheGerontologist,17(3): 235-41.-167-Trela, J.E., Simmons, L.W. (1971). Health and Other Factors Affecting Membershipand Attrition in a Senior Center. Journal of Gerontology, 26(1): 46-51.Uman, G.C., Richardson, P.L. (1986). Application of Classroom Health Education toHealth Behavior in Daily Life. In Peterson, D.A., Thornton, J.E., Birren, J E Educationand Aging.  Prentice-Hall, Inc. Englewood Cliffs, New Jersey. 185-204.U.S. Department of Health, Education and Welfare (1979) Healthy_Peoplei_The_Surgeon General's Report on Health Promotion and Disease Prevention. Office of theAssistant Secretary for Health and Surgeon General, DHEW (PHS) Publication No.79-55071. Washington.Valecha, G.K., Ostrom, T.M. (1974). An Abbreviated Measure of Internal-ExternalLocus of Control. dournaLaRersonality  Assessment, 38' 369-76.Volden, C., Langemo, D., Adamson, M.,, Oechsle, L. (1990). The Relationship of Age,Gender, and Exercise Practices to Measures of Health, Life-Style, and Self-Esteem.Applied Nursing Research, 3(1): 20-26.Ware, J.E. (1987). Standards for Validating Health Measures: Definition and Content.dournaLo_f_CtirorticDisease_40(Suppl. 1): 473-80.Ware, J.E., Brook, R.H., Davies, A.R., Lohr, K.N. (1981). Choosing Measures of HealthStatus for Individuals in General Populations Amedcan_Journatof_Publiciiealth„7_1(6): 620-25.Ware, J.E. (1985). Monitoring and Evaluating Health Services. Medical Care, 23(5):705-709.Ward. W.B., Salisbury, Z.T., Zapka, J.G., Kar, S.B. (eds). (1986) Advancesdn±lealth_Education and Promotion.  Vol. 1, Part A. Jai Press. Greenwich, Conn.Ward. W.B., Salisbury, Z.T., Zapka, J.G., Kar, S.B. (eds). (1986). Advances in Health Education and Promotion  Vol. 1, Part B. Jai Press. Greenwich, Conn.Wechler, R., Minkler, M. (1986). A Community-Oriented Approach to HealthPromotion: The Tenderloin Senior Outreach Project. In Wellness an_d_EealthPromotion for the Elderly Dychtwald, K. (editor), Aspen Systems Corp., Rockville,Maryland. 301-312.Weiler,P.G. (1986). Education and Training in Wellness for Health Care Providers. InWellness anditealth Promotion for the Elderly Dychtwald, K. (editor), Aspen SystemsCorp., Rockville, Maryland. 71-86.Weiler, P.G., Chi, I., Lubben, J.E. (1989). A Statewide Preventive Health CareProgram for the Aged. Public Health Reports 104(3): 215-221.-168-Weiss, J.C. (1988). The "Feeling Great!" Wellness Program for Older Adults. Activities,Adaptation_and_Aging_12(3/4): 1-16.Whitbourne, S.K., Sperbeck, D.J. (1982). Health Care Maintenance for the Elderly. InWells, T. (ed.). Aging and Health Promotion Aspens Systems Corporation. Rockville,Maryland. 111-128.Whitelaw, N.A., Liang, J. (1991). The Structure of the OARS Physical HealthMeasures. Medical Care, 29(4): 332-47.Williams, J.H., Drinka, T.J.K., Greenberg, J.R., Farrell-Holtan, J., Euhardy, R., Schram,M. (1991). Development and Testing of the Assessment of Living Skills andResources (ALSAR) in Elderly Community-Dwelling Veterans. The Gerontologist, 31(1): 84-91.Windsor, R.A. (1986). The Utility of Time Series Designs and Analysis in EvaluatingHealth Promotion and Education Programs. In Ward. W.B., Salisbury, Z.T., Zapka,J.G., Kar, S.B. (eds ) Advances in Health Education and Promotion Vol. 1, Part A. JaiPress. Greenwich, Conn. 435-65.APPENDIX AORDER OF VISITATION OFKERRISDALE BLOCKSAPPENDIX AORDER OF VISITATION OF KERRISDALE BLOCKS (with corresponding computergenerated number)1.^41 Ave. - Larch St. - 39 Ave. - Balsam St.3. 44 Ave. - Boulevard - 43 Ave. - Maple St.4. 45 Ave. - Larch St. - 43 Ave. - Balsam St.16.^44 Ave. - Maple St. - 43 Ave. - Balsam St.18.^Balsam St. - Vine St. between 47 and 45 Ave.21.^Larch St. - Balsam St. on either side of 45 Ave.23. 49 Ave. - Yew St. - 48 Ave. - Boulevard25. Balsam St. - Vine St. on either side of 45 Ave.26. Balsam St. - Vine St. on either side of 41 Ave.27. 37 Ave. - Laburnum St. - 39 Ave. - Cypress St.31.^Elementary School-- 37 Ave, Maple St.33. 39 Ave. - Yew St. - 38 Ave. - Boulevard34. Maple Grove Elementary School35. 49 Ave. - Cypress St.41. Vine St. - Yew St., either side of 45 Ave.42. 42 Ave. - Yew St. - 41 Ave. - Bouldevard43.^45 Ave. - Boulevard - 44 Ave. - Maple St.45. 43 Ave. - Boulevard - 42 Ave. - Maple St.46. 37 Ave. - Cypress St.48. 38 Ave. - Yew St. - 37 Ave. - Boulevard51.^41 Ave. - Maple St. - 39 Ave. - Laburnum St.57.^49 Ave. - Balsam St. - 47 Ave. - Vine St.61.^41 Ave. - Boulevard- Kerrisdale Park - Maple St. (Park and School)67. 40 Ave. - Yew St. - 39 Ave. - Boulevard67.^49 Ave. - Boulevard - 48 Ave. - Maple St.69.^47 Ave. - Balsam St. - Vine St.77.^Elm St. - Larch St., either side of 46 Ave.77.^39 Ave. - Balsam St. - Vine St.81.^47 Ave. - Boulevard - 46 Ave. - Maple St.83.^42 Ave. - Boulevard - 41 Ave. - Maple St.94. 45 Ave. - Maple St. - 44 Ave. - Cypress St.95. Elm St. - Larch St., either side of 44 Ave.97.^Kerndale Park99. Larch St. - Balsam St., either side of 46 Ave.100. 41 Ave - Cypress St.100. 39 Ave. - Elm St. - Larch St.100. 43 Ave. - Elm St. - Larch St.103. Elm St. - Larch St., either side of 45 Ave.105. 41 Ave. -Yew St. - 40 Ave. - Boulevard110. Vine St. - 43 Ave. - Yew St.-171-110. 45 Ave. - Yew St. - 46 Ave. - Boulevard113. 46 Ave. - Yew St. - 47 Ave. - Boulevard114. Point Grey Park117. 39 Ave. - Larch St. - Elm St.117. 41 Ave. -Maple St. -42 Ave. - Cypress St.120. Balsam St. - 42 Ave. - Vine St. - 43 Ave.121. Laburnum St. - 38 Ave. - Cypress St. - 39 Ave.124. 47 Ave. - Boulevard - 48 Ave. - Maple St.128. 47 Ave. - Yew St. - 48 Ave. - Boulevard136. 43 Ave. - Balsam St. - 44 Ave. - Vine St.138. Elm St. - 49 Ave. - Larch St. - 47 Ave.138. Point Grey Park145. Elm Park145. Vine St. - 49 Ave. - Yew St. - 47 Ave.147. Vine St. - 47 Ave. - Yew St. - 46 Ave.150. 45 Ave. - Yew St. - 44 Ave. - Boulevard152. 41 Ave., between Elm and Larch155. Boulevard - 45 Ave. - Maple St. 46 Ave.156. 37 Ave. - Larch St. - 39 Ave. - Balsam St.160. 43 Ave. - Yew St. - 42 Ave. - Boulevard165. Vine St. - 37 Ave. - Yew St. - 39 Ave.169. 37 Ave., between Vine and Yew St.176. Cypress St. 49 Ave. - Laburnum St. - 47 Ave.186. 43 Ave. - Maple St. - 42 Ave. - Cypress St.37 Ave., between Yew and Balsam St.APPENDIX BIOWA SELF-ASSESSMENT INVENTORYAND ITS SCORE PROFILEAppendix ADirectionsThe statements on the following pages are about things that can affect our lives in one way or another. We ask you todescribe your own situation using these statements. In this way we hope to understand some of your problems and needs.Please use the following key in rating each statement:4 - True3 - More often true than not2 - More often false than not1 - FalsePlease read each statement carefully and then encircle the number corresponding to the answer that best applies to you.We realize that some of the statements may not apply directly to you all the time, but try to do the best you can. Do notworry about giving exactly the right answer; your answer may simply mean the statement is true or false to some degree.Please try to make an answer to every statement.4 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 11. I have enough money to meetunexpected emergencies.2. I sometimes get tense as I think of theday's happenings.3. I have no physical disabilities orillnesses at this time.4. People secretly say bad things aboutme.5. I need a cane, crutches, walker, orwheelchair to get around.6. I have trouble remembering things thathappened recently.7. There is no one I can turn to in times ofstress.8. I have enough money to buy those littleextras.9. I frequently find myself worrying.10. I take 3 or more medicines each day.11. Friends are disloyal to me behind myback.12. I do my own shopping without help.13. I forget where I put things.14. There is no one I can depend on for aidif I really need it.15. I have enough money to meet myregular daily expenses.16. I lose sleep over worry.17. My overall health is excellent.18. I believe I am being plotted against.19. I do my own laundry.20. I have trouble remembering the namesof people I know.21. There is someone I can talk to aboutimportant decisions.22. I need financial help.23. I am bothered by thoughts I can't getout of my head.24. My health is better than it was 5 yearsago.25. Someone has it in for me.26. Getting around town is a problem forme.27. I lose my train of thought in the middleof a conversation.28. There is no one I feel comfortabletalking about problems with.29. My finances at the present time areexcellent.30. I am a very nervous person.31. My ability to carry on my daily activitiesis worse than it was 5 years ago.32. I am sure I am being talked about.33. I am not able to prepare my own meals.34. Learning new things is harder for methan it used to be.35. No one shares my concerns.36. My monthly expenses are so high Icannot always pay my bills.37. I get upset over things.38. I have fewer health problems than mostolder people I know.39. Someone is controlling my thoughts.40. I walk without help.41. I forget appointments.42. I know people I can depend on to helpme if I really need it.43. I have some savings and/orinvestments.44. I worry over past mistakes.45. During the past year I have been so sickI was unable to carry on my usualactivities.46. Strangers look at me critically.47. I can visit a friend or relative who livesout of town for overnight or longer.48. My mind is just as sharp as ever.49. If something went wrong, no onewould come to my assistance.50. I use food stamps.51. I have more ups and downs than mostpeople.52. During the past year I have been to adoctor fewer than 4 times.53. I see things when others do not.54. I visit friends in their homes.55. I forget to take medicine when I amsupposed to.56. I do not have close relationships withother people.4 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 14 3 2 1174248^ The GerontologistTHE IOWA SELF ASSESSMENT INVENTORY^ page 11Name DateGenderTScores70^-(circle one):ER^EBFemalePH TOMaleMOAge onCSlast birthdaySSTScores70-32 3265- 31 -8532 30 31 -31 29 3060^ 30^ 28^ 32 32 29 6032 27 2829 26 3231 31 2728 25 -55- 30 27 24 30 26 31 -5529 25 -29 28 23 31 aa28 22 28 2450^ 27 25^ 21^ 30 27 23 29 502426 20 26 22- 25 23 19 29 28 -- 25 21 -45- 24 22 18 24 20 27 -45- 21 17 -23 28 23 1922 20 16 2B -22 18 -40^ 21^ 19^ 15^ 27 21 17 25 40- 1420 18 13 20 1619 26 24 -17 12 19 15 -35- 18 16 11 23 :3518 1417 15 10 2516 s 17 13 2214 -30- 15 8 24 16 12 21 -3013 15 11 -1413 12 23 14 10 2025- 12 11 13 9 -2510 8 19 -11 22 1210 9 1811 -20- 9 8 21 10 17 -208 920 16B -15- -1515 -19 1410- 18 13 -1012 -5- 17 11 - 51810 -15 9^ 0RawScores175Figure 1: Iowa Self Assessment Inventory Profile FormAPPENDIX CQUESTIONNAIRE AS IT WENT OUTTO STUDY PARTICIPANTSTHE UNIVERSITY OF BRITISH COLUMBIADepartment of Health Care and EpidemiologyFaculty of MedicineMather Building, 5804 Fairview AvenueVancouver, B.C. Canada V6T 1Z3Tel: (604) 822-2772Fax: (604) 822-4994Graduate Programs: (604) 822-5405DIFFERENCES BETWEEN PARTICIPANTS AND NON-PARTICIPANTS OFA COMMUNITY BASED SENIOR HEALTH EDUCATION PROGRAMInformation about who participates or does not participate in community healtheducation programs, and why they make that choice, is important to those whoorganise such services so that participants can obtain maximum benefit from theservice.The National Health Research and Development Program (NHRDP) has fundeda study through the University of British Columbia Department of Health Care andEpidemiology, to determine whether there are differences between seniors whoparticipate in the Kerrisdale Senior Center Health Drop In Program, and those whodon't. The names of the persons conducting the study are Dr. Samuel Sheps andShannon Berg, and they can be reached at UBC (604) 822-3910.You are being asked, along with a sample of community residents, to participateby filling out two questionnaires which answer some questions about why you do or donot participate in the Drop In program. The questionnaires should take approximatelyone hour to complete. They will be left with you overnight for you to complete and willbe collected the next day.You do not have to participate if you do not want to, and if you choose not to,this will in no way jeopardise any services that you receive at the Kerrisdale SeniorCenter. If you complete and return the questionnaire, researchers at UBC will assumeyou are consenting to participate. Do not place your name on the questionnaire. Youridentity will remain anonymous.The questionnaire will only be seen by Dr. Sheps and Shannon Berg. Allquestionnaires will be destroyed after the study is completed.Thank you for your time and effort in assisting us to determine reasons forparticipation in health education programs.-i77-If you would like to receive a summary of the results of this study, please provideyour name and address below. This information will be kept separate from yourquestionnaire to ensure that your identity remains confidential.I wish to receive a summary of the study results: Yes^ NoIf "yes", please provide your name and address:178IOWA SELF ASSESSMENT INVENTORY1DIRECTIONSThe statements on the following pages are about things that can affectour lives in one way or another. We ask you to describe your own situationusing these statements.Please use the following key in rating each statement:1 - True2 - More often true than not3 - More often false than not]4 - FalsePlease read each statement carefully and then circle the numbercorresponding to the answer that best applies to you. We realise that someof the statements may not apply directly to you all the time, but try to do thebest you can. Do not worry about giving exactly the right answer; youranswer may simply mean the statement is true or false to some degree.Please try to circle an answer for every statement.-3-1791 © 1985 by Woodrow W. Morris and Kathleen C. Buckwalter1 = TRUE2 = MORE OFTEN TRUE THAN NOT3 = MORE OFTEN FALSE THAN NOT4 = FALSE1 2 3 41. I have enough money to meet unexpectedemergencies. ^  1 2 3 42. I sometimes get tense as I think of the day'shappenings.  1 2 3 4I have no physical disabilities or illnesses atthis time. ^  1 2 3 44. People secretly say bad things about me. . . ^ 1 2 3 45. I need a cane, crutches, walker, orwheelchair to get around. ^ 1 2 3 46. I have trouble remembering things thathappened recently^  1 2 3 47. There is no one I can turn to in times ofstress. ^  1 2 3 48. I have enough money to buy those littleextras.  1 2 3 49. I frequently find myself worrying. ^ 1 2 3 410. I take 3 or more medicines each day.^ 1 2 3 411. Friends are disloyal to me behind my back. ^ 1 2 3 412. I do my own shopping without help^ 1 2 3 413. I forget where I put things^ 1 2 3 41801 = TRUE2 = MORE OFTEN TRUE THAN NOT3 = MORE OFTEN FALSE THAN NOT4 = FALSE123414. There is no one I can depend on for aid if Ireally need it. ^  1 2 3 415. I have enough money to meet my regulardaily expenses  1 2 3 416. I lose sleep over worry^  1 2 3 417. My overall health is excellent. ^ 1 2 3 418. I believe I am being plotted against^ 1 2 3 419. I do my own laundry^  1 2 3 420. I have trouble remembering the names ofpeople I know.  1 2 3 421. There is someone I can talk to aboutimportant decisions. ^  1 2 3 422. I need financial help  1 2 3 423. I am bothered by thoughts I can't get out ofmy head ^  1 2 3 .424. My health is better than it was 5 years ago. . ^ 1 2 3 425. Someone has it in for me^ 1 2 3 426. Getting around town is a problem for me^ 1 2 3 427. I lose my train of thought in the middle of aconversation. ^  1 2 3 41811 = TRUE2 = MORE OFTEN TRUE THAN NOT3 = MORE OFTEN FALSE THAN NOT4 = FALSE123428. There is no one I feel comfortable talkingabout problems with^  1 2 3 429. My finances at the present time are excellent. 1 2 3 430. I am a very nervous person. ^ 1 2 3 431. My ability to carry on my daily activities isworse than it was 5 years ago. ^ 1 2 3 432. I am sure I am being talked about  1 2 3 433. I am not able to prepare my own meals^ 1 2 3 434. Learning new things is harder for me than itused to be^  1 2 3 435. No one shares my concerns^ 1 2 3 436. My monthly expenses are so high I cannotalways pay my bills ^  1 2 3 437. I get upset over things.  1 2 3 438. I have fewer health problems than most olderpeople I know. ^  1 2 3 439. Someone is controlling my thoughts. ^ 1 2 3 440. I walk without help  1 2 3 441. I forget appointments^  1 2 3 442. I know people I can depend on to help me if Ireally need it.  1 2 3 41821 = TRUE2 = MORE OFTEN TRUE THAN NOT3 = MORE OFTEN FALSE THAN NOT4 = FALSE123443. I have some savings and/or investments^ 1 2 3 444. I worry over past mistakes. ^ 1 2 3 445. During the past year I have been so sick Iwas unable to carry on my usual activities. . . . 1 2 3 446. Strangers look at me critically^ 1 2 3 447. I can visit a friend or relative who lives out oftown for overnight or longer. ^ 1 2 3 448. My mind is just as sharp as ever. ^ 1 2 3 449. If something went wrong, no one wouldcome to my assistance^  1 2 3 450. I receive the GIS (Guaranteed Income Supplement). .1 2 3 451. I have more ups and downs than mostpeople. ^  1 2 3 452. During the past year I have been to a doctorfewer than 4 times^  1 2 3 453. I see things when others do not. ^ 1 2 3 454. I visit friends in their homes.  1 2 3 455. I forget to take medicine when I amsupposed to ^  1 2 3 456. I do not have close relationships with otherpeople  1 2 3 4QUESTIONNAIRE RE PARTICIPATION/NON-PARTICIPATION IN ACOMMUNITY BASED SENIORS HEALTH EDUCATION PROGRAM1) What is your birthdate? ^Month^Day^Year2) Sex M^ F3) What is the highest grade you completed in school? ^4) Do you live:5) Do you live in an:^alone ^with a spouse with another relative or friend ^other apartment^single family residenceother (please specify) 6) Do you know about the Health Drop In Program at Kerrisdale Seniors Center?Yes^ No^7) Have you ever attended the program? Yes^ No^8) If so, for how long?Once^More than once but less than 3 months 3 months to 1 year ^More than 1 year 9) Do you attend the Health Drop In at Kerrisdale Senior Center now?Yes ^No (go to question 14)10) What do you participate in:Exercise program  ^Blood Pressure Monitoring ^Neck and Shoulder Massage ^-8 of 10-18411) How did you hear about the Kerrisdale Senior Center Drop In?^Read about it  ^Public health nurse ^A friend told me  My doctor told me Other (specify) 12) How frequently do you attend the Health Drop In at Kerrisdale Senior Center?Every week^On a regular basis (for example, every 2 weeks)Not on a regular basis^13) Why do you attend the Health Drop In at Kerrisdale Senior Center? (select allappropriate)Health is important to me ^ Have lots of time^To make friends Center is close and easy to get to ^Need the exercise I'm good at exercise^Want to stay active ^I go to other activities so I'm there^To participate with a group I want to continue living at home^Enjoy exercising  My doctor told me to go ^Relief from boredom ^To be accepted by others^Break in routine My friends and/or spouse go ^To participate/volunteer in community work^I'm interested in learning how to keep healthy^Others (please list)^Please go to question 1514) Why don't you attend the Health Drop In at Kerrisdale Senior Center? (select allappropriate responses)Don't know about it ^Would not enjoy it ^Too far to centerCan't do exercise^Don't like to leave home^Won't do me any good ^Can't afford it^My doctor says not to^Don't need itNot interested^Not enough time^Don't know anyone there^Poor health^Health is not important to me^My friends and/or spouse think its silly ^My friends and/or spouse don't go ^I don't feel welcome there ^No transportation^Others (please list) ^-9 of 10-18515) What other community activities are you involved in and in what capacity?Program Attend Leader or volunteerOther programs at Kerrisdale Seniors Center West Main Health Unit activities Attendance at church Church groups, organizationsCultural centers, organizations ^Attend Symphony, opera, etc. Recreational activities ^Other Senior center, organizations ^Community Center^Others . ^16) On average, during one week, how many hours do you spend on the followingactivities:household activities ^exercise hobbies ^visiting friends or relatives community activities ^gardening sleepThank you for completing these questionnaires.-10 of 10-186APPENDIX DLIST OF "OTHER" ACTIVITIESSPECIFIED BY RESPONDENTSUNDER COMMUNITY ACTIVITIESAPPENDIX DLIST OF SPECIFIED COMMUNITY ACTIVITIES IN THE "OTHERS"CATEGORYA) PARTICIPANTS:• Art and Choir• Board Member for Kiwana's Neighbourhood House• Board Member of False Creek Recreation Association• Board member for A.S.K.• Board member for apartment building• Board member for Operating Committee• Bowling• Children's Hospital Auxiliary for 40 years• Entertain seniors in nursing homes, etc.• Garden Clubs• Golf Club• Hospital volunteer (2)• Piano lessons• Rotary Club• South Slope YMCA• University Women's Club of Vancouver• Whist at different churches on Saturday nightsB) NON-PARTICIPANTS• Arbutus Club• Ballroom dancing• Boundary Bay Flying Club (2)• Bridge (5)•Carpet Bowling• Chess• Classes at centres• Family gatherings, visit family out of town, etc. (3)• Golf (2)• Help a "hard of seeing" friend• Keep in contact with some elderly people• Library (3)• Night school courses• Recorder group• Red Cross• Royal Canadian Legion• RCMP Veteran's Ladies Auxiliary• Seminars: New Age Group• Square dancing• Volunteer Canadian Cancer• Volunteer work for hospitals (2)• Volunteer for Meals on Wheels (2)• Woodpen Club• Work part-timeAPPENDIX EMEANS OF RESPONSES FOR PARTICIPANTSCHOOSING EACH REASONCOMPARED TO MEANSFOR ALL PARTICIPANTSAPPENDIX EAPPENDIX E1: Means of Responses for Participants Choosing Each ReasonCompared to Means for All Participants (Age and ISAI Scores)AGE (yrs) ISAI SCALES:BR B3 PH TO MO CS SSALL^PARTICIPANTS 74.8 30.1 24.4 24.4 30.1 2 9 24.2 29.3PREDISPOSING (n = 51) 74.4 30.2 24.3 23.8 30.8 28.9 24 29.5Do volunteer work (n = 7) 73.9 30.4 24.9 22.7 30.9 30 25.1 30.9Health important to me (n = 39) 73.4 30.2 24.2 24.2 30.7 - 28.9 24 29.9Health learning interest (n = 21) 73.6 29.5 24.5 23.6 30.4 29.1 23.6 29.7Want to stay home (n = 19) 77.5 29.9 25 24.4 31.21 28.6 24.7 29Want to stay active (n = 28) 74.1 29.9 24.5 24.5 30.3 29.2 23.8 29.1Need the exercise (n = 24) 73.8 29.7 24.7 24.6 30.5 29.6 23.3 29.1ENABLING (n = 26) 75.3 30.7 25.3 24.8 30.6 29.6 23.8 29.4Good at exercise (n = 14) 76 30.6 25.1 26.2 30.9 29.9 23.2 30Centre close (n = 22) 75.4 30.8 25.5 25.5 30.7 29.3 24.7 29There for other activity (n =10) 74.2 30.5 25.7 24.9 30.3 30 22.8 29.7Lots of time (n = 9) 73.4 31.3 24.1 26.9 30.1 28.7 22.8 27.3REINFORCING (n = 37) 74.6 29.8 24.3 24.6 30 28.6 24.6 29.1Enjoy exercising (n = 19) 74.4 30.3 23.7 25.7 30.4 29.1 23.1 28.8Dr. said to go (n = 6) 77.5 31.2 26.8 24.2 30.6 26.8 22 29.2Relief from boredom (n = 7) 73.4 29.1 23.6 23.9 30.6 26.9 21.7 26.4"Break in routine (n = 7) 72.3 29.9 21.6 25.7 29.6 27.1 23.1 25.9Participate with group (n = 16) 75.6 30.1 23.9 25.1 31 29.9 - 24.7 29.6Friends/Spouse go (n = 15) 76 30.2 25.7 24.9 30.1 28.4 25.2 29.6Be accepted by others (n = 2) 71 24.5 22.5 19 32 22.5^ 24.5 23.5Make friends (n = 14) 72.6 28.6 23.5 24.7 29.7 28 23.9 27.7ASignificantly higher using Mann Whitney U tests at the p s .05 levelt Significantly higher using Mann Whitney U tests at the p s .01 levelA Significantly lower using Mann Whitney U tests at the p s .05 levelMeans for respondents choosing each reason are compared to the overall participant group means.-191-APPENDIX E2: Means of Responses for Participants Choosing Each ReasonCompared to Means for All Participants (Hours of Activity and Community Activity)Hours.>,->-0-6;0=of01,x.-LIm.20.1.Activityto-,..._>(hrs.).3'...C=g0UQ m.,.=. Ei)ct^.c_,97,--)CI)..?,,-?.....0'-'-'75H2ci.,ugci).7y"7. E„c.)Community.0'5Ut4EU- 55 ..=UActivity.=..==-a,,,,!T!2.-U..,.2U>.,r..3,'';Cl).c.)'<Fit:6c' 728Lr(proportion)egu O'5,t',-.)-ECe.a.c..)u. .a.`.Ut0TOTAL^PARTIC. 11.5 5.6 5.1 5.8 2.9 2.1 47 76.3 0.3 0.3 0.2 0.1 0.4 0.3 0.2 0.1 0.2PREDISPOSING (51) 11.9 5.1^4.7^ 6 2.9 2.2 1, 48.6 76.71 0.3 0.3 0.3 0.2 0.5 0.4 0.1 0.1 0.2Volunteer work ( 7) 12.1 4.4 4.6 3.3 7.7t 2.7 43.5 73.3 0.91. 0.4 0.4 0.1 0.7 0.4 0.3 0 0.3Health^important (39 12.4 5.9 5.4 5.1 3.5 2 49.9 80.3 0.3 0.3 0.3 0.2 0.5 0.4 0.2 0.1 0.2Health^learning^(21) 11.6 4.3 7.3 4.3 3 2.1 51.8 76.4 0.4 0.4 0.3 0.1 0.4 0.4 0.3" 0.1 0.2To stay home (19) 12.7 5 4.6 4.1 2.2 2.9 51.6 78.4 0.3 0.3 0.2 0.1 0.4 0.3 0.3" 0.1 0.2To stay active (28) 12.7 5 5.5 4.4 2.1 2.4 51.3 78.5 0.4 0.3 0.3 0.2 0.4 0.4 0.2 0.1 0.2Need exercise (24) 9.1 5.6 4 6.1 2 2 46.7 71.6^0.3 0.3 0.2 0.2 0.5 0.4 0.2^0.2 0.2ENABLING (n = 26) 10.1 6 4.9 3.8 1.8 2 49.2 74.5 0.4 0.4 0.2 0.2 0.5 0.4 0.3" 0.2 0.2Good at exer. (14) 12.8 8.2 5.5 4.8 2.2 2.2 49.4 85^0.4 0.4 0.2 0.1 0.5 0.4 0.2^0.2 0.1Centre close (22) 9.9 5.8 4.6 4 1.6 1.3 49.4 73^0.3 0.4 0.2 0.1 0.6 0.3 0.2^0.1 0.2There^(10) 14 8.6 5.8 5.1 4 3.1 48.8 91.6 0.8t 0.2 0.2 0.2 0.4 0.3 0.5t 0.2 0.2Lots of time (n = 9) 14.4 5.7 8.6 3.7 1.6 1.7 49.1 86.9 0.4 0.2 0.2 0.2 0.6 0.2 0.1^0.2 0REINFORCING (37) 10.4 6 5.1 4.7 2.3 1.9 47.3 73.5 0.3 0.3 0.3 0.2 0.4 0.4 0.2^0.1 0.1Enjoy exercise^(19) 11.9 6.8 5.8 4.8 1.9 2.1 50.2 81.8^0.4 0.2 0.3 0.2 0.5 0.3 0.2^0.2 0.2Dr. said to go (n = 6) 10.4 9.8 6^17.8 8.2 1 47.4 90.6 0.5 0.2 0 0 0.2 0.2 0.2^0.2 0.2Relieve boredom (7) 11.7 7.1 5.1 5.3 1.7 1.3 52 87.1^0.6 0 0.1 0.1 0.6 0.1 0.1^0.1 0Break^in^routine (7) 12.4 3.6 9.1 3 0.3 1.7 51.5 77.1^0.4 0.4 0.1 0.1 0.4^0 0.1^0.3 0Partic.^in^group^(16) 12.5 6.5 2.6 4.8 3.3 1.5 49.6 74.8 0.5" 0.3 0.3 0.2 0.610.3 0.3^0.1 0.1Fr./Spouse go (15) 9.2 5.1 6 4.5 0.9A 2.3 46.6 75.9 0.2 0.3 0.2 0.1 0.3^0.5 0.2^0.2 0.1Be accepted^(2) 15 1.5 0 0 0 4 63 83.5^0 0 0.5 0 0.5^0 0^0 0Make friends^(14) 12.6 6.6 5 4.3 1.3 2.3 51.6 76.8 0.4 0.2 0.1 0.1 0.3^0.3 0.2^0.1 0Significantly higher at the p s .05 level.^ -I-Significantly higher at the p s .01 level." Significantly lower at the p s .01 level. ° Significantly lower at the p s .01 level.Mann Whitney U test used for hours of activity; Contingency table analysis used for Community activities.Means for respondents choosing each reason are compared to the overall participant group means.-192-APPENDIX FMEANS OF RESPONSES FOR NONPARTICIPANTSCHOOSING EACH REASONCOMPARED TO MEANS FORALL NONPARTICIPANTSAPPENDIX F1: Means of Responses for Nonparticipants Choosing EachReason Compared to Means for All Nonparticipants.AGE (yrs)BRISAI SCALES:EB PH TO MO CS SSALL^NONPARTICIPANTS 77.1 2 9 2 6 21.3 30.8 27.2 24.9 29.9PREDISPOSING (n = 50) 76.4 29.7 27.3" 23.2t 30.9 28.2 25.9" 30.6Don't like to leave home (n = 3) 80.321.3° 28.3 26 26.3 26.3 25 25.7Health not important (n = 1) 76 32 29 27 32 27 27 29Won't do me any good (n = 2) 74.5 26 26.5 19 28.5 23 28.5 26.5Would not enjoy (n = 14) 79.3 29.5 26.6 24.7" 31.2 26.7 24.9 29.3Not interested (n = 27) 76.8 29.9 27.8" 22.2 31.3 28.3 26.2 30.9"Don't need it (n = 25) 76.1 29.9 27.4 25.61 31.4 .'29.8" 25.8 30.2ENABLING (n = 70) 77.2 28.7 25.7 20.9 30.8 26.3" 25.1 29.8Not enough time (n = 22) 74.5 29.6 26.7 24.71 31 28.1 26.5" 31.1Don't know about it (n = 43) 75.8 29.1 25.4 20.2 30.8 26.9 25.1 29.4Too far to centre (n = 1) 88 25 16 19 31 21 19 29No transportation (n = 8) 86.4t 29.8 25.9 19.8 31.321.5° 22.4 29.5Can't afford it (n = 4) 81.3 1 5° 23.3 23.7 29.3 A 27.7 25.3 25.7Poor health (n = 9) 82.1" 26.423.2°12.9° 30.618.1°21.7^ 29.4Can't do exercise (n = 14) 821' 27.5 24.617.2° 30.521.4°22.2 ,1 23.2^REINFORCING (n = 33) 80.7t 28.2 24.8 20.1 30.224.0° 24.127.5°Don't know anyone there (n =19) 81" 27.4 24.3A 20.2 30.123.3° 24.326.4°Friends/Spouse don't go (n = 7) 78.4 30.7 26.7 21.7 30.7 28.1 23.3 29.6Don't feel welcome (n = 3) 79.3 28.3 27.7 23.3 29 25.3 28.7 28Fr/Sp think^its silly (n^= 2) 69.5 32 26 24.5 31 28 26.5 30.5Significantly higher using Mann Whitney U tests at the p s .05 levelt Significantly higher using Mann Whitney U tests at the ps .01 level° Significantly lower using Mann Whitney U tests at the p 5 .01 levelA Significantly lower using Mann Whitney U tests at the p s .05 levelMeans for respondents choosing each reason are compared tooverall nonparticipant group means.-194-APPENDIX F2: Means of Responses for Nonparticipants Choosing Each ReasonCompared to Means for All Nonparticipants (Hours of Activity and Comm. Activity)Hours.>,.-,T-69of,,,)5Activityys-a2.0E.5.(hrs.)›,.;.-.?'gEgUti). r,,-ai 2e3 .c,..E iie-.•>".---<...)z0-3geE,c-`-'c)cf-'c.)7i'"O1'Community-'==2c:,'-.="'fActivitym0. 0....,,,. '.2.g01)6-P..Uci.;-,..:)..2.c.):-.>. _5<:""-5 5 ,;)(proportion)Et._5U'5. -ac.)cf-'1C2 5ej..?;•-^Ee;,...)-5TOTAL^NON-PAR.^13.7 4.3 8.2 7.2 3.7 3.7 51.5^88.3PRED. (50)^13.3 5.4 7.5 5.8 0.6^ 4.5 53.1^89.6 0.1 0.3 0.2 0.1 0.3 0.2 0 0.1 0.3No leave home (3)^0.3 1 4.7 0 0 0 64.3^93.3 0 0.3 0 0 0 0 0 0 0Health not NB (n = 1)^31 4 24 0 0 0 39^98 0 0 0 0 0 0 0 0 1Do me no good (n = 2)^3 0 5 2 0 0 50^60 0.5 0 0 0 0.5 1 0 0 0Would not enjoy (n = 14)^9.6 6.2 12.8 4.8 0 5.5 48.5^88.2 0 0.4 0.3 0 0.3 0.1 0.1 0.1 0.3Not interested (n = 27)^15.2 4.8 8.3 5.7 0.6 4.5 53.1^90.2 0.1 0.4 0.3 0 0.3 0.3 0 0.1 0.3Don't need it (n = 25)^12.1 5.9 8.5 7 0.5 5.9 50.2^90.6 0.1 0.4 0.3 0.1 0.3 0.2 0 0.1 0.3ENABLING (n =70)^11.4 4.1 10.1 5 1.4 4.4 49.1°^84.9 0.1 0.4 0.2 0.1 0.3 0.2 0.1 0.1 0.3Not enough time (n = 22)^9 5.2 8 4.8 1.6 7.1 - 49.3^84.8 0.1 0.4 0.3 0.1 0.5 0.2 0.1 0.1 0.4No knowledge (n = 43)^14.4 3.8 10.9 5.2 1.4 4.3 49.3^89.8 0 ^ 0.3 0.1 0.1 0.3 0.1 0.1 0.1 0.3Too far to centre (n = 1)^3 0 0 0 0 1 49^53 0 1 1 0 0 0 0 0 1No transportation (n = 8) 12.3 2.1 17.1 3.1 0 2.1 44.4^79.9 0  0.3 0 0 0.1 0 0.1 0.3 0.3Can't afford it (n = 4)^8.3 2.3 8.8 2 0 2.5 49.8^73.5 0.3 0.5 0 0 0 0 0 0 0Poor health (n = 9)^4.2^ 1^ 5.8 5.8 0 0.2 48.4 57.3^ 0.1 0.3 0.1 0 0.1 0.1 0.1 0.2 0.1Can't exercise (n = 14)^6.5 ° 2.8 7.8 4 0 1.8 45.2^ 64.2° 0.2 0.4 0.1 0.1 0.1 0.1 0.1 0.1 0.1REINFORCING (n = 33)^10.3 3.2 9.2 4.3 0 ° 1.3° 48.1^73.7^ 0.1 0.3 0.1 0 0.1 0.1 0 0.1 0.2Know no one (n =19)^10.5 2.5^ 10 4.1 0 ^ 1.5^ 48.2^73.7 0.1 0.3 0.1 0 0.1 0.1 0 0.1 0.2Fr./Sp. don't go (n = 7)^11.6 4.3 4.7 6.1 0 0.1 A 47.1^74 0.3 0.6 0.4 0 0.4 0.3 0 0.1 0.3No feel welcome (n =3)^23.5 9.5 2 7 - 7.5 0 4 51^122 0.3 0 0 0 0 0.3 0 0 0.3Fr/Sp think silly (n = 2)^29.5 5 14 6 0 0 47.5^102 0 0.5 0.5 0 0 0.5 0 0 1Significantly higher at the p s .05 level.^ t Significantly higher at the p s .01 level." Significantly  lower at the p s .01 level. ° Significantly lower at the p s .01 level.Mann Whitney U test used for hours of activity; Contingency table analysis used for Community activities.Means for respondents choosing each reason are compared to the overall nonparticipant group means.-195-


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items