Which Features of the Environment Impact Community Participation of Older Adults? A Systematic Review and Meta-Analysis

Abstract

Objective: The objective of this study is to identify features of the environment associated with community participation of older adults. Method: A systematic review of studies that examined associations between environment and community participation in older adults was conducted. Environmental features were extracted and grouped using the International Classification of Functioning, Disability, and Health’s environmental categories. Meta-analysis of environmental categories was conducted by calculating combined effect size (ES) estimates. Results: Significant, small to moderate, random ESs were found for six of seven environmental categories: neighborliness (ES = 0.52, 95% confidence interval [CI] = [0.18, 0.87]), social support (ES = 0.38, 95% CI = [0.13, 0.62]), land-use diversity (ES = 0.29, 95% CI = [0.16, 0.42]), transportation (ES = 0.29, 95% CI = [0.15, 0.43]), civil protection (e.g., safety; ES = 0.27, 95% CI = [0.0, 0.54]), and street connectivity/walkability (ES = 0.20, 95% CI = [0.15, 0.26]). Discussion: Community initiatives that address specific features of the social environment and street-level environment may increase community participation of older adults.

Keywords

aging in place International Classification of Functioning Disability and Health social support transportation

Introduction

Community participation, that is, engagement in activities occurring outside the home that are complex in nature, social and nondomestic (Chang, Coster, & Helfrich, 2013), is a key component of healthy aging (Hsu & Jones, 2012; Johnson & Mutchler, 2014; Rowe & Kahn, 1997). Community participation is a broad term referring to people’s engagement in social and community activities, such as getting together with family or friends, taking part in religious or recreational activities, working or volunteering, and taking part in social and cultural activities (Theis & Furner, 2011). As older adults develop mobility limitations or medical conditions with multiple comorbidities common to aging, for example, arthritis, respiratory disease, or diabetes, it can become increasingly difficult to participate in social and community activities (Ahem & Hendryx, 2008; Chang et al., 2013). Up to 20% of adults in the 2002 National Health Interview Survey (NHIS) who were above the age of 50 and had a chronic condition (e.g., arthritis, hypertension) reported community participation restrictions, with the prevalence of restrictions increasing for people with more comorbidities (Theis & Furner, 2011).

Physical, social, and policy features of the environment in which people live can facilitate or restrict people’s engagement in community activities (Freedman, Grafova, Schoeni, & Rogowski, 2008; Theis & Furner, 2011). For example, good street connectivity, level sidewalks, curb-cuts, transportation, positive attitudes toward people with disabilities, supportive relationships, and public programs and services may facilitate engagement in social and community activities as people age in place (Whyte, 1998; Lum et al., 2014; Theis & Furner, 2011). The importance of the person–environment “fit” is recognized as a key element of minimizing disability or participation restrictions (Whyte, 1998; Iwarsson, 2005; Lawton & Nahemow, 1973). For older adults, features of the environment may pose barriers to participation, particularly for people with more functional limitations, which are common with increasing age. For example, an older adult with arthritis who has difficulty walking for long distances may need more environmental supports (e.g., places to sit/rest, social support) to facilitate his or her community participation, as compared with an older adult with fewer limitations. Over the past 15 years, a growing number of efforts to foster older adults’ full inclusion in their community, many of which address person–environment fit characteristics, have been adopted by cities worldwide, spearheaded by the World Health Organization’s Checklist of Essentials of Age-Friendly Cities (Plouffe & Kalache, 2010; World Health Organization, 2007).

Despite the growing community initiatives, there is limited empirical evidence linking the environment to community participation among older adults. A few observational studies show that physical aspects of the environment (e.g., public transit, sidewalk conditions, places to sit and rest) and social aspects of the environment (e.g., social support, safety) are associated with community participation (Levasseur et al., 2011; Liu & Besser, 2003), whereas others show mixed associations (Beard et al., 2009; Bowling & Stafford, 2007; Haak, Fange, Horstmann, & Iwarsson, 2008; C. Hand, Law, Hanna, Elliott, & McColl, 2012; C. L. Hand, Wilkins, Letts, & Law, 2013; Keysor et al., 2010; Richard, Gauvin, Gosselin, & Laforest, 2009; Richard et al., 2013; Theis & Furner, 2011). In two recent scoping reviews, features of the environment were identified as being associated with participation of older adults, including safety, neighborhood mobility barriers, transportation, and proximity to resources in the community; however, neither an assessment of study quality nor synthesis of statistical results was conducted (C. Hand et al., 2012; Levasseur et al., 2015). Other similar reviews measure the environment-participation link in adults with chronic conditions, not in general older adults (C. Hand et al., 2012; C. L. Hand et al., 2013).

Multiple studies have used the World Health Organization’s International Classification of Functioning, Disability, and Health (ICF) environmental taxonomy as a framework to organize environmental features (Duggan, Albright, & Lequerica, 2008; Levasseur et al., 2015; Theis, Murphy, Hootman, & Wilkie, 2013; van Roekel et al., 2014). The taxonomy categorizes the environment into five domains: (a) products and technology (i.e., products, equipment, or technology that are part of the environment; for example, assistive devices, computers, or urban planning), (b) natural environment and human-made changes to environment (e.g., physical geography, population density, climate), (c) support and relationships (i.e., people or animals in an environment that provide physical or emotional support), (d) attitudes (i.e., attitudes of people in a person’s environment that are the result of cultural norms or other beliefs which may influence behavior or social life), and (e) services, systems, and policies (i.e., services, systems, and policies provided by society to meet the individuals’ needs or to enable control or organization). The environment in the ICF is classified at different levels of geographic detail, from immediate environment, such as street-level factors (e.g., curb ramps), to the broader environment, such as policy-level factors (e.g., transportation policy) to provide a comprehensive taxonomy useful for categorizing environmental features identified in the literature (World Health Organization, 2001).

To date, no study has systematically summarized and statistically combined the quantitative literature on the relationship between the environment and community participation of community-dwelling older adults, which could inform age-friendly community initiatives. Hence, the objective of this article is to identify features of the environment that are associated with community participation of community-dwelling older adults via rigorous systematic review and meta-analysis approaches. Based on the literature, we hypothesized that most environmental categories would show small but significant relationships with community participation of older adults.

Design and Methods

The systematic review and meta-analysis were conducted according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines (Moher, Liberati, Tetzlaff, & Altman, 2009). Comprehensive database searches were conducted in Web of Science, CINAHL, and PsycINFO on January 26, 2013; the search strategy was repeated using the three original databases and PubMed from February 19 to 20, 2014. The search terms were organized into population (e.g., aged, elderly, older adults), environment (e.g., neighborhood, environment, community, social factors), and participation (e.g., community participation, participation, social participation, social activities); a full electronic search strategy can be found in the appendix. Reference searches of relevant articles were conducted. Titles and abstracts were reviewed by one reviewer (MV); full article screening was then conducted by two reviewers (MV, RA).

Inclusion criteria for articles in the review were as follows: (a) adult samples above age 55; (b) examination of the neighborhood community; (c) assessment of the environment using census or administrative data, self-report, or observational approaches; (d) assessment of community participation using a broad operationalization of the term as defined as “active involvement in activities that are intrinsically social and either occur outside the home or are part of a nondomestic role” (Chang et al., 2013, p. 772); (e) cross-sectional or longitudinal study design that statistically examined the association between the environment and community participation; (f) published in 2001 or later (this date was chosen to focus on recent studies and based on the introduction of environment into the ICF framework); and (g) written in English. Exclusion criteria included (a) assessment of a specific type of community participation (e.g., volunteering) without a measure that broadly operationalized the domain; (b) studies that reported descriptive statistics of the environment or community participation, but not a statistical relationship; (c) studies whose study samples were exclusively disease-based populations (e.g., Parkinson’s disease), unless the disease was a prevalent chronic condition for older adults (e.g., arthritis).

Two reviewers (MV, RA) independently assessed the quality of the included studies and extracted relevant information during the review process. Discrepancies were resolved through discussion or with a third reviewer (JK). The eight-item Newcastle–Ottawa Scale (NOS) for cohort studies was used to rate the quality of cohort studies. The NOS, created using a Delphi method, has different scales for cohort and case-control studies and has been used in systematic reviews (Narducci, Waltz, Gorski, Leppla, & Donaldson, 2011; Wells, Smyth, & Brown, 2012). For cross-sectional studies, an eight-item checklist was used to assess bias (van Abbema et al., 2011; von Elm et al., 2007). Both rating scales’ items are found in Table 2. In addition, the following information was extracted from the selected articles: study year, study design, sample size, sample characteristics, environmental features, environmental assessment, type of community participation assessed, statistical methods, and statistical results. If more than one measure of community participation was reported in a study, the one that most closely aligned with the definition of community participation was used. Extracted features of the environment were organized into environmental categories within the ICF’s five environmental domains: (a) products and technology, (b) natural environment and human-made changes, (c) support and relationships, (d) attitudes, and (e) services, systems, and policies; discrepancies on classification were resolved with two outside reviewers (JK, ML). To statistically pool the quantitative results, environmental categories were required to be measured in no less than three studies.

Statistical Analyses

Study measures of association between environmental categories and community participation were converted into the standardized mean difference measure of effect to provide a consistent effect measure across the studies. These conversions were made using formulas from The Handbook of Research Synthesis and Meta-Analysis (Borenstein, 2009). The standardized mean difference effect size (ES) estimates were combined for each of the environmental categories using the DerSimonian and Laird random-effects method to arrive at a combined estimate and its 95% confidence interval (CI; DerSimonian & Laird, 1986). The null hypothesis value (no association) for standardized mean difference estimates is 0, so if the 95% CI excludes 0 the combined estimate is statistically significant and rejects the null hypothesis. Inconsistency in study results, or statistical heterogeneity, was evaluated by the Q test, which is used to test for extra variation in the study standardized mean difference values beyond what would be expected if all were estimating the same effect. The amount of extra variation was quantified using the I² estimate for inconsistency (Higgins & Thompson, 2002). Values of I² less than 0.25 were considered low heterogeneity, values between 0.25 and 0.75 were considered moderate heterogeneity, and values above 0.75 were considered high heterogeneity. Selective publication, or publication bias, was assessed by the visual evaluation of the symmetry of funnel plots and by use of the linear regression test of Egger (Egger, Davey Smith, Schneider, & Minder, 1997) An asymmetric funnel plot, or a significant result of an Egger test, would suggest that the meta-analytic combined estimates may be exaggerated from a lack of publication of nonsignificant study results. All tests were conducted with statistical significance at alpha level .05. Statistical analyses were performed using SAS (Version 9.3) and R (Version 3.2.0) statistical software packages.

Results

The initial search (January 2013) resulted in 575 publications, 11 of which met inclusion criteria; additional searches in February 2014 identified 720 publications, one of which was included (Figure 1). In total, 12 studies (one longitudinal, 11 cross-sectional) met eligibility criteria for the review and meta-analysis, including 8,694 older adults between 58 and 95 years old (Table 1).

Figure 1.

Process of study selection.

Table 1.

Studies Examining Environment and Community Participation of Older Adults.

Author, year, study type	Sample	Environmental feature(s)• Assessment approach/method (perceived, objective, or self-report; individual- or community-level data)	Environmental category (ICF domain)	Community participation (method of measurement)
Anaby et al. (2009), cross-sectional	N = 200, older adults (age 65-90 years old, M = 75 years) with two or more chronic conditions	Social support • Interpersonal Support Evaluation List (perceived; individual-level data)	Support and relationships (3)	Total daily activities (Assessment of Life Habits)
Beard et al. (2009), cross-sectional	N = 2,138, older adults age 65+ years	Land-use diversity, negative street characteristics, crime • New York City Mayor’s Management Report (objective; community level) Neighborhood collective efficacy • 10-item structured questionnaire assessing social cohesion and informal social control (perceived; individual level)	Land-use diversity (2), street connectivity and walkability (1), neighborliness (4), civil protection services (5)	Going outside the home disability^a (question on the long form of the 2000 U.S. Census: Q18c asking about disability affecting going outside the home)
Bowling and Stafford (2007), cross-sectional	N = 761, older adults age 65+ years	No nice places to walk, low perceived neighborliness, and living in a very unsafe neighborhood • U.K. national survey individual questions (perceived; community level)	Street connectivity and walkability (1), neighborliness (4), civil protection services (5)	Social participation (frequency of social contacts over the past month)
Dahan-Oliel, Mazer, Gelinas, Dobbs, and Lefebvre (2010), cross-sectional	N = 90, older adults age 65+ years (M = 76.3 years)	Lives with spouse • Individual question (self-report; individual level)	Living with or near to friends/family (3)	Participation (Craig Handicap Assessment & Reporting Technique [CHART], Nottingham Leisure Questionnaire)
Haak, Fange, Horstmann, and Iwarsson (2008), longitudinal	N = 314, older adults age 80+ years	Good access to shops/services, living where the action is, living close to friends or relatives, neighborliness, and having good local transport • ENABLE-AGE survey study questionnaire (perceived; community level)	Land-use diversity (2), street connectivity and walkability (1), living with/near friends/family (3), neighborliness (4), transportation services (5)	Togetherness-oriented participation^b (ENABLE-AGE survey questionnaire)
Hand, Law, Hanna, Elliott, and McColl (2012), cross-sectional	N = 248, older adults age 60+ years, with one or more chronic conditions	Neighborhood amenities Neighborhood Environment • Walkability Survey–Abbreviated version (perceived; individual level) Social support Medical Outcomes Study Social Support Survey (perceived; individual level) Neighborhood cohesion • Seven-item scale asking about trust/reciprocity among neighbors (perceived; individual level) Neighborhood problems • Six-item scale asking about traffic, noise, crime, air quality, litter/garbage, and graffiti (perceived; individual level) Neighborhood safety • Two-item scale asking about safety at night and during day (perceived; individual level)	Land-use diversity (2), support and relationships (3), neighborliness (4), civil protection services (5)	Satisfaction with participation (Keele Assessment of Participation)
Keysor et al. (2010), cross-sectional	N = 435, older adults age 65+ years, with knee osteoarthritis	Neighborhood mobility barriers, neighborhood transportation facilitators • Home and Community Environment survey (perceived; individual level)	Land-use diversity (2), transportation services (5)	Daily activity limitation (Late Life Disability Instrument)
Levasseur et al. (2011), cross-sectional	N = 554 (n = 296 females, n = 258 males) older adults age 65+ years—Stratified by gender	Greater perceived proximity to neighborhood resources • Perceived Accessibility to Key Resources for Older Adults scale (perceived; individual level)	Land-use diversity (2)	Social participation (Elderly Activity Inventory Questionnaire and Statistics Canada’s Participation and Activity Limitation Survey)
Liu & Besser (2003), cross-sectional	N = 2,802, older adults age 65+ years	Perceived support for community involvement • Three individual questions (perceived; individual level) Sense of community • Two individual-level questions (perceived; individual level)	Support and relationships (3), neighborliness (4)	Community participation (individual question of self-rated community involvement)
Richard, Gauvin, Gosselin, and Laforest (2009), cross-sectional	N = 282, older adults age 58+ years	Perceived accessibility to neighborhood resources • Perceived Accessibility to Key Resources for Older Adults (perceived; individual level) Greater perceived neighborhood walkability, lives with at least one other person or friends/relatives in neighborhood, excellent adequacy of neighborhood for a senior’s lifestyle, strong sense of belonging to the neighborhood, and bus stop/metro stop <5 min walk • Eight individual questions (perceived; individual level)	Land-use diversity (2), street connectivity and walkability (1), support and relationships (3), living with or near to friends/family (3), neighborliness (4), transportation services (5)	Social participation (Elderly Activity Inventory Questionnaire and Statistics Canada’s Participation and Activity Limitation Survey)
Richard et al. (2013), cross-sectional	N = 520, older adults age 68 to 84 years (M = 74.7 years)	Highest proportion of perceived services/amenities located <5 min walk • Perceived Accessibility to Key Resources for Older Adults (perceived; individual level) User friendliness of walking environment, perceived social support,^c and availability of a bus stop/metro stop <5 min walk • Individual questions (perceived; individual level)	Land-use diversity (2), street connectivity and walkability (1), support and relationships (3), neighborliness (4), transportation services (5)	Social participation (Elderly Activity Inventory Questionnaire and Statistics Canada’s Participation & Activity Limitation Survey)
Therrien & Desrosiers (2010), cross-sectional	N = 350 (n = 100 metropolitan, n = 200 urban, n = 50 rural), older adults age 65 to 95 years—Stratified by urban–rural classification	Access to stores, living with others, satisfaction with social support, access to public transportation, and feeling of security in the neighborhood • Questionnaire developed for study (perceived; individual level)	Land-use diversity (2), living with or near to friends/family (3), support and relationships (3), transportation services (5), civil protection services (5)	Participation (Assessment of Life Habits, Short Form 3.1) analyzed associations separately by city type

Note. ICF = International Classification of Functioning, Disability, and Health.

Going outside the home disability: Disability affecting going outside of the home.

Togetherness-oriented participation: “Sharing experiences with others and getting impressions from the outside world.”

Measured via items assessing availability of help in case of illness, disability, or problem, and (individually) the availability of someone who could take care of respondents (a) for a short period of time, (b) for a long period of time, and (c) from time to time.

Of the 12 studies, seven recruited participants from urban locations, including cities in Canada (Anaby et al., 2009; Dahan-Oliel, Mazer, Gelinas, Dobbs, & Lefebvre, 2010; Levasseur et al., 2011; Richard et al., 2009; Richard et al., 2013), the United States (Keysor et al., 2010), and Sweden (Haak et al., 2008). Two studies had a mix of urban and rural participants (Bowling & Stafford, 2007; Therrien & Desrosiers, 2010), and one study included only a rural sample (Liu & Besser, 2003). The majority of articles included participants greater than 65 years old; however, two studies used a lower age cutoff between 58 and 60 years (C. Hand et al., 2012; Richard et al., 2009), whereas another study included only people 80 years or older (Haak et al., 2008). While most studies did not include certain health diagnoses, two studies included older adults with common chronic conditions, for example, arthritis (Anaby et al., 2009; C. Hand et al., 2012), and one included participants with knee pain (Keysor et al., 2010). In addition, two studies calculated associations after stratifying their sample population into subgroups according to gender (Levasseur et al., 2011) or city size (Therrien & Desrosiers, 2010).

Environmental Features

Environmental features extracted from the articles were grouped into seven different environmental categories within the five ICF environmental domains. Land-use diversity and planning (n = 7), perceived social support (n = 6), and neighborliness (n = 6) were the most commonly measured environmental categories. Street connectivity and walkability (n = 5), living with or near family and friends (n = 4), transportation services (n = 5), and civil protection services (n = 4; for example, crime/safety) were examined, but less frequently (Table 1).

Community Participation Outcomes

Community participation was ascertained by self-report in all 12 studies, although a number of different assessment tools were used, including single-item census questions and more extensive multi-item surveys. Nine of the 12 studies used different approaches, whereas three studies used the Elderly Activity Inventory Questionnaire (Levasseur et al., 2011; Richard et al., 2009; Richard et al., 2013; Table 1).

Study Ratings

Study quality ranged from fair to good (Table 2). The cross-sectional studies’ failure to receive a full score was typically due to methodology weaknesses. Also, nearly half of the studies did not report clear inclusion or exclusion criteria, and several studies did not use validated assessment methods.

Table 2.

Quality Ratings of Studies.

Cross-sectional studies^a	1	2	3	4	5	6	7	8	Total score	Total possible
Anaby et al. (2009)	1	1	1	1	0	1	1	1	7	8
Beard et al. (2009)	1	0	1	1	0	1	1	1	6	8
Bowling & Stafford (2007)	1	0	1	0	0	1	1	1	5	8
Dahan-Oliel, Mazer, Gelinas, Dobbs, and Lefebvre (2010)	1	1	0	1	0	1	1	1	6	8
Hand, Law, Hanna, Elliott, and McColl (2012)	1	1	1	1	0	1	1	1	7	8
Keysor et al. (2010)	1	1	1	1	0	1	1	1	7	8
Levasseur et al. (2011)	1	0	1	1	0	0	1	1	5	8
Liu et al. (2003)	1	0	1	0	0	1	1	1	5	8
Richard, Gauvin, Gosselin, and Laforest (2009)	1	1	1	1	0	1	1	1	7	8
Richard et al. (2013)	1	1	1	1	0	1	1	1	7	8
Therrien & Desrosiers (2010)	1	1	0	1	0	1	1	1	6	8
Longitudinal studies^b	9	10	11	12	13	14	15	16	Total score	Total possible
Haak, Fange, Horstmann, and Iwarsson (2008)	1	1	1	0	0	0	0	1	4	8

Note. 0 = absent, 1 = present.

Questions 1 to 8: Checklist created by van Abbema et al. (2011) for cross-sectional studies. Low quality is a score less than 5 out of 8; high quality is 6 out of 8 or higher.

Questions 9 to 16: Newcastle–Ottawa Scale for cohort studies. 1: Source selection of participants is clear, and a representative sample of the population intended in the study was selected. 2: Inclusion and exclusion criteria were clearly described. 3: Methods used to ascertain exposure were valid and reliable. 4: Methods used to ascertain outcome were valid and reliable. 5: Researchers were blinded for test outcomes. 6: Appropriate univariate statistical method was used to establish the relationship between exposure and outcome. 7: Appropriate multivariate statistical methods were used to establish the relative contribution of the exposure to the outcome. 8: Statistical analysis included control for confounding factors. 9: Representation of the exposed cohort. 10: Selection of the nonexposed cohort. 11: Ascertainment of exposure. 12: Demonstration that the outcome of interest was not present at the time of study. 13: Comparability of cohorts on the basis of the design or analysis. 14: Assessment of outcome. 15: Was the follow-up long enough for outcomes to occur? 16: Adequacy of follow-up of cohorts.

Environmental Categories and Community Participation

Street connectivity/walkability

Environmental features that were considered part of the street connectivity or walkability category were measured in five studies (Beard et al., 2009; Bowling & Stafford, 2007; Haak et al., 2008; Richard et al., 2009; Richard et al., 2013)—These typically included street-level characteristics measured via a composite measure, such as sidewalk conditions, curb-cuts, places to sit/rest, and nice places to walk (Bowling & Stafford, 2007; Keysor et al., 2010). Four of the five studies found significant associations between walkability measures and community participation (Beard et al., 2009; Bowling & Stafford, 2007; Richard et al., 2009; Richard et al., 2013). The one study that did not find a significant association measured user friendliness of the walking environment (i.e., how easy it is to walk in their neighborhood) (Richard et al., 2013).

Land-use diversity

Six out of seven studies that showed a positive relationship between land-use diversity and community participation used a “quantity of social participation” measure; the one study that showed a nonsignificant relationship assessed “satisfaction with participation” (C. Hand et al., 2012). Thus, land-use diversity may be related to amount of community participation, but more research is needed to assess its relationship to satisfaction with participation.

Living with/near to friends/family

All four studies measuring the environmental category, “living with or near to friends/family,” found significant associations with community participation (Dahan-Oliel et al., 2010; Haak et al., 2008; Richard et al., 2009; Therrien & Desrosiers, 2010). However, one of these studies by Therrien & Desrosiers (2010) stratified by urban, rural, and metropolitan location and found only a significant relationship for people in rural areas.

Perceived social support

Five of the seven studies measuring perceived social support found significant associations. One of the two studies with insignificant findings measured social support as the availability of help or care in case of illness or disability (Richard et al., 2013), as compared with other studies, which measured perceived social support more broadly. The Therrien & Desrosiers (2010) study found insignificant associations regardless of city size (e.g., urban, metropolitan, or rural locations) (Therrien & Desrosiers, 2010).

Neighborliness

Five of the six studies measuring “neighborliness” found significant positive associations between neighborliness and community participation. Haak (2008) found that the association of perceived neighborliness with “togetherness-oriented participation,” defined as “sharing experiences with others and getting impressions from the outside world,” after 1 year in its sample of older adults greater than 80 years old was not statistically significant, but the estimate was in the positive direction (Haak, Fange, Horstmann, & Iwarsson, 2008). However, this study did find a significant association between living close to friends or relatives and participation.

Transportation services

Of the five studies measuring transportation services, the two that found significant associations with community participation were conducted with a more physically limited population (Keysor et al., 2010) or older sample than the other studies (Haak, Fange, Hortmann, & Iwarsson, 2008). Haak, Fange, Hortmann, & Iwarsson (2008) measured associations between “having good local transport” and togetherness-oriented participation in a sample of adults greater than 80 years old, and Keysor et al. (2010) measured associations between community participation and a composite transportation variable, including having personal transportation, nearby public transportation, available adaptations for people with physical limitations, and handicap parking availability in a sample of people with knee pain.

Civil protection services (safety)

Very mixed associations were found in the four studies measuring the relationship between civil protection services (e.g., safety, crime) and community participation. The two studies in which crime was significantly associated with community participation were the only studies conducted in a strictly urban setting (Beard et al., 2009; C. Hand et al., 2012); the other two were national studies that included both rural and urban samples from Canada and the United Kingdom (Bowling & Stafford, 2007; Therrien & Desrosiers, 2010).

Meta-Analysis of Associations Between Environmental Categories and Community Participation

With exception of the category, “Living with/near family and friends,” all environmental categories were found to have statistically significant, small to moderate, positive combined associations with community participation (Table 3). Neighborliness (ES = 0.52, 95% CI = [0.18, 0.87]) and perceived social support (ES = 0.38, 95% CI = [0.13, 0.62]) had the largest ESs, though the large CIs reflect limited power given the literature to date (Figure 2).

Table 3.

Statistical Results From Meta-Analysis.

ICF domain	Environmental category^a	Stratified studies	Publication bias p value	I² estimate	Hedges D Summary ES[95% CI]
Products and technology	Street connectivity or walkability	5	.21	0.0 ^b	0.20*** [0.15, 0.26]
Natural and built environment	Land-use diversity	10	.20	0.66^c	0.29*** [0.16, 0.42]
Support and relationships	Support and relationships, general (perceived social support)	8	.46	0.90^d	0.38** [0.13, 0.62]
Support and relationships	Living with/near to family or friends	6	.21	0.84^d	0.06 [−0.29, 0.40
Attitudes	Neighborliness	6	.42	0.99^d	0.52** [0.18, 0.87]
Services, systems, and policies	Transportation services, systems, and policies	6	.09	0.0^e	0.29*** [0.15, 0.43]
Services, systems, and policies	Civil protection services, systems, and policies (safety)	6	.30	0.95^d	0.27* [0.00, 0.54]

Note. ICF = International Classification of Functioning, Disability, and Health; ES = effect size; CI = confidence interval.

Terminology from the ICF environmental framework.

Heterogeneity p = .4

Heterogeneity p < .01.

Heterogeneity p < .001

Heterogeneity p = .5

p ≤ .05. ** p < .01. ***p < .001.

Figure 2.

Forest plot of effect sizes.

Inconsistency findings

We found considerable statistical heterogeneity in most categories with I² values above 0.65 and statistically significant Q-test results, indicating that the majority of the variation in study effect estimates is due to differences between studies. The exceptions to this heterogeneity finding were the categories of “Street connectivity or walkability” and “Transportation services, systems, and policies,” where no excess variation between study results was detected (Table 3).

Selective publication findings

The seven environmental categories did not demonstrate publication bias in either the funnel plot or the Egger test. However, although the Egger test result for “Transportation services, systems, and policies” does not reach statistical significance, it is close enough (p = .09) that there could be some concern that the statistically significant combined association of .29 for this category is somewhat inflated due to publication bias.

Discussion

This is the first systematic review and meta-analysis to be conducted on associations between environmental features and community participation of older adults. Small to moderate ESs were found between community participation and neighborliness, perceived social support, land-use diversity, street connectivity/walkability, civil protection services (e.g., safety or crime), and transportation services. An insignificant association was found for the category, “Living with or near to family or friends.” Significant heterogeneity was found for five of seven categories, indicating inconsistency in study results probably due, at least in part, to differences in study methodology. For example, we found differences between studies in age ranges, city size, as well as the methods used to assess participation and environmental categories. No studies tested positive for publication bias. Our meta-analysis results generally concur with previous reviews’ findings that features of the environment are significantly associated with community participation of older adults (C. L. Hand et al., 2013; Levasseur et al., 2015) and strengthen the empirical evidence supporting certain age-friendly initiatives (Plouffe & Kalache, 2010; World Health Organization, 2007). Specifically, community participation of older adults may be improved with initiatives that aim to improve social factors, including social support, neighborliness, and safety, as well as street-level factors, such as land-use diversity, street connectivity/walkability, and transportation services.

This review found wide variability in assessment methods of the environment, including census or administrative data, geographic information system data, and self-report measures using individual questions, which could explain the high degree of heterogeneity. Within certain environmental categories, such as land-use diversity, there does not appear to be clarity in terms of which items are needed to comprehensively measure a category. More consistency in measurement approaches of specific environmental categories (e.g., safety) via contemporary measurement techniques, such as item response theory or computer adapted testing, would move the field forward (Magasi et al., 2015; Whiteneck & Dijkers, 2009). Environmental assessment is complex and needs to reflect the research question, specific domains, and the context of the person–environment interaction; as a result, it presents substantial challenges to the field and potentially requires novel use of mapping and statistical modeling approaches.

Most of the current research measuring the environment–community participation relationship is cross-sectional. To better inform policy decisions, additional study designs to better measure the causality of associations between the environment and participation are needed, for example, prospective cohort or experimental studies. In addition, only two of the studies in this review assessed the more complex interaction of personal factors (e.g., sex, age) with the environment and community participation (Levasseur et al., 2011; Therrien & Desrosiers, 2010). Instead, most of the articles included in this review examined the specific association of one or more environmental feature with participation. As conceptualized in the ICF and other disablement models, engagement in life roles is influenced by the interaction of multiple factors, such as health, motivation, and environmental features, and future research should explore these relationships.

This study has several limitations. First, due to ambiguity in the ICF terminology and limited guidelines on mapping environmental features, subjective decisions were made to group environmental features into categories. This potential inconsistency, which may reflect a newly evolving field of study, could introduce misclassification of environmental factors. Second, this review summarized the role of individual environmental categories in participation, yet there may be summative effects of the environment on participation that were not examined in this review. Third, the interaction between environment categories and function of older adults in the face of community participation was not explored in this manuscript and may explain some of the heterogeneity found across studies. Fourth, formal testing of the independence of the ICF environmental domains has not been conducted; hence, there is no guidance as to whether analyzing each environmental category’s association with participation separately is a better approach than to group them into a composite measure (e.g., all social environmental categories under social environment). However, the method taken was intended to provide empirical evidence of the relationship between specific environmental categories to inform age-related community initiatives.

However, the use of the detailed ICF environmental framework to categorize the environment and conduct a systematic review and meta-analysis provides clarification to the mixed findings from the current literature and supports small to moderate associations between the environment and community participation. More importantly, the results of this review underlie the importance of considering the social and physical features of older adults’ environments and indicate that environmental initiatives may be necessary to increase community participation for older adults. Improving community participation outcomes and opportunities for healthy aging may require forging alliances and collaborations with planners, policymakers, and public health advocates. Additional research in this field is necessary to increase the depth of knowledge about specific features of the environment, the interactions between them, and their long-term impact on community participation of older adults.

Footnotes

Appendix

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors’ work is supported by NIDILRR center grant #HHS90RT5009 (Enhancing Activity & Participation among Persons with Arthritis (ENACT)) and NIAMS Multidisciplinary Clinical Research Center grant #5P60AR047785-15.

References

Ahem

M. M.

Hendryx

(2008). Community participation and the emergence of late-life depressive symptoms: Differences between women and men. Journal of Womens Health, 17(9), 1463-1470. doi:10.1089/jwh.2007.0752

Anaby

Miller

W. C.

Eng

J. J.

Jarus

, & PACC Research Group. (2009). Can personal and environmental factors explain participation of older adults? Disability and Rehabilitation, 31, 1275-1282. doi:10.1080/09638280802572940

Beard

J. R.

Blaney

Cerda

Frye

Lovasi

G. S.

Ompad

. . .Vlahov

(2009). Neighborhood characteristics and disability in older adults. Journals of Gerontology, Series B: Psychological Sciences & Social Sciences, 64, 252-257. doi:10.1093/geronb/gbn018

Borenstein

(2009). Effect sizes for continuous data. In Cooper

Hedges

L. V.

Valentine

J. C.

(Eds.), The handbook of research synthesis and meta-analysis (2nd ed., pp. 221-235). New York, NY: Russell Sage.

Bowling

Stafford

(2007). How do objective and subjective assessments of neighbourhood influence social and physical functioning in older age? Findings from a British survey of ageing. Social Science & Medicine, 64, 2533-2549. doi:10.1016/j.socscimed.2007.03.009

Chang

F. H.

Coster

W. J.

Helfrich

C. A.

(2013). Community participation measures for people with disabilities: A systematic review of content from an International Classification of Functioning, Disability, and Health perspective. Archives of Physical Medicine and Rehabilitation, 94, 771-781. doi:10.1016/j.apmr.2012.10.031

Dahan-Oliel

Mazer

Gelinas

Dobbs

Lefebvre

(2010). Transportation use in community-dwelling older adults: Association with participation and leisure activities. Canadian Journal on Aging/Revue Canadienne Du Vieillissement, 29, 491-502. doi:10.1017/S0714980810000516

DerSimonian

Laird

(1986). Meta-analysis in clinical trials. Controlled Clinical Trials, 7, 177-188. doi:10.1016/0197-2456(86)90046-2

Duggan

C. H.

Albright

K. J.

Lequerica

(2008). Using the ICF to code and analyse women’s disability narratives. Disability and Rehabilitation, 30, 978-990. doi:10.1080/09638280701797549

10.

Egger

Davey Smith

Schneider

Minder

(1997). Bias in meta-analysis detected by a simple, graphical test. British Medical Journal (Clinical Research Ed.), 315, 629-634.

11.

Freedman

V. A.

Grafova

I. B.

Schoeni

R. F.

Rogowski

(2008). Neighborhoods and disability in later life. Social Science & Medicine, 66, 2253-2267. doi:10.1016/j.socscimed.2008.01.013

12.

Haak

Fange

Horstmann

Iwarsson

(2008). Two dimensions of participation in very old age and their relations to home and neighborhood environments. American Journal of Occupational Therapy, 62, 77-86. doi:10.5014/ajot.62.1.77

13.

Hand

Law

McColl

M. A.

Hanna

Elliott

(2012). Neighborhood Influences on Participation among Older Adults with Chronic Health Conditions: A Scoping Review. OTJR: Occupation, Participation and Health, 32, 95-109.

14.

Hand

C. L.

Wilkins

Letts

L. J.

Law

M. C.

(2013). Renegotiating environments to achieve participation: A metasynthesis of qualitative chronic disease research. Canadian Journal of Occupational Therapy/Revue Canadienne D’Ergotherapie, 80, 251-262. doi:10.1177/0008417413501290

15.

Higgins

J. P.

Thompson

S. G.

(2002). Quantifying heterogeneity in a meta-analysis. Statistics in Medicine, 21, 1539-1558. doi:10.1002/sim.1186

16.

Hsu

H. C.

Jones

B. L.

(2012). Multiple trajectories of successful aging of older and younger cohorts. The Gerontologist, 52, 843-856. doi:10.1093/geront/gns005

17.

Iwarsson

(2005). A long-term perspective on person-environment fit and ADL dependence among older Swedish adults. The Gerontologist, 45, 327-336.

18.

Johnson

K. J.

Mutchler

J. E.

(2014). The emergence of a positive gerontology: From disengagement to social involvement. The Gerontologist, 54, 93-100. doi:10.1093/geront/gnt099

19.

Keysor

J. J.

Jette

A. M.

LaValley

M. P.

Lewis

C. E.

Torner

J. C.

Nevitt

M. C.

, . . . MOST Group. (2010). Community environmental factors are associated with disability in older adults with functional limitations: The MOST study. Journals of Gerontology, Series A: Biological Sciences & Medical Sciences, 65, 393-399. doi:10.1093/gerona/glp182

20.

Lawton

M. P.

Nahemow

(1973). Ecology and the aging process. In Lawton

M. P.

Eisdorfer

(Eds.), The psychology of adult development and aging (pp. 619-674). Washington, DC: American Psychological Association.

21.

Levasseur

Gauvin

Richard

Kestens

Daniel

, & NuAge Study Group. (2011). Associations between perceived proximity to neighborhood resources, disability, and social participation among community-dwelling older adults: Results from the VoisiNuAge study. Archives of Physical Medicine and Rehabilitation, 92, 1979-1986. doi:10.1016/j.apmr.2011.06.035

22.

Levasseur

Genereux

Bruneau

J. F.

Vanasse

Chabot

Beaulac

Bedard

M. M.

(2015). Importance of proximity to resources, social support, transportation and neighborhood security for mobility and social participation in older adults: Results from a scoping study. BMC Public Health, 15, Article 503. doi:10.1186/s12889-015-1824-0

23.

Liu

A.Q.

Besser

(2003). Social capital and participation in community improvement activities by elderly residents in small towns and rural communities. Rural Sociology, 68 (3), 343-365.

24.

Lum

T. Y.

Lou

V. W.

Chen

Wong

G. H.

Luo

Tong

T. L.

(2014). Neighborhood support and aging-in-place preference among low-income elderly Chinese city-dwellers. The Journals of Gerontology, Series B: Psychological Sciences & Social Sciences. Advance online publication. doi:10.1093/geronb/gbu154

25.

Magasi

Wong

Gray

D. B.

Hammel

Baum

Wang

C. C.

Heinemann

A. W.

(2015). Theoretical foundations for the measurement of environmental factors and their impact on participation among people with disabilities. Archives of Physical Medicine and Rehabilitation, 96, 569-577. doi:10.1016/j.apmr.2014.12.002

26.

Moher

Liberati

Tetzlaff

Altman

D. G.

(2009). Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. Annals of Internal Medicine, 151, 264-269. doi:10.7326/0003-4819-151-4-200908180-00135

27.

Narducci

Waltz

Gorski

Leppla

Donaldson

(2011). The clinical utility of functional performance tests within one-year post-ACL reconstruction: A systematic review. International Journal of Sports Physical Therapy, 6(4), 333-342.

28.

Plouffe

Kalache

(2010). Towards global age-friendly cities: Determining urban features that promote active aging. Journal of Urban Health: Bulletin of the New York Academy of Medicine, 87, 733-739. doi:10.1007/s11524-010-9466-0

29.

Richard

Gauvin

Gosselin

Laforest

(2009). Staying connected: Neighbourhood correlates of social participation among older adults living in an urban environment in Montreal, Quebec. Health Promotion International, 24, 46-57. doi:10.1093/heapro/dan039

30.

Richard

Gauvin

Kestens

Shatenstein

Payette

Daniel

. . .Mercille

(2013). Neighborhood resources and social participation among older adults: Results from the VoisiNuAge study. Journal of Aging and Health, 25, 296-318. doi:10.1177/0898264312468487

31.

Rowe

J. W.

Kahn

R. L.

(1997). Successful aging. The Gerontologist, 37, 433-440. doi:10.1093/geront/37.4.433

32.

Theis

K. A.

Furner

S. E.

(2011). Shut-in? Impact of chronic conditions on community participation restriction among older adults. Journal of Aging Research, 2011, 759158. doi:10.4061/2011/759158

33.

Theis

K. A.

Murphy

Hootman

J. M.

Wilkie

(2013). Social participation restriction among US adults with arthritis: A population-based study using the International Classification of Functioning, Disability, and Health. Arthritis Care & Research, 65, 1059-1069. doi:10.1002/acr.21977

34.

Therrien

Desrosiers

(2010). Participation of metropolitan, urban and rural community-dwelling older adults. Archives of Gerontology and Geriatrics, 51, E52-E56. doi:10.1016/j.archger.2009.11.017

35.

van Abbema

Lakke

S. E.

Reneman

M. F.

van der Schans

C. P.

van Haastert

C. J.

Geertzen

J. H.

Wittink

. (2011). Factors associated with functional capacity test results in patients with non-specific chronic low back pain: A systematic review. Journal of Occupational Rehabilitation, 21, 455-473. doi:10.1007/s10926-011-9306-4

36.

van Roekel

E. H.

Bours

M. J.

de Brouwer

C. P.

Ten Napel

Sanduleanu

Beets

G. L.

Weijenberg

M. P

. (2014). The applicability of the International Classification of Functioning, Disability, and Health to study lifestyle and quality of life of colorectal cancer survivors. Cancer Epidemiology, Biomarkers & Prevention: A Publication of the American Association for Cancer Research, Cosponsored by the American Society of Preventive Oncology, 23, 1394-1405. doi:10.1158/1055-9965.EPI-13-1144

37.

von Elm

Altman

D. G.

Egger

Pocock

S. J.

Gotzsche

P. C.

Vandenbroucke

J. P.

, & STROBE Initiative. (2007). The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: Guidelines for reporting observational studies. The Lancet, 370, 1453-1457. doi:10.1016/S0140-6736(07)61602-X

38.

Wells

S. A.

Smyth

Brown

T. G.

(2012). Patient attitudes towards change in adapted motivational interviewing for substance abuse: A systematic review. Substance Abuse and Rehabilitation, 3, 61-72. doi:10.2147/SAR.S23287

39.

Whiteneck

Dijkers

M. P.

(2009). Difficult to measure constructs: Conceptual and methodological issues concerning participation and environmental factors. Archives of Physical Medicine and Rehabilitation, 90(11, Suppl.), S22-S35. doi:10.1016/j.apmr.2009.06.009

40.

Whyte

(1998). Enabling America: A report from the Institute of Medicine on rehabilitation science and engineering. Archives of Physical Medicine and Rehabilitation, 79(11), 1477-1480. doi:10.1016/s0003-9993(98)90248-9

41.

World Health Organization. (2001). International Classification of Functioning, Disability, and Health. Geneva, Switzerland: Author.

42.

World Health Organization. (2007). Checklist of essential features of age-friendly cities. Retrieved from http://www.who.int/ageing/publications/Age_friendly_cities_checklist.pdf