The relationship between aging expectations and ...

Psychology of Sport and Exercise 14 (2013) 569e576

Contents lists available at SciVerse ScienceDirect

Psychology of Sport and Exercise journal homepage: www.elsevier.com/locate/psychsport

The relationship between aging expectations and various modes of physical activity among aging adultsq Brad A. Meisner a, *, Patricia L. Weir b, Joseph Baker c a

Department of Psychology, Ryerson University, Toronto, Ontario M5B 2K3, Canada Department of Kinesiology, University of Windsor, Windsor, Ontario, Canada c School of Kinesiology and Health Science, York University, Toronto, Ontario, Canada b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 28 June 2011 Received in revised form 22 February 2013 Accepted 26 February 2013 Available online 19 March 2013

Objective: Previous research has indicated that older adults who expect decline and disease with age are less likely to engage in aerobic exercise. This study explores the influence that different types of aging expectations have on various modes of physical activity (PA) among aging adults. Design & methods: Community-dwelling adults aged 41e97 years (M ¼ 70.8, SD ¼ 12.8, n ¼ 247) completed a questionnaire including the Physical Activity Scale for the Elderly, the Expectations Regarding Aging (ERA) Survey, and a number of demographic, socio-economic, and medical covariates. Results: Bivariate analyses revealed significant relationships between overall ERA scores and multiple modes of PA, but not at multivariate levels. Bivariate analyses of the ERA sub-scales revealed significant associations among PA and the physical health ERA sub-scale but not mental health or cognitive function ERA sub-scales. In the multivariate analyses, higher physical health ERA was correlated with strenuous sport and recreational physical activities after adjusting for all covariates among aging adults without restrictions of daily activity (OR ¼ 1.01, CI ¼ 1.00e1.02, n ¼ 194). Conclusions: The association between aging expectations and PA appears to be dependent on the type of aging expectation (i.e., physical health), the mode or intensity of PA, and the functional abilities of the aging adult. Ó 2013 Elsevier Ltd. All rights reserved.

Keywords: Gerontology Social perception & cognition Sports Recreation & leisure Health psychology

Individuals born within the years of the baby boom (1946e 1964) are now middle-aged and older adults. As these individuals continue to advance into later life, there will be increases in rates of morbidity, disability, and dependency as well as unprecedented demands on health care services (Parker & Thorslund, 2007; Werblow, Felder, & Zweifel, 2007). In light of these bleak forecasts, there has been a surge of research interest on understanding what it means to age healthily. Emerging from this body of research is the concept of ‘successful aging’, which discerns the risk factors of ill health as well as the promoting factors of optimal health among aging individuals. Finding effective ways to facilitate health, and negate decline, is crucial to encourage successful aging among middle aged and older adultsdone way to do so is through participating in physical activity.

q This research was supported by a Social Sciences and Humanities Research Council (SSHRC) of Canada Doctoral Fellowship awarded to BAM (752-2009-2047) and an SSHRC and Sport Canada Research Initiatives standard research grant (8622007-0002) awarded to JB and PLW. * Corresponding author. E-mail address: [email protected] (B.A. Meisner). 1469-0292/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.psychsport.2013.02.007

There is an increasing amount of literature positioning physical activity as an effective practice to attain and/or maintain successful aging. For instance, moderate and high levels of physical activity are associated with a number of successful aging outcomes, such as living to an advanced age (Wen et al., 2011), having little to no disability prior to death (Menec, 2003), as well as having a low probability of disease/disease-related disability, high cognitive/ physical functioning, and being actively engaged with life (Baker, Meisner, Logan, Kungl, & Weir, 2009). Furthermore, Vaillant and Mukamal (2001) investigated the role that regular exercise had on successful aging, which was defined by longevity with high levels of physical, mental, and social well-being. Participants who expended 500 kcal per week or more through exercise were over three times more likely to be classified as aging successfully than those who expended less than 500 kcal per week. Evidence has shown that the influence that physical activity (or inactivity) has on overall successful aging is through independent associations with many of the measures popularly used to define ‘success’. For example, physical inactivity was found to be positively associated with increased likelihood of reporting disease and disability, low functional capacities, and being socially disengaged

570

B.A. Meisner et al. / Psychology of Sport and Exercise 14 (2013) 569e576

with life (Meisner, Dogra, Logan, Baker, & Weir, 2010)dthree commonly used criteria used to determine unsuccessful aging (Depp & Jeste, 2006; Rowe & Kahn, 1987, 1998). Even though physical activity has a greater influence on some successful aging criteria (e.g., functional capacity) than others (e.g., active life engagement), the multidimensional benefits of physical activity for promoting well-being are undeniable, particularly among aging adults (King & King, 2010; Meisner et al., 2010). Additionally, the significant associations between physical activity and these diverse definitions and criteria of successful aging provide further support for the multifaceted and important influence that physical activity has on aging successfully. Despite the evidence, the majority of middle-aged and older adults do not engage in physically active lifestyles (e.g., Craig, Russell, Cameron, & Bauman, 2004) and adherence to physical activity programs among these age groups is often low (e.g., Thurston & Green, 2004). In an effort to enhance these poor participation rates, a considerable amount of research has investigated the barriers to participating in physical activity experienced by aging adults and many notable factors have been identified (please refer to Lim & Taylor, 2005; Trost, Owen, Bauman, Sallis, & Brown, 2002). One of these important barriers includes beliefs about the aging process itself. It has been found that beliefs toward aging can act as an important associate of physical activity. For instance, both real and perceived risks that relate to one’s advanced chronological age have been found to result in decreased levels of physical activity among aging adults (O’Brien Cousins, 2000, 2003) and, as a result, negative aging beliefs may render an aging person to accept that they are ‘too old’ to engage in physical activity. This acceptance may lead to the rejection of physical activity, its benefits, and/or the usefulness of changing physical activity practices (Horton, 2010). The extent to which beliefs toward aging influence physical activity is likely due to both the number and strength of negative age stereotypes that relate to physical and functional performance, specifically. In fact, Levy (1996) discovered that some of the most widely-supported age stereotypes held uniquely by aging adults relate to physical health, functioning, and performance (e.g., decline, decrepit, and diseased). These age stereotypes provide ‘knowledge’ to aging adults on how to behave and function in ‘age-appropriate ways’ within the physical domain (Levy & Leifheit-Limson, 2009; Meisner, 2012). Research shows that aging adults who believe that functional decline is inevitable with advanced age often disengage from activities that support functioning-related outcomes, such as physical activity (O’Brien Cousins, 2000). For example, a longitudinal study by Levy and Myers (2004) found in the United States that middleaged and older adult’s beliefs toward aging predicted preventive health behaviors, such as participating in exercise, over a 20-year period. Specifically, after adjusting for covariates, positive age self-perceptions reported in 1975 significantly predicted healthier behavioral practices in 1995. Similarly, Kim (2009) found a moderate positive association between aging expectations (i.e., expecting to maintain health with aging) and six combined healthpromoting behaviors among older Korean adults. One of these six health behaviors was physical activity. To date, only one study has investigated the relationship between aging expectations and physical activity practices in particular. In the greater Los Angeles region, Sarkisian and her colleagues (Sarkisian, Prohaska, Wong, Hirsch, & Mangione, 2005) discovered a significant positive association between aging expectations and aerobic physical activities such as walking for exercise, swimming, bicycling (both stationary and road), and ‘other’ aerobic exercises. Results indicated that participants with the lowest aging expectations (i.e., expecting declines in health and functioning with age) were: (a) 1.6

times less likely to engage in 60 min of moderate-to-vigorous physical activity in the previous week; (b) 2.6 times more likely to report performing less than 30 min of moderate-to-vigorous physical activity in the previous week; and (c) 2.9 times more likely to report no physical activity in the previous week, compared to those with the highest aging expectations (i.e., expecting to maintain health and function with age). These findings were independent of a number of covariates and confounding variables such as age, sex, ethnicity, comorbidity, and functional impairment. When considered together, the above research indicates that the endorsement of negative aging beliefs (i.e., ‘low’ aging expectations of decline) may predict low physical activity patterns among the aging population. In fact, it has been hypothesized that the contribution that ageist beliefs have on health, physical activity, and successful aging extends to the entire aging population (Levy, 2003; Meisner, 2012; Ory, Hoffman, Hawkins, Sannerc, & Mockenhaupt, 2003). However, despite this potential wide-scale impact, the association between aging beliefs and physical activity among aging adults is relatively unexplored. Also, there is a lack of quantitative research that discerns the influence that aging beliefs have on physical activity in an aging context at group-levels. As such, the objective of this study was to explore the association between aging expectations and physical activity to deepen and expand the current body of literature. One way this objective was achieved in this study was by assessing multiple, diverse modes of physical activity that are commonly practiced by aging adults (vs. specific, traditional aerobic exercises; Sarkisian, Prohaska, et al., 2005) to investigate the ubiquity of the association between aging expectations and physical activity. Despite the exploratory nature of this study, based on previous literature (i.e., Sarkisian, Prohaska, et al., 2005), it was hypothesized that higher aging expectations would be associated with increased involvement in physical activity. Method Participant recruitment These analyses are part of a larger research project examining the relationships between beliefs toward aging and involvement in a range of health behaviors (e.g., Meisner & Baker, 2013). Recruitment took place in the Greater Toronto Area in retirement housing complexes, senior centers, and a broad range of recreational activity groups (e.g., mall walking, bridge clubs, etc.) from June to December of 2009. At these locations, adults were approached and asked whether they were interested in participating in a paper-based research survey. To be eligible, participants had to be able to read English and be at least 40 years of age. This age range is represented in previous research (e.g., Levy & Myers, 2004) and it was used to provide a more inclusive perspective of the influence that aging expectations have on health-related behavior among aging adults. Upon informed consent, participants were given a self-reported questionnaire that was later mailed-in. Measures Physical activity The Physical Activity Scale for the Elderly (PASE; Washburn, McAuley, Katula, Mihalko, & Boileau, 1999; Washburn, Smith, Jette, & Janney, 1993) was used to measure physical activity practices. This scale has previously demonstrated acceptable validity and reliability in community-dwelling aging adult samples (Bonnefoy et al., 2001; Harada, Chiu, King, & Stewart, 2001; Washburn et al., 1993). This scale was used to measure 12 leisure time, household, and work-related modes of physical activity that are commonly practiced among aging adults, based on the previous


seven days. First, participants reported whether they practiced (yes/no) any of the following six modes of physical activity: (a) lawn work or yard care (e.g., snow or leaf removal); (b) caring for another person (e.g., child, dependent spouse, and another adult); (c) home repairs (e.g., painting or wall papering); (d) heavy housework or chores (e.g., vacuuming or scrubbing floors); (e) light housework (e.g., dusting or washing dishes); and (f) outdoor gardening. Then an additional six modes of physical activity were measured: (a) muscle strength and endurance exercises (e.g., lifting weights, push-ups, etc.); (b) strenuous sport and recreational activities (e.g., jogging, swimming, cycling, singles tennis, etc.); (c) moderate sport and recreational activities (e.g., doubles tennis, softball, golf without a cart, etc.); (d) light sport and recreational activities (e.g., bowling, golf with a cart, etc.); (e) job for pay or volunteer involving standing or walking; and (f) walking outside the home for any reason (e.g., fun, exercise, walking to work, etc.). For each of these latter six modes of physical activity, participants indicated their participation by frequency (i.e., ‘never’, ‘seldom [one to two days]’, ‘sometimes [three to four days]’, or ‘often [five to seven days]’) and duration (i.e., ‘less than 1 h’, ‘between one but less than 2 h’, ‘between 2 and 4 h’, or ‘greater than 4 h’). These data were then used to quantify ‘activity frequency values’ for each of the 12 physical activities and each activity frequency value was then multiplied by its standardized PASE activity weight according to the PASE scoring procedures, resulting in 12 standardized physical activity frequency products. Also, these products were summed across all 12 modes of physical activity to provide an overall PASE score. The overall PASE score was represented on a continuous scale, on which higher scores indicate greater levels of overall physical activity in the previous week (Washburn et al., 1993). More detailed information on the PASE and how it is scored can be found elsewhere (New England Research Institutes, 1991; Washburn et al., 1993). Expectations Regarding Aging Aging expectations were measured with the 12-item version of the Expectations Regarding Aging Survey (Sarkisian, Steers, Hays, & Mangione, 2005). This scale includes three four-item scales on physical health, mental health, and cognitive function aging expectations. Example items from this survey include: “The human body is like a car: When it gets old, it gets worn out” and “Having more aches and pains is an accepted part of aging” (physical health sub-scale); “Being lonely is just something that happens when people get old” and “As people get older they worry more” (mental health sub-scale); as well as “I expect that as I get older I will become more forgetful” and “It is impossible to escape the mental slowness that happens with aging” (cognitive function sub-scale). Participants were asked to what extent they agreed or disagreed with each item on a four-point scale (i.e., definitely true, somewhat true, somewhat false, and definitely false). According to the ERA scoring procedures, these data were used to compute overall and sub-scale continuous scores that can range from 0 to 100 (Sarkisian, Steers, et al., 2005). Higher overall scores indicate expectations of achievement and maintenance of health with age while lower scores indicate expectations of decline with age. This instrument has previously demonstrated acceptable reliability and construct validity and detailed information can be found elsewhere (Sarkisian, Steers, et al., 2005). In the current study, Cronbach alpha statistics demonstrated acceptable internal consistency among the sub-scale items as alpha values ranged from .74 to .86. Covariates In concordance with previous research (e.g., Levy & Myers, 2004; Sarkisian, Hays, & Mangione, 2002), a number of demographic, socio-economic, and medical measures were included in the

571

multivariate analyses. Participants were asked for their gender and age. Ethnicity was asked with an open-ended question, which was coded into the following recommended categories (Statistics Canada, 2008): Caucasian/White; Canadian; British Isles; European (East, West, South, and Other combined); and Asian (South, West, East and Southeast Asian, and Arab combined). Education was categorized into those who completed high school or less, a college diploma, an undergraduate degree, or a graduate degree. Income was separated into those with annual household earnings of $0e $39,999; $40,000e$79,999; or $80,000 and greater (Canadian currency). Both education and income measures were entered as ordinal scale variables. A question about employment was asked that categorized participants as working either part-time, full-time, or whether they had retired from employment. Also, depressive symptoms were measured by the 15-item Geriatric Depression Scale (Brink et al., 1982; Yesavage et al., 1983), which was developed as a basic screening measure for depression in aging adults and scores on this scale were summed and entered as a continuous variable. Participants also reported (yes/no) if they experienced any pain while sitting or standing during the previous seven days, experienced any restrictions in daily activity due to discomfort or pain during the previous seven days, and whether they have been diagnosed with a chronic condition by a health care professional that has lasted, or is expected to last, six months or longer. Statistical procedures and analyses Continuous scores of the overall aging expectation scale and its three sub-scales served as the four focal independent variables. The standardized physical activity frequency products of the 12 individual modes of physical activity as well as the overall PASE score were examined as dependent variables. Given the large number of dependent measures, biserial and Pearson correlation tests examined the associations between the dichotomous (i.e., lawn work or yard care, caring for another person, home repairs, heavy housework or chores, light housework, and outdoor gardening) and continuous (i.e., muscle strength and endurance exercises, strenuous sport and recreational activities, moderate sport and recreational activities, light sport and recreational activities, job for pay or volunteer involving standing or walking, and walking outside home for any reason) physical activity measures with the continuous aging expectations scales. Significant associations (p < .05) from these preliminary correlations determined which associations warranted further examination at the multivariate level. For the main multivariate analyses, given that the physical activity data were both dichotomous or continuous, either logistic and linear regression techniques were used, respectively. To compare the effect sizes observed between logistic and linear regression techniques, the beta coefficients and 95% confidence intervals from the linear regression analyses were exponentiated (i.e., ex, where x ¼ beta) to be analogous to odds ratios and confidence intervals estimated in logistic regression analyses. To quantify the associations between the significant dependent measures and aging expectations, bivariate (Model 1, to supplement the bivariate correlations) and step-wise multivariate regression models were developed. Covariates were adjusted for in three blocks (Models 2 through 4): 1) demographic factors (age, gender, and ethnicity); 2) socio-economic factors (education, household income, and employment status); and 3) medical factors (depressive symptoms, pain, restriction of activities, and chronic disease). Two sets of identical models were developed for each dependent physical activity variable, one for overall ERA scale and the other for the ERA subscales, as done in previous research (Sarkisian, Prohaska, et al., 2005). All tests were performed using SPSS version 19.0 using a critical alpha level of .05.

572


Results Descriptive statistics Complete details on participant characteristics are presented in Table 1 (n ¼ 247). The average age of this sample was 70.8 years (SD ¼ 12.8, range ¼ 41e97), 67% were female, and 19% identified with being White or Caucasian. Regarding the socioeconomic factors, 29% completed high school levels of education or less while 20% had graduate degrees; 26% had less than $40,000 of household annual income while 45% had $80,000 and more; and 75% were retired from employment. Few participants reported depressive symptoms (7%; M ¼ 1.8, SD ¼ 2.3, range ¼ 0e12); however, 52%

Table 1 Descriptive statistics of sample (n ¼ 247). Measure Physical activity dependent Lawn work or yard care1 Caring for another person2 Home repairs1 Heavy housework or chores1 Light housework3 Outdoor gardening3 Muscle strength and endurance4 Strenuous sport and recreation1 Moderate sport and recreation2 Light sport and recreation1 Active job or volunteer3 Walking outside5 Overall PASE score6 ERA independent variables Overall7 Physical health sub-scale7 Mental health sub-scale7 Cognitive function sub-scale7 Covariates Age Gender Ethnicity5

Education status4

Annual household income8 Employment status4 Depressive symptoms7 Pain9 Restriction of activities4 Chronic disease

Category variables Yes Yes

n (valid %) or M (SD, range) 79 (33.1) 94 (39.7)

Yes Yes

30 (12.6) 128 (53.6)

Yes Yes e

227 (94.6) 116 (48.3) .14 h/day (.29, .00e1.50)

e

.20 h/day (.46, .00e2.57)

e

.20 h/day (.48, .00e2.57)

e

.47 h/day (.71, .00e4.29)

e e e

.99 h/day (1.42, .00e4.00) .46 h/day (.50, .00e2.57) 127.51 (70.22, 2.20e409.65)

e e

46.33 (18.09, 0e97.20) 38.38 (21.28, 0e100)

e

61.10 (22.47, 0e100)

e

39.52 (22.82, 0e100)

reported experiencing pain, 34% reported restriction of daily activities, and 82% reported having one or more chronic diseases. Regarding physical activity, based on the previous seven days, the mean overall PASE score was 127.5 (SD ¼ 70.2, range ¼ 2.20e 409.65). For the modes of physical activity assessed dichotomously (i.e., yes/no), the vast majority of the participants reported engaging in light housework (95%) while about half reported engaging in heavy housework and chores (54%) or outdoor gardening (48%). Less frequently reported physical activities were caring for another person (40%), lawn work and yard care (33%), and home repairs (13%). For the modes of physical activity measured continuously (i.e., units of hours per day and then averaged across the previous week), the activity that participants engaged in for the longest amount of time was having a job or volunteer position that involved standing or walking (59.4 min per day). Both walking outside and light sport and recreational activities were practiced for approximately one half of an hour on an average day (27.6 and 28.2 min per day, respectively). Lastly, participants practiced an average of 12.0 min per day of both moderate and strenuous sport and recreational activities and about 8.4 min per day of muscle strength and endurance exercises. The mean score of the overall aging expectation scale was 46.3 (SD ¼ 18.1, range .00e97.20); however differences were observed among the sub-scales. Mean scores for the physical health and cognitive function aging expectation sub-scales were similar (i.e., M ¼ 38.4, SD ¼ 21.3 and M ¼ 39.5, SD ¼ 22.8, respectively), while the mental health aging expectation sub-scale mean was notably higher (i.e., mean ¼ 61.1, SD ¼ 22.5). A Pearson correlation matrix of the overall sample revealed that the subscales were all significantly related to one another (all p < .001) and coefficients ranged from .47 to .52. Preliminary correlation analyses As reported in Table 2, the biserial (i.e., ‘rb’ for dichotomous physical activity measures) and Pearson (i.e., ‘r’ for continuous

Table 2 Bivariate correlations between each mode of physical activity and aging expectations. Physical activity

e Female Caucasian/White Canadian British European Asian High school or less College Undergraduate Graduate $0e$39,999 $40,000e$79,999 $80,000 Retired e Yes Yes

70.81 (12.76, 41e97) 165 (66.8) 47 (19.4) 50 (20.7) 61 (25.2) 50 (20.7) 34 (14.0) 70 (28.7) 61 (25.0) 64 (26.2) 49 (20.1) 53 (25.5) 62 (29.8) 93 (44.7) 184 (75.4) 1.79 (2.31, 0e12) 129 (52.4) 84 (34.4)

Yes

202 (81.8)

Abbreviations: PASE: Physical Activity Scale for the Elderly, ERA: Expectations Regarding Aging. Note: Missing data from: 1 8 (3.2%) participants; 2 10 (4.0%) participants; 3 7 (2.8%) participants; 4 3 (1.2%) participants; 5 5 (2.0%) participants; 6 24 (9.7%) participants; 7 2 (.8%) participants; 8 39 (15.8%) participants; 9 1 (.4%) participant.

Overall ERA

Physical ERA Mental ERA Cognitive ERA

.15 (237)* .18 (237)** Lawn work or yard carea Caring for another .01 (235) .07 (235) persona Home repairsa .06 (237) .11 (237) Heavy housework or .04 (237) .10 (237) a chores a Light housework .04 (238) .14 (238)* .11 (238) .14 (238)* Outdoor gardeninga .01 (242) .09 (242) Muscle strength and enduranceb Strenuous sport and .15 (237)* .24 (237)** recreationb Moderate sport and .10 (235) .15 (235)* recreationb Light sport and .07 (237) .14 (237)* recreationb Active job or .12 (238)* .14 (238)* volunteerb b Walking outside .01 (241) .04 (241) Overall PASE scoreb .19 (222)** .25 (222)**

.12 (237)

.07 (237)

.05 (235)

.01 (235)

.05 (237) .04 (237)

.01 (237) .05 (237)

.01 (238) .11 (238) .04 (242)

.03 (238) .03 (238) .02 (242)

.05 (237)

.08 (237)

.10 (235)

.01 (235)

.04 (237)

.01 (237)

.07 (238)

.13 (238)

.01 (241) .13 (222)

.02 (241) .09 (222)

Abbreviations: ERA: Expectations Regarding Aging, PASE: Physical Activity Scale for the Elderly. **p < .001, *p < .05. a Biserial correlation coefficients (sample size) for associations between continuous and dichotomous ordinal data. b Pearson correlation coefficients (sample size) for two continuous variables.


physical activity measures) correlation coefficients revealed significant associations between overall aging expectations and four modes of physical activity: Overall PASE scores (r ¼ .19, p < .001, n ¼ 222), strenuous sport and recreational activities (r ¼ .15, p < .05, n ¼ 237), lawn work or yard care (rb ¼ .15, p < .05, n ¼ 237), and engaging in an active job or volunteer position (r ¼ .12, p < .05, n ¼ 238). Regarding the correlations between the aging expectation sub-scales and each mode of physical activity, eight statistically significant associations were found for the physical health subscale only, while none were significant for either of the mental health or cognitive function aging expectation sub-scales. More specifically, physical health aging expectations were significantly related to the following eight modes of physical activity: Overall PASE scores (r ¼ .25, p < .001, n ¼ 222), strenuous, moderate, and light sport and recreational activities (r ¼ .24, p < .001, n ¼ 237; r ¼ .15, p < .05, n ¼ 235; r ¼ .14, p < .05, n ¼ 237, respectively), lawn work or yard care (rb ¼ .18, p < .001, n ¼ 237), engaging in an active job or volunteer position (r ¼ .14, p < .05, n ¼ 238), light housework (rb ¼ .14, p < .05, n ¼ 238), and outdoor gardening (rb ¼ .14, p < .05, n ¼ 238). It is important to note that four of these eight modes of physical activity were also associated with overall aging expectations (i.e., overall PASE scores, strenuous sport and recreational activities, lawn work or yard care, and engaging in an active job or volunteer position) and that the effect sizes involving these four modes of physical activity were stronger for physical health aging expectations than for overall aging expectations.

573

statistically significant (i.e., strenuous sport and recreational activities, OR ¼ 1.01, CI ¼ 1.00e1.02, n ¼ 232) while three were marginally significant (i.e., overall PASE score, OR ¼ 1.50, CI ¼ .99e 2.28, n ¼ 217; lawn work or yard care, OR ¼ 1.01, CI ¼ .99e1.03, n ¼ 232; and outdoor gardening, OR ¼ 1.01, CI ¼ .99e1.02, n ¼ 233). Three of these four associations (i.e., strenuous sport and recreational activities, lawn work or yard care, and outdoor gardening) maintained their effect sizes and levels of significance after adjusting for education, household income, and employment status (Table 4, Model 3); however, overall PASE scores were no longer marginally associated with physical health aging expectations after adjusting for these socio-economic factors (p > .05, Table 4, Model 3). In the final model that adjusted for depressive symptoms, pain, restriction of activities, and chronic disease, the previously marginal associations between physical health aging expectations and engaging in lawn work or yard care and outdoor gardening were no longer present (p > .05). Additionally, the previously significant association between physical health aging expectations and strenuous sport and recreational activities found in Model 3 was only marginally significant in the final model (OR ¼ 1.01, CI ¼ .99e1.03, n ¼ 191). Inspection of this latter association using post hoc linear regression analyses revealed a significant positive association between physical health aging expectations and strenuous sport and recreational activities after controlling for all variables except restriction of daily activities (Table 5, OR ¼ 1.01, CI ¼ 1.00e1.02, n ¼ 194). For every one unit increase in physical health aging expectations (i.e., a scale that ranged from zero to 100) there was a 1% increase in the likelihood that participants reported engaging in strenuous sport and recreational activities among participants who did not report a restriction of daily activities.

Multivariate regression analyses The significant correlations described above were explored further with multivariate regression analyses. Of the four physical activities associated with overall aging expectations (i.e., overall PASE scores, strenuous sport and recreational activities, lawn work or yard care, and engaging in an active job or volunteer position), after adjusting for age, gender, and ethnicity, only one of these associations remained statistically significant (i.e., lawn work or yard care, Table 3, Model 2, OR ¼ 1.01, CI ¼ 1.00e1.03, n ¼ 232). However, this association became non-significant after adjusting for education, household income, and employment status (Table 3, Model 3) and remained non-significant after adjusting for depressive symptoms, pain, restriction of activities, and chronic disease (Table 3, Model 4). Regarding the physical health aging expectation sub-scale and its association with eight physical activities (i.e., overall PASE scores, strenuous sport and recreational activities, lawn work or yard care, and engaging in an active job or volunteer position, light housework, outdoor gardening, and both light and moderate sport and recreational activities, Table 4), after adjusting for age, gender, and ethnicity (Model 2), only one of these associations remained

Discussion Among a heterogeneous sample of community-dwelling middle aged and older adults, it was anticipated that higher aging expectations would be associated with increased odds of being physically active across diverse modes of physical activity. Overall, the current study provides little support for this hypothesis as aging expectations were not widely associated with various physical activities commonly practiced by aging individuals. However, the novel and significant contributions of the current study can be found by comparing the various associations in the multivariate models that demonstrate the influence of aging expectations on physical activities is contextually based on the type of aging expectation, the mode of physical activity, and the functional abilities of the aging individual. The most notable finding was a positive association between physical health aging expectations and participation in strenuous sport and recreational activities among participants

Table 3 Bivariate and multivariate associations between overall aging expectations and practicing different modes of physical activity. Physical activity dependent variable

Model 1a OR

Lawn work or yard caree Strenuous sport and recreationf Active job or volunteerf Overall PASE scoref

1.02 1.01 1.02 2.07

(237) (237) (238) (222)

Model 2b CI

OR

1.00e1.03 1.00e1.02 1.00e1.04 1.24e3.46

1.01 1.01 1.01 1.21

(232) (232) (233) (217)

Model 3c CI

OR

1.00e1.03 .98e1.03 .97e1.05 .74e1.98

1.01 1.00 1.01 1.06

(194) (194) (195) (181)

Abbreviations: OR: odds ratio (sample size); CI: 95% confidence interval. a Model 1 represents the bivariate association. b Model 2 adjusts for demographics: age, gender, and ethnicity. c Model 3 adjusts for Model 2 and socio-economic status factors: education, household income, and employment status. d Model 4 adjusts for Model 3 and medical factors: depressive symptoms, pain, restriction of activities, and chronic disease. e ORs reported from logistic regression models. f ORs (exponentiated beta coefficients and CIs) reported from linear regression models.

Model 4d CI

OR

.98e1.03 .98e1.02 .98e1.04 .61e1.83

1.01 1.00 1.00 1.01

CI (191) (191) (194) (180)

.98e1.03 .97e1.03 .98e1.02 .39e1.41

574


Table 4 Bivariate and multivariate associations between physical health aging expectations and practicing different modes of physical activity. Physical activity dependent variable

Model 1a OR

Lawn work or yard caree Light houseworke Outdoor gardeninge Strenuous sport and recreationf Moderate sport and recreationf Light sport and recreationf Active job/volunteerf Overall PASE scoref

1.02 1.03 1.01 1.01 1.01 1.02 1.02 2.30

(237) (238) (238) (237) (235) (237) (238) (222)

Model 2b CI

OR

1.01e1.03 1.00e1.07 1.00e1.03 1.00e1.02 1.00e1.02 1.00e1.03 1.00e1.02 1.49e3.58

1.01 1.03 1.01 1.01 1.00 1.01 1.00 1.50

(232) (233) (233) (232) (230) (232) (233) (217)

Model 3c CI

OR

.99e1.03 .98e1.06 .99e1.02 1.00e1.02 .98e1.01 .98e1.03 .97e1.02 .99e2.28

1.01 1.03 1.01 1.01 1.00 1.00 1.00 1.21

(194) (195) (195) (194) (192) (193) (197) (181)

Model 4d CI

OR

.99e1.03 .98e1.08 .99e1.03 1.00e1.02 .98e1.02 .97e1.03 .98e1.01 .77e1.92

1.01 1.01 1.00 1.01 1.00 1.00 .99 1.11

CI (191) (192) (192) (191) (191) (192) (196) (180)

.98e1.03 .96e1.06 .98e1.03 .99e1.03 .98e1.02 .97e1.02 .98e1.01 .60e1.59

Abbreviations: OR: odds ratio (sample size); CI: 95% confidence interval. a Model 1 represents the bivariate association. b Model 2 adjusts for demographics: age, gender, and ethnicity. c Model 3 adjusts for Model 2 and socio-economic status factors: education, household income, and employment status. d Model 4 adjusts for Model 3 and medical factors: depressive symptoms, pain, restriction of activities, and chronic disease. e ORs reported from logistic regression models. f ORs (exponentiated beta coefficients and CIs) reported from linear regression models.

without restrictions of daily living, in particular. This finding was independent of age, gender, ethnicity, education, household income, employment status, depressive symptoms, reports of pain, and the presence of chronic disease. Observing the significance levels for each mode of physical activity and their relation to aging expectations across the step-wise multivariate models demonstrates the aforementioned contextual nature of the association between these two measures. To illustrate, the associations between physical health aging expectations and light housework, moderate and light sport and recreational activities, as well as having an active job or volunteer position were no longer significant after adjusting for age, gender, and ethnicity. These findings illustrate that these four associations were explained by demographic factors. Furthermore, the association between physical health aging expectations and overall physical activity (i.e., PASE scores) was explained by socioeconomic factors while the result for lawn and yard care as well as outdoor gardening were explained by medical factors. These confounding variables have been found to contribute to aging expectations and/or physical activity previously (e.g., Sarkisian, Liu, Ensrud, Stone, & Mangione, 2001; Sarkisian, Shunkwiler, Aguilar, & Moore, 2006; Trost et al., 2002). Unfortunately, these confounding factors are, to a large extent, non-modifiable; however, interventions can target different groups at different levels of these variables differently. As a result, the practical implications of the current findings suggest that specific types of physical activity practices may be promoted by improving aging expectations if interventions adopt specific participant inclusion criteria (e.g., functional ability). Future research is warranted to discern the specific participant inclusion characteristics with evidence-based results using validated and reliable measures. The positive association between physical health aging expectations and participating in strenuous sport and recreational activities among participants without restrictions of daily living is a new finding, but it is not surprising. Previous research that has Table 5 Post hoc multivariate models of the association between physical health aging expectations and participating in strenuous sport and recreational physical activities.

Model Model Model Model

4 4 4 4

without without without without

depressive symptoms pain chronic disease restriction of activities

OR

CI

n

1.00 1.00 1.01 1.01

.98e1.01 .98e1.01 .98e1.02 1.00e1.02

191 191 191 194

Abbreviations: OR: odds ratio; CI: 95% confidence interval. Note: ORs (exponentiated beta coefficients and CIs) reported from linear regression models.

examined aging expectations and physical activity used the Lorig Self-Management Behavior Exercise Survey (Sarkisian, Prohaska, et al., 2005). Rather than assessing diverse modes of physical activity, this particular instrument measures participation rates in aerobic physical activities only, such as walking for exercise, swimming, bicycling (both stationary and road), and ‘other’ aerobic exercises (Sarkisian, Prohaska, et al., 2005). According to the PASE scoring manual, these physical activities would be classified as ‘strenuous sport and recreational physical activities’, with the exception of walking (New England Research Institutes, 1991). These previous findings, in conjunction to the results found in the current study, suggest that the influence of aging expectations on physical activity may be restricted to strenuous physical activities rather than to physical activity in general. This finding may be a result of the fact that many aging adults cite advanced chronological age as a barrier against physical activity practices despite being physically and functionally able (O’Brien Cousins & Gillis, 2006). Also, this influence is even more pronounced with physical activities that are vigorous or strenuous in intensity due to the fear and risks of overexertion (O’Brien Cousins, 2000). Thus, negative aging expectations may result in someone thinking they are unable to engage in strenuous physical activities due to their advanced age and they may subsequently dis-identify and disengage with this mode of physical activity (Horton, 2010). Future research should explore these associations in greater detail to establish why, how, and when age-related beliefs predict strenuous types of physical activity. It is also important to note that the analyses involving physical activity and the aging expectation sub-scales demonstrated significant associations for physical health aging expectations only. Furthermore, when comparing the associations found between the modes of physical activity and overall verses physical health aging expectations, the significant associations observed with physical health aging expectations were more pronounced as results were greater in frequency (i.e., number of significance associations with p < .05), stronger in association (i.e., larger effect sizes), and more consistent across the step-wise multivariate models. These findings not only indicate that the physical health aging expectations are explaining the association between overall aging expectations and physical activities, but they also suggest a notable and important ‘matching effect’ found between specific types of aging expectations and their domain-specific outcomes. The matching effect observed between physical health aging expectations and strenuous sport and recreational activities may be useful for the design and implementation of specific and targeted strategies to help aging adults attain and/or maintain higher


physical activity levels. Perhaps an intervention designed to increase sport and recreational activity in this age group may be more effective if it also promotes realistic (i.e., more affirming) physical health aging expectations at the same time. Such interventions may be particularly advantageous given that physical health aging expectation scores were the lowest among the three sub-scales in this sample and engagement in strenuous sport and recreational activities was among the least reported modes of physical activity. These rather dismal scores indicate that psychosocial-behavioral interventions that are designed to promote physical activity and higher physical health aging expectations are justified and would be a valuable endeavor for future research (Nelson, Rejeski, Blair, et al., 2007; Vogel et al., 2009). The findings of the current study are limited by the fact that these data were self-reported that may result in social desirability or response bias, which is of particular importance to the assessment of physical activity. Also, due to the amount of data collected from participants, single-item indicators were used to measure a number of covariates rather than validated psychometric instruments in an effort to minimize survey non-response. Also, the use of an open-ended question to discern ethnicity resulted in relatively non-descriptive categories. Data were cross-sectional so reverse causality cannot be excluded. Furthermore, participants were community-dwelling so results may not represent those living in long-term care or in other institutionalized settings. Also, it may be possible that individuals with poorer health were less likely to participate, which may have generated a healthier sample that may not be representative of the overall aging population. In summary, the results of this study provided evidence that aging expectations of physical decline is a notable correlate of strenuous sport and recreational physical activity participation among functionally able adults in mid- and later-life. Future longitudinal and experimental research is required to authenticate these cross-sectional findings to discern whether low physical health aging expectations act as a risk factor of inactivity and functional decline and/or whether high aging expectations act as a promoting factor of physical activity and functional maintenance or achievement. Also, the findings involving the matching effect between different domains of aging expectations and domain-specific outcomes warrants further investigation to discern the extent to which this effect applies to other health- and illness-related measures. With this knowledge, effective strategies may be developed and implemented to increase the likelihood of successful aging among our population. References Baker, J., Meisner, B. A., Logan, A. J., Kungl, A. M., & Weir, P. L. (2009). Physical activity and successful aging in Canadian older adults. Journal of Aging & Physical Activity, 17, 223e235. Retrieved from. http://journals.humankinetics.com/japa. Bonnefoy, M., Normand, S., Pachiaudi, C., Lacour, J. R., Laville, M., & Kostka, T. (2001). Simultaneous validation of ten physical activity questionnaires in older men: a doubly labeled water study. Journal of the American Geriatrics Society, 49, 28e35. http://dx.doi.org/10.1046/j.1532-5415.2001.49006.x. Brink, T. L., Yesavage, J. A., Lum, O., Heersema, P., Adey, M. B., & Rose, T. L. (1982). Screening tests for geriatric depression. Clinical Gerontologist, 1, 37e44. Retrieved from. http://www.tandf.co.uk/journals/WCLI. Craig, C. L., Russell, S. J., Cameron, C., & Bauman, A. (2004). Twenty-year trends in physical activity among Canadian adults. Canadian Journal of Public Health, 95, 59e63. Retrieved from. http://www.cpha.ca/en/cjph.aspx. Depp, C. A., & Jeste, D. V. (2006). Definitions and predictors of successful aging: a comprehensive review of larger quantitative studies. The American Journal of Geriatric Psychiatry, 14, 6e20. Retrieved from. http://focus.psychiatryonline.org. Harada, N. D., Chiu, V., King, A. C., & Stewart, A. L. (2001). An evaluation of three selfreport physical activity instruments for older adults. Medicine and Science in Sports and Exercise, 33, 962e970. Retrieved from. http://journals.lww.com/ acsm-msse. Horton, S. (2010). Masters athletes as role models? Combating stereotypes of aging. In J. Baker (Ed.), The masters athlete: Understanding the role of sport and exercise in optimizing aging (pp. 122e136). London: Routledge.

575

Kim, S. H. (2009). Older people’s expectations regarding ageing, health-promoting behaviour and health status. Journal of Advanced Nursing, 65, 84e91. http:// dx.doi.org/10.1111/j.1365-2648.2008.04841.x. King, A. C., & King, D. K. (2010). Physical activity for an aging population. Public Health Reviews, 32, 1e20. Retrieved from. http://www.publichealthreviews.eu. Levy, B. (1996). Improving memory in old age by implicit self-stereotyping. Journal of Personality and Social Psychology, 71, 1092e1107. http://dx.doi.org/10.1037/ 0022-3514.71.6.1092. Levy, B. R. (2003). Mind matters: cognitive and physical effects of aging self-stereotypes. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 58B, P203eP211. http://dx.doi.org/10.1093/geronb/58.4.P203. Levy, B. R., & Leifheit-Limson, E. (2009). The stereotype-matching effect: greater influence on functioning when age stereotypes correspond to outcomes. Psychology and Aging, 24, 230e233. http://dx.doi.org/10.1037/a0014563. Levy, B. R., & Myers, L. M. (2004). Preventive health behaviours influenced by selfperceptions of aging. Preventive Medicine, 39, 625e629. http://dx.doi.org/ 10.1016/j.ypmed.2004.02.029. Lim, K., & Taylor, L. (2005). Factors associated with physical activity among older people: a population-based study. Preventive Medicine, 40, 33e40. http:// dx.doi.org/10.1016/j.ypmed.2004.04.046. Meisner, B. A. (2012). A meta-analysis of positive and negative age stereotype priming effects on behavior among older adults. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 67B, 13e17. http://dx.doi.org/ 10.1093/geronb/gbr062. Meisner, B. A., & Baker, J. (2013). An exploratory analysis of aging expectations and health care behavior among aging adults. Psychology and Aging, 28, 99e104. http://dx.doi.org/10.1037/a0029295. Meisner, B. A., Dogra, S., Logan, A. J., Baker, J., & Weir, P. L. (2010). Do or decline: comparing the effects of physical inactivity on biopsychosocial components of successful aging. Journal of Health Psychology, 15, 688e696. http://dx.doi.org/ 10.1177/1359105310368184. Menec, V. H. (2003). The relation between everyday activities and successful aging: a 6-year longitudinal study. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 58B, 74e82. http://dx.doi.org/10.1093/geronb/ 58.2.S74. Nelson, M. E., Rejeski, W. J., Blair, S. N., Duncan, P. W., Judge, J. O., King, A. C., et al. (2007). Physical activity and public health in older adults: recommendation from the American College of Sports Medicine and the American Heart Association. Circulation, 16, 1094e1105. http://dx.doi.org/10.1161/circulationaha.107.185650. New England Research Institutes. (1991). Physical activity scale for the elderly: Administration and scoring instruction manual. Watertown, MA: New England Research Institutes Inc. O’Brien Cousins, S. (2000). “My heart couldn’t take it”: older women’s beliefs about personal risks in physical exercise. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 55B, P283eP294. O’Brien Cousins, S. (2003). A self-referent thinking model: how older adults may talk themselves out of being physically active. Health Promotion Practice, 4, 439e448. O’Brien Cousins, S., & Gillis, M. (2006). “Just do it”: voices of active and inactive older adults thinking about physically active lifestyles. Psychology of Sport & Exercise, 6, 313e334. http://dx.doi.org/10.1016/j.psychsport.2004.03.001. Ory, M., Hoffman, M. K., Hawkins, M., Sannerc, B., & Mockenhaupt, R. (2003). Challenging aging stereotypes: strategies for creating a more active society. American Journal of Preventive Medicine, 25, 164e171. http://dx.doi.org/10.1016/ S0749-3797(03)00181-8. Parker, M. G., & Thorslund, M. (2007). Health trends in the elderly population: getting better and getting worse. The Gerontologist, 47, 150e158. http:// dx.doi.org/10.1093/geront/47.2.150. Rowe, J. W., & Kahn, R. L. (1987). Human aging: usual and successful. Science, 237, 143e149. http://dx.doi.org/10.1126/science.3299702. Rowe, J. W., & Kahn, R. L. (1998). Successful aging. New York: Pantheon Books. Sarkisian, C. A., Hays, R. D., & Mangione, C. M. (2002). Do older people expect to age successfully? The association between expectations regarding aging and beliefs regarding healthcare seeking among older people. The Journal of American Geriatrics Society, 50, 1837e1843. http://dx.doi.org/10.1046/j.1532-5415.2002. 50513.x. Sarkisian, C. A., Liu, H., Ensrud, K. E., Stone, K. L., & Mangione, C. M. (2001). Correlates of attributing new disability to old age. Journal of the American Geriatrics Society, 49, 134e141. http://dx.doi.org/10.1046/j.1532-5415.2001.49033.x. Sarkisian, C. A., Prohaska, T. R., Wong, M. D., Hirsch, S., & Mangione, C. M. (2005). The relationship between expectations for aging and physical activity among older people. Journal of General Internal Medicine, 20, 911e915. http:// dx.doi.org/10.1111/j.1525-1497.2005.0204.x. Sarkisian, C. A., Shunkwiler, S. M., Aguilar, I., & Moore, A. A. (2006). Ethnic differences in expectations for aging among older people. The Journal of American Geriatrics Society, 54, 1277e1282. http://dx.doi.org/10.1111/j.1532-5415.2006.00834.x. Sarkisian, C. A., Steers, W. N., Hays, R. D., & Mangione, C. M. (2005). Development of the 12-item Expectations Regarding Aging survey. The Gerontologist, 45, 240e248. http://dx.doi.org/10.1093/geront/45.2.240. Statistics Canada. (2008). Appendix C: Comparison of ethnic origins disseminated in 2006, 2001 and 1996. Accessed online: http://www12.statcan.gc.ca/censusrecensement/2006/ref/dict/app-ann003-eng.cfm#15. Thurston, M., & Green, K. (2004). Adherence to exercise in later life: how can ‘exercise on prescription’ programmes be made more effective? Health Promotion International, 19, 379e386 http://dx.doi.org/10.1093/heapro/dah311.

576


Trost, S. G., Owen, N., Bauman, A. E., Sallis, J. F., & Brown, W. (2002). Correlates of adults’ participation in physical activity: review and update. Medicine and Science in Sports and Exercise, 34, 1996e2001. Retrieved from. http://journals.lww. com/acsm-msse. Vaillant, G. E., & Mukamal, K. (2001). Successful aging. American Journal of Psychiatry, 158, 839e947. Retrieved from. http://ajp.psychiatryonline.org. Vogel, T., Brechat, P. H., Lepretre, P. M., Kaltenbach, G., Berthel, M., & Lonsdorfer, J. (2009). Health benefits of physical activity in older patients: a review. International Journal of Clinical Practice, 63, 303e320. http://dx.doi.org/10.1111/ j.1742-1241.2008.01957.x. Washburn, R. A., McAuley, E., Katula, J., Mihalko, S. L., & Boileau, R. A. (1999). The Physical Activity Scale for the Elderly (PASE): evidence for validity. Journal of Clinical Epidemiology, 52, 643e651. http://dx.doi.org/10.1016/S0895-4356(99) 00049-9.

Washburn, R. A., Smith, K. W., Jette, A. M., & Janney, C. A. (1993). The Physical Activity Scale for the Elderly (PASE): development and evaluation. Journal of Clinical Epidemiology, 46, 153e162. http://dx.doi.org/10.1016/0895-4356(93)90053-4. Wen, C. P., Wai, J. P., Tsai, M. K., Yang, Y. C., Cheng, T. Y., Lee, M. C., et al. (2011). Minimum amount of physical activity for reduced mortality and extended life expectancy: a prospective cohort study. The Lancet, 378, 1244e1253. http:// dx.doi.org/10.1016/S0140-6736(11)60749-6. Werblow, A., Felder, S., & Zweifel, P. (2007). Population ageing and health care expenditure: a school of ‘red herrings’? Health Economics, 16, 1109e1126 http:// dx.doi.org/10.1002/hec.1213. Yesavage, J. A., Brink, T. L., Rose, T. L., Lum, O., Huang, V., Adey, M. B., et al. (1983). Development and validation of a geriatric depression screening scale: a preliminary report. Journal of Psychiatric Research, 17, 37e49. Retrieved from. www. stanford.edu.