Behavioural Science Section / Original Paper Gerontology 2014;60:65–72 DOI: 10.1159/000354538
Received: April 17, 2013 Accepted: July 17, 2013 Published online: October 25, 2013
Successful Aging in Canada: Prevalence and Predictors from a Population-Based Sample of Older Adults Xiangfei Meng a Carl D’Arcy b a
Department of Psychiatry, College of Medicine, and b Department of Psychiatry and School of Public Health, University of Saskatchewan, Saskatoon, Sask., Canada
Abstract Background: Little research has been conducted to thoroughly explore the prevalence and predicators of successful aging (SA) from a national point of view. Objectives: The objectives of this study were (1) to estimate the prevalence of SA as defined by Rowe and Kahn using a large populationbased dataset and (2) to determine the roles of sociodemographic, psychological, and lifestyle factors in SA among Canadian seniors. Methods: Data was from the Canadian Community Health Survey: Healthy Aging (CCHS-HA) (n = 25,864) conducted in 2008–2009. Rowe and Kahn’s concept was used to measure SA. Descriptive analyses were used to estimate the prevalence of SA in those aged 45+ living in private dwellings in the ten provinces of Canada. Multivariate logistic regression was used to assess predicators of SA among those aged 65+ and having complete data on cognition. Results: The prevalence of SA was 50.1% among those aged 50+, then decreased to 46.2% for those aged 55+, 42.0% for those aged 60+, and 37.2% among those aged 65+. Assuming those 65+ living in institutions as unsuccessful agers, then the prevalence of SA was 35.3% among Canadian seniors aged 65+. There were no differences in prevalence rate of SA between males and females. We did
© 2013 S. Karger AG, Basel 0304–324X/14/0601–0065$39.50/0 E-Mail [email protected] www.karger.com/ger
not find higher income associated with SA. Being younger, married, regular drinkers, exercisers, perceived better health, satisfied with life, and taking calcium in the past month were associated with SA. The predicted probability of being a successful ager was 41% for those aged 65–74 years, 33% for 75–84 years, and 22% for those 85+ years, while controlling for other covariates. Presence of disease led to a major decline in SA, levels of functioning and engagement in contrast remained relatively constant. Conclusion: Over one third of the seniors in Canada met the criteria for SA, largely because the operationalization of the definition only considered severe chronic diseases that may not be well managed and thus likely to result in functional impairment. The modifiable risk factors identified provide direction for prevention efforts to increase SA at a population level. © 2013 S. Karger AG, Basel
Introduction
Population aging is a global phenomenon affecting both developed and developing countries. The percentage of those 60 and over has risen from 8% of the world’s The data reported in this paper were based on data from Statistic Canada. The opinions presented here do not represent the views of Statistic Canada.
Xiangfei Meng, PhD Department of Psychiatry University of Saskatchewan, Box 92, RUH 103 Hospital Drive, SK S7N 0W8 (Canada) E-Mail xiangfei.meng @ usask.ca
Downloaded by: NYU Medical Center Library 128.122.253.212 - 5/30/2015 6:03:08 PM
Key Words Aging · Epidemiology · Health aging · Risk factors
66
Gerontology 2014;60:65–72 DOI: 10.1159/000354538
Canadian provinces. Weir et al. [4] used the same data source, and found that age was a significant predictor of disability and impaired physical functioning. However, these studies did not have a direct measure of cognitive status nor did they have data on the Canadian population as a whole. The objectives of this study were (1) to estimate the prevalence of SA as defined by Rowe and Kahn using a large national population-based dataset and (2) to determine the roles of sociodemographic, psychological, and lifestyle factors in SA among Canadian seniors. An examination of national prevalence of SA as conceptualized by Rowe and Kahn is a constructive addition to the literature. First, it is important to look at national prevalence of SA with more recent and well-measured population-based data. This will provide a general profile of SA in Canada, and also add information to the worldwide literature on SA. Second, it provides a unique opportunity to evaluate the extent to which SA varies across a range of different sociodemographic, psychological, and lifestyle groups, as was done in this study. Finally, a detailed examination of SA using Rowe and Kahn’s concept may provide a basis for refining the concept and suggest appropriate changes in individual criteria of SA. Methods Data Source Data analyzed was from the Canadian Community Health Survey: Healthy Aging (CCHS-HA), which was a population-based, national cross-sectional survey that collected information about the factors, influences and processes that contribute to healthy aging focusing on health, social and economic determinants. This survey was conducted by Statistics Canada between December 2008 and November 2009 collecting data on those aged 45 and over living in private dwellings in the ten provinces of Canada. The response rate was 62.3% for a final sample of 25,864 respondents. The survey excluded residents of the three territories, some remote regions, institutions, Indian reserves or Crown lands, military bases, and full-time members of the Canadian Forces. The survey used a two-step strategy to allocate the sample to the provinces, based on census age and gender population distributions. The sample was subsequently proportionally allocated to rural and urban strata to ensure good representation. A survey weight was given to each survey respondent. This weight corresponds to the number of persons in the Canadian population that the respondent represents. The survey sample was weighted to represent the population 45 years of age and over living in the ten provinces of Canada at 2008–2009 [15, 16]. Study Sample for Predicators In this study, only those who had completed the Cognition Module of the survey and were aged 65 and over were used to explore predicators of SA (n = 8,154) (fig. 1).
Meng/D’Arcy
Downloaded by: NYU Medical Center Library 128.122.253.212 - 5/30/2015 6:03:08 PM
population (200 million) in 1950 to around 11% (760 million) in 2011, and has been projected to reach 22% (2 billion) by 2050 with the rate dramatically increasing [1]. The aging of Canada’s population is also accelerating. Around 14.4% of Canadians were aged 65 and over in 2011, and it is estimated this percentage will account for 23–25% by 2036. The old-old group (those aged 85+ years) is the fastest-growing age group in Canada [2, 3]. Population aging impacts healthcare, economics, education, employment, and social engagement, etc. [4]. Since Rowe and Kahn [5] distinguished ‘successful aging’ (SA) from ‘normal’ aging, there has been considerable attention paid to this concept [1, 4, 6–9]. Rowe and Kahn suggested that SA consisted of a combination of three components: (a) low probability of disease and diseaserelated disability, (b) high cognitive and physical functional capacity, and (c) active engagement with life [5]. The concept of SA developed by Rowe and Kahn has been generally accepted and used [4, 10]. However, Bowling and Dieppe [11] suggest that it is unrealistic to be completely disease-free in older age, and SA should be recognized as a continuum of achievement. Therefore, the list of major diseases considered in any definition of SA should only include major diseases likely to result in a big impact on a person’s life. Recently, a comprehensive review identified 28 studies with 29 different definitions for SA. The mean reported proportion of successful agers aged 60+ was 35.8%, but varied from 0.4 to 95%. Although the review identified the differences among these definitions of SA, several components were consistently identified, including absence of disability, physical functioning, social engagements, cognitive functioning, and absence of depression [12]. All these components are present in Rowe and Kahn’s model [5]. Though the criteria are similar, their operationalizations have been diverse. In terms of specific studies, a US study using five major diseases and disability, cognitive and physical functioning, social connections, and active engagement as the definition of SA, found that no greater than 11.9% of older adults (aged 65+) were aging successfully at four time points: 1998, 2000, 2002, and 2004 [13]. Using the US study’s [13] definition of SA, a cross-nation study found about 8.5% of Europeans reported SA with a substantial country variation in terms of percentages of SA, 21.1% of Danes were successful agers, while only 1.6% of Poles were successful agers [14]. Baker et al. [10] using a broader range of diseases to measure SA with data from the 2003 Canadian Community Health Survey (CCHS 2.1) found 11% of older adults were aging successfully in four
Exclude those without complete data on Cognition Module (n = 8,020) Survey sample with complete data on Cognition Module (n = 17,844) Exclude those who were aged between 45 and 64 years (n = 9,690) Study sample with complete data on Cognition Module and aged 65 and over (n = 8,154)
Fig. 1. Flowchart of the study sample.
Dependent Variables Using Rowe and Kahn’s conceptual framework, we defined SA as a multidimensional construct composed of (1) absence of major diseases, (2) high cognitive and physical functioning, and (3) active engagement with life. This operationalization of SA is similar to that used in some recent studies [5, 13, 17]. Absence of Major Diseases. Self-reported long-term diseases, which were expected to last or have already lasted 6 months or more that had been diagnosed by a health professional, were used to measure whether or not the respondent had a major disease. The diseases tapped were cancer, chronic lung disease, diabetes, heart disease, stroke, and depression. Respondents reporting any of these diseases were classified as having a major disease. High Cognitive and Physical Functioning. Respondents reporting no problems with instrumental activities of daily living (IADL) and basic activities of daily living (BADL) were judged to have met criteria for having an appropriate physical functioning. The CCHS-HA Cognition Module includes four cognitive tests: immediate and delayed recall for memory, and animal-naming and the Mental Alternation Test [18] for executive functioning. These tests were also used in other previous aging studies [19, 20]. Items were dichotomized as ‘having difficulties in cognition’ versus ‘having no difficulties in cognition’. Respondents who reported no limitations on cognition were considered as having appropriate cognitive functioning. Respondents with both appropriate physical and cognitive functioning were categorized as having met the second criteria for SA. Active Engagement with Life. Rowe and Kahn’s concept of active engagement refers to social connections and networking in productive way [21]. Participants were defined as ‘actively engaged’ if they reported (a) ‘had a job throughout last year’ or (b) ‘engaged in any type of community-related activity in the last year’. Respondents who met all three sets of criteria for SA were classified as successful agers, otherwise they were considered as normal agers. Independent Variables Sociodemographic factors (age, gender, marital status, place of residence, education, immigrant status, and personal and household income), psychological factors (self-perceived health and life
Successful Aging in Canada
satisfaction), and lifestyle factors (type of smokers, type of drinkers, physical activity and dietary supplement use) were examined as independent variables. Data Analysis The CCHS-HA survey used a multistage stratified sampling design. All results presented here used a bootstrap procedure recommended by Statistics Canada utilizing a set of 500 replicate sampling weights to correct for non-response and oversampling. All analyses were conducted using the Survey’s Master data files at the Saskatchewan Research Data Centre on the University of Saskatchewan campus. The prevalence of individual SA’s criterion and SA was calculated for the total sample 45+ as well as for those aged 65+ with complete data on cognition available. To determine if demographic, psychological, and lifestyle factors were statistically significant, unadjusted odds ratio (OR) and 95% confidence interval (CI) were also calculated. Multivariate logistic regression was used to examine predicators for SA among those aged 65+. The goodness of fit was tested. SAS software, version 9.2 (SAS Institute, Inc., Cary, N.C., USA) was used for all analyses.
Results
Characteristics of the Representative Population Table 1 shows demographic characteristics of the representative population. The CCHS-HA study has 25,864 participants, which represent the Canadian population of 13,633,927 aged 45+ at 2008–2009. The largest segment of the survey population were those aged 45–64 years, women, married, non-immigrant, having post-secondary graduation education, living in an urban area, making a middle-level income, regular drinkers, former smokers, exercisers, and those with self-perceived good health and life satisfaction. Prevalence of Successful Aging The prevalence of SA as defined in this analysis was 50.1% among community-dwelling Canadian residents aged 50+, then decreased to 46.2% for those aged 55+, 42.0% for those aged 60+, and 37.2% for those aged 65+ (fig. 2). In 2008–2009, about 238,000 Canadians aged 65 years and older lived in institutions [22]. Assuming these seniors in institutions to be non-successful agers, then the adjusted prevalence of SA among all Canadians aged 65+ was 35.3% at 2008–2009. Figure 2 also shows detailed information on proportions of older adults who met individual criterion of SA. There were 61.1% of those aged 50+ without any of the selected major diseases, but only 47.1% of seniors aged 65+ had none of the selected major diseases. There were minor decreasing trends in proportions of physical and cognitive functioning, and active engagement among older adults (table 2). Gerontology 2014;60:65–72 DOI: 10.1159/000354538
67
Downloaded by: NYU Medical Center Library 128.122.253.212 - 5/30/2015 6:03:08 PM
Survey sample with Cognition Module (n = 25,864)
Table 1. Characteristics of CCHS-HA covered population (aged
Absence of major diseases High physical and cognitive functioning Active social engagement Successful aging
45+) in 2008/2009 n
%
Total Age, years 45–64 65–74 75–84 85+ Gender Males Females Marital status Married/common law Never married Separated/divorced/widowed Immigrant status Yes No Education Less than secondary Secondary graduation Other post-secondary Post-secondary graduation Missing Place of residence Urban Rural Personal income, CAD
Received: April 17, 2013 Accepted: July 17, 2013 Published online: October 25, 2013
Successful Aging in Canada: Prevalence and Predictors from a Population-Based Sample of Older Adults Xiangfei Meng a Carl D’Arcy b a
Department of Psychiatry, College of Medicine, and b Department of Psychiatry and School of Public Health, University of Saskatchewan, Saskatoon, Sask., Canada
Abstract Background: Little research has been conducted to thoroughly explore the prevalence and predicators of successful aging (SA) from a national point of view. Objectives: The objectives of this study were (1) to estimate the prevalence of SA as defined by Rowe and Kahn using a large populationbased dataset and (2) to determine the roles of sociodemographic, psychological, and lifestyle factors in SA among Canadian seniors. Methods: Data was from the Canadian Community Health Survey: Healthy Aging (CCHS-HA) (n = 25,864) conducted in 2008–2009. Rowe and Kahn’s concept was used to measure SA. Descriptive analyses were used to estimate the prevalence of SA in those aged 45+ living in private dwellings in the ten provinces of Canada. Multivariate logistic regression was used to assess predicators of SA among those aged 65+ and having complete data on cognition. Results: The prevalence of SA was 50.1% among those aged 50+, then decreased to 46.2% for those aged 55+, 42.0% for those aged 60+, and 37.2% among those aged 65+. Assuming those 65+ living in institutions as unsuccessful agers, then the prevalence of SA was 35.3% among Canadian seniors aged 65+. There were no differences in prevalence rate of SA between males and females. We did
© 2013 S. Karger AG, Basel 0304–324X/14/0601–0065$39.50/0 E-Mail [email protected] www.karger.com/ger
not find higher income associated with SA. Being younger, married, regular drinkers, exercisers, perceived better health, satisfied with life, and taking calcium in the past month were associated with SA. The predicted probability of being a successful ager was 41% for those aged 65–74 years, 33% for 75–84 years, and 22% for those 85+ years, while controlling for other covariates. Presence of disease led to a major decline in SA, levels of functioning and engagement in contrast remained relatively constant. Conclusion: Over one third of the seniors in Canada met the criteria for SA, largely because the operationalization of the definition only considered severe chronic diseases that may not be well managed and thus likely to result in functional impairment. The modifiable risk factors identified provide direction for prevention efforts to increase SA at a population level. © 2013 S. Karger AG, Basel
Introduction
Population aging is a global phenomenon affecting both developed and developing countries. The percentage of those 60 and over has risen from 8% of the world’s The data reported in this paper were based on data from Statistic Canada. The opinions presented here do not represent the views of Statistic Canada.
Xiangfei Meng, PhD Department of Psychiatry University of Saskatchewan, Box 92, RUH 103 Hospital Drive, SK S7N 0W8 (Canada) E-Mail xiangfei.meng @ usask.ca
Downloaded by: NYU Medical Center Library 128.122.253.212 - 5/30/2015 6:03:08 PM
Key Words Aging · Epidemiology · Health aging · Risk factors
66
Gerontology 2014;60:65–72 DOI: 10.1159/000354538
Canadian provinces. Weir et al. [4] used the same data source, and found that age was a significant predictor of disability and impaired physical functioning. However, these studies did not have a direct measure of cognitive status nor did they have data on the Canadian population as a whole. The objectives of this study were (1) to estimate the prevalence of SA as defined by Rowe and Kahn using a large national population-based dataset and (2) to determine the roles of sociodemographic, psychological, and lifestyle factors in SA among Canadian seniors. An examination of national prevalence of SA as conceptualized by Rowe and Kahn is a constructive addition to the literature. First, it is important to look at national prevalence of SA with more recent and well-measured population-based data. This will provide a general profile of SA in Canada, and also add information to the worldwide literature on SA. Second, it provides a unique opportunity to evaluate the extent to which SA varies across a range of different sociodemographic, psychological, and lifestyle groups, as was done in this study. Finally, a detailed examination of SA using Rowe and Kahn’s concept may provide a basis for refining the concept and suggest appropriate changes in individual criteria of SA. Methods Data Source Data analyzed was from the Canadian Community Health Survey: Healthy Aging (CCHS-HA), which was a population-based, national cross-sectional survey that collected information about the factors, influences and processes that contribute to healthy aging focusing on health, social and economic determinants. This survey was conducted by Statistics Canada between December 2008 and November 2009 collecting data on those aged 45 and over living in private dwellings in the ten provinces of Canada. The response rate was 62.3% for a final sample of 25,864 respondents. The survey excluded residents of the three territories, some remote regions, institutions, Indian reserves or Crown lands, military bases, and full-time members of the Canadian Forces. The survey used a two-step strategy to allocate the sample to the provinces, based on census age and gender population distributions. The sample was subsequently proportionally allocated to rural and urban strata to ensure good representation. A survey weight was given to each survey respondent. This weight corresponds to the number of persons in the Canadian population that the respondent represents. The survey sample was weighted to represent the population 45 years of age and over living in the ten provinces of Canada at 2008–2009 [15, 16]. Study Sample for Predicators In this study, only those who had completed the Cognition Module of the survey and were aged 65 and over were used to explore predicators of SA (n = 8,154) (fig. 1).
Meng/D’Arcy
Downloaded by: NYU Medical Center Library 128.122.253.212 - 5/30/2015 6:03:08 PM
population (200 million) in 1950 to around 11% (760 million) in 2011, and has been projected to reach 22% (2 billion) by 2050 with the rate dramatically increasing [1]. The aging of Canada’s population is also accelerating. Around 14.4% of Canadians were aged 65 and over in 2011, and it is estimated this percentage will account for 23–25% by 2036. The old-old group (those aged 85+ years) is the fastest-growing age group in Canada [2, 3]. Population aging impacts healthcare, economics, education, employment, and social engagement, etc. [4]. Since Rowe and Kahn [5] distinguished ‘successful aging’ (SA) from ‘normal’ aging, there has been considerable attention paid to this concept [1, 4, 6–9]. Rowe and Kahn suggested that SA consisted of a combination of three components: (a) low probability of disease and diseaserelated disability, (b) high cognitive and physical functional capacity, and (c) active engagement with life [5]. The concept of SA developed by Rowe and Kahn has been generally accepted and used [4, 10]. However, Bowling and Dieppe [11] suggest that it is unrealistic to be completely disease-free in older age, and SA should be recognized as a continuum of achievement. Therefore, the list of major diseases considered in any definition of SA should only include major diseases likely to result in a big impact on a person’s life. Recently, a comprehensive review identified 28 studies with 29 different definitions for SA. The mean reported proportion of successful agers aged 60+ was 35.8%, but varied from 0.4 to 95%. Although the review identified the differences among these definitions of SA, several components were consistently identified, including absence of disability, physical functioning, social engagements, cognitive functioning, and absence of depression [12]. All these components are present in Rowe and Kahn’s model [5]. Though the criteria are similar, their operationalizations have been diverse. In terms of specific studies, a US study using five major diseases and disability, cognitive and physical functioning, social connections, and active engagement as the definition of SA, found that no greater than 11.9% of older adults (aged 65+) were aging successfully at four time points: 1998, 2000, 2002, and 2004 [13]. Using the US study’s [13] definition of SA, a cross-nation study found about 8.5% of Europeans reported SA with a substantial country variation in terms of percentages of SA, 21.1% of Danes were successful agers, while only 1.6% of Poles were successful agers [14]. Baker et al. [10] using a broader range of diseases to measure SA with data from the 2003 Canadian Community Health Survey (CCHS 2.1) found 11% of older adults were aging successfully in four
Exclude those without complete data on Cognition Module (n = 8,020) Survey sample with complete data on Cognition Module (n = 17,844) Exclude those who were aged between 45 and 64 years (n = 9,690) Study sample with complete data on Cognition Module and aged 65 and over (n = 8,154)
Fig. 1. Flowchart of the study sample.
Dependent Variables Using Rowe and Kahn’s conceptual framework, we defined SA as a multidimensional construct composed of (1) absence of major diseases, (2) high cognitive and physical functioning, and (3) active engagement with life. This operationalization of SA is similar to that used in some recent studies [5, 13, 17]. Absence of Major Diseases. Self-reported long-term diseases, which were expected to last or have already lasted 6 months or more that had been diagnosed by a health professional, were used to measure whether or not the respondent had a major disease. The diseases tapped were cancer, chronic lung disease, diabetes, heart disease, stroke, and depression. Respondents reporting any of these diseases were classified as having a major disease. High Cognitive and Physical Functioning. Respondents reporting no problems with instrumental activities of daily living (IADL) and basic activities of daily living (BADL) were judged to have met criteria for having an appropriate physical functioning. The CCHS-HA Cognition Module includes four cognitive tests: immediate and delayed recall for memory, and animal-naming and the Mental Alternation Test [18] for executive functioning. These tests were also used in other previous aging studies [19, 20]. Items were dichotomized as ‘having difficulties in cognition’ versus ‘having no difficulties in cognition’. Respondents who reported no limitations on cognition were considered as having appropriate cognitive functioning. Respondents with both appropriate physical and cognitive functioning were categorized as having met the second criteria for SA. Active Engagement with Life. Rowe and Kahn’s concept of active engagement refers to social connections and networking in productive way [21]. Participants were defined as ‘actively engaged’ if they reported (a) ‘had a job throughout last year’ or (b) ‘engaged in any type of community-related activity in the last year’. Respondents who met all three sets of criteria for SA were classified as successful agers, otherwise they were considered as normal agers. Independent Variables Sociodemographic factors (age, gender, marital status, place of residence, education, immigrant status, and personal and household income), psychological factors (self-perceived health and life
Successful Aging in Canada
satisfaction), and lifestyle factors (type of smokers, type of drinkers, physical activity and dietary supplement use) were examined as independent variables. Data Analysis The CCHS-HA survey used a multistage stratified sampling design. All results presented here used a bootstrap procedure recommended by Statistics Canada utilizing a set of 500 replicate sampling weights to correct for non-response and oversampling. All analyses were conducted using the Survey’s Master data files at the Saskatchewan Research Data Centre on the University of Saskatchewan campus. The prevalence of individual SA’s criterion and SA was calculated for the total sample 45+ as well as for those aged 65+ with complete data on cognition available. To determine if demographic, psychological, and lifestyle factors were statistically significant, unadjusted odds ratio (OR) and 95% confidence interval (CI) were also calculated. Multivariate logistic regression was used to examine predicators for SA among those aged 65+. The goodness of fit was tested. SAS software, version 9.2 (SAS Institute, Inc., Cary, N.C., USA) was used for all analyses.
Results
Characteristics of the Representative Population Table 1 shows demographic characteristics of the representative population. The CCHS-HA study has 25,864 participants, which represent the Canadian population of 13,633,927 aged 45+ at 2008–2009. The largest segment of the survey population were those aged 45–64 years, women, married, non-immigrant, having post-secondary graduation education, living in an urban area, making a middle-level income, regular drinkers, former smokers, exercisers, and those with self-perceived good health and life satisfaction. Prevalence of Successful Aging The prevalence of SA as defined in this analysis was 50.1% among community-dwelling Canadian residents aged 50+, then decreased to 46.2% for those aged 55+, 42.0% for those aged 60+, and 37.2% for those aged 65+ (fig. 2). In 2008–2009, about 238,000 Canadians aged 65 years and older lived in institutions [22]. Assuming these seniors in institutions to be non-successful agers, then the adjusted prevalence of SA among all Canadians aged 65+ was 35.3% at 2008–2009. Figure 2 also shows detailed information on proportions of older adults who met individual criterion of SA. There were 61.1% of those aged 50+ without any of the selected major diseases, but only 47.1% of seniors aged 65+ had none of the selected major diseases. There were minor decreasing trends in proportions of physical and cognitive functioning, and active engagement among older adults (table 2). Gerontology 2014;60:65–72 DOI: 10.1159/000354538
67
Downloaded by: NYU Medical Center Library 128.122.253.212 - 5/30/2015 6:03:08 PM
Survey sample with Cognition Module (n = 25,864)
Table 1. Characteristics of CCHS-HA covered population (aged
Absence of major diseases High physical and cognitive functioning Active social engagement Successful aging
45+) in 2008/2009 n
%
Total Age, years 45–64 65–74 75–84 85+ Gender Males Females Marital status Married/common law Never married Separated/divorced/widowed Immigrant status Yes No Education Less than secondary Secondary graduation Other post-secondary Post-secondary graduation Missing Place of residence Urban Rural Personal income, CAD
