Social Science & Medicine 104 (2014) 23e30

Contents lists available at ScienceDirect

Social Science & Medicine journal homepage: www.elsevier.com/locate/socscimed

Relationships between the perceived neighborhood social environment and walking for transportation among older adults Jelle Van Cauwenberg a, b, c, *,1, Liesbeth De Donder d,1, Peter Clarys a, Ilse De Bourdeaudhuij b, Tine Buffel e, Nico De Witte d, f, Sarah Dury d, Dominique Verté d, Benedicte Deforche a, b a Department of Human Biometry and Biomechanics, Faculty of Physical Education and Physical Therapy, Vrije Universiteit Brussel, Pleinlaan 2, B-1050 Brussels, Belgium b Department of Movement and Sport Sciences, Faculty of Medicine and Health Sciences, Ghent University, Watersportlaan 2, B-9000 Ghent, Belgium c Fund for Scientific Research Flanders (FWO), Egmontstraat 5, B-1000 Brussels, Belgium d Department of Educational Sciences, Faculty of Psychology and Educational Sciences, Vrije Universiteit Brussel, Pleinlaan 2, B-1050 Brussels, Belgium e School of Social Science, The University of Manchester, Oxford Road, Manchester M13 9PL, United Kingdom f University College Ghent, Keramiekstraat 78-80, 9000 Gent, Belgium

a r t i c l e i n f o

a b s t r a c t

Article history: Available online 18 December 2013

Ecological models state that physical activity (PA) behaviors can be explained by the interplay between individuals and their surrounding physical and social environment. However, the majority of research on PA-environment relationships has focused upon the physical environment. The purpose of the current study was to investigate the relationship between the perceived social environment and older adults’ walking for transportation, while adjusting for individual and perceived physical environmental factors. Questionnaires were used to collect data on walking for transportation, individual, perceived physical and social environmental factors in 50,986 Flemish older adults (65 years) in the period of 2004e2010. Multilevel logistic regression analyses were applied to examine the relationships between perceived social environmental factors and the odds of daily walking for transportation. The final models showed significant positive relationships for frequency of contacts with neighbors, neighbors’ social support, too many immigrants residing in the neighborhood, neighborhood involvement, participation, and volunteering. These results emphasize the need for including social environmental factors in future studies examining correlates of older adults’ physical activity. Current findings suggest that projects stimulating interpersonal relationships, place attachment, and formal community engagement might promote walking for transportation among older adults. Future research should try to further disentangle the complex (inter)relationships and causal mechanisms between older individuals, their environments, and their walking for transportation behavior. Ó 2014 Elsevier Ltd. All rights reserved.

Keywords: Belgium Ecological model Social environment Walking Physical activity Older adults

Introduction

* Corresponding author. Vrije Universiteit Brussel, Faculty of Physical Education and Physical Therapy, Department of Human Biometry and Biomechanics, Pleinlaan 2, B-1050 Brussels, Belgium. E-mail addresses: [email protected], [email protected] (J. Van Cauwenberg), [email protected] (L. De Donder), [email protected] (P. Clarys), [email protected] (I. De Bourdeaudhuij), tine.buffel@vub. ac.be (T. Buffel), [email protected] (N. De Witte), [email protected] (S. Dury), [email protected] (D. Verté), [email protected] (B. Deforche). 1 Shared first authorship. 0277-9536/$ e see front matter Ó 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.socscimed.2013.12.016

The physical, social and mental health benefits of regular physical activity (PA) for older adults (65 years) are well-established (Chodzko-Zajko et al., 2009). Given that the population of older adults is growing, with many suffering from chronic diseases and few (30e40%) accumulating enough PA to maintain health benefits, the promotion of PA in this age category is needed (Centers for Disease Control and Prevention, 2012; Eurobarometer, 2010). Walking represents an ideal activity to promote among older adults, as it is healthy, safe, accessible and well-liked (De Fré, De Martelaer, Philippaerts, Scheerder, & Lefevre, 2009; Manson et al., 2002; Murtagh, Murphy, & Boone-Heinonen, 2010). In particular, walking for transportation seems promising as it can be easily integrated into

24

J. Van Cauwenberg et al. / Social Science & Medicine 104 (2014) 23e30

older adults’ daily routines (e.g. walking to a shop). In order to promote walking for transportation among older adults, knowledge of its correlates is required (Baranowski, Anderson, & Carmack, 1998). According to social ecological models of health promotion, human well-being, behavior and environmental conditions are strongly interrelated. These models emphasize that the level of congruence between people and their surrounding environment is an important predictor of their health behaviors (e.g. PA) and wellbeing (Stokols, 1996). Social ecological models conceptualize the environment as a multi-dimensional concept, including physical and social aspects (Stokols, 1996). The physical environment can be defined as the characteristics of the physical context in which people spend their time (e.g. home, neighborhood), including aspects of urban design (e.g. presence of sidewalks), traffic density and speed, distance to and design of venues for PA (e.g. parks), crime and safety (Davison & Lawson, 2006). In conceptualizing the social environment, definitions vary from author to author, depending on the different perspectives of their research paradigms. As a consequence of these different approaches, there are certain inconsistencies in the concepts, which lead to theoretical and methodological ambiguity (Carpiano, 2006). Social epidemiological studies, for instance, often refer to social capital (Putnam, 2000), whereas others use related concepts such as social cohesion (Echeverria, Diez-Roux, Shea, Borrell, & Jackson, 2008), perceived neighborhood climate (Brown et al., 2011), or social neighborhood environment (Kamphuis et al., 2009). According to Buffel et al. (2012) the social environment can encompass several dimensions, including: interpersonal relationships (i.e. social networks and social support) (Brown et al., 2011; Forrest & Kearns, 2001; McNeill, Kreuter, & Subramanian, 2006), place attachment (i.e. social composition of the neighborhood and neighborhood involvement) (Brown et al., 2011), and formal community engagement (i.e. participation in organizations and volunteering) (Forrest & Kearns, 2001; Putnam, 1993, 2000). Social ecological models do not only describe the environment as multi-dimensional, they also underscore the behavior-specificity of environmental correlates (Alfonzo, 2005; Sallis et al., 2006). For example, the presence of a variety of shops in the neighborhood can be expected to stimulate walking for transportation, but might be less relevant for walking for recreation. Since 2000, PA-environment relationships have received considerable research attention (Rhodes & Nasuti, 2011). While much of the research on the relationship between environmental factors and health-related behaviors has focused on aspects of the physical environment, less is known about social environmental conditions and PA (Mahmood et al., 2012; McNeill et al., 2006). Furthermore, the vast majority of previous studies were conducted among younger populations, whereas studies on older adults are rather scarce (Rhodes & Nasuti, 2011; Van Cauwenberg et al., 2011). Nevertheless, several authors point to an increasing importance of the neighborhood environment as people age (Buffel et al., 2012), for a number of reasons. First, the long period of time spent in the same place, resulting in cumulative memories (Phillipson, 2007); second, the greater time spent at home and in the neighborhood following retirement (Peace, Wahl, Mollenkopf, & Oswald, 2007); third, increased reliance upon neighborhood relationships for support in old age (Krause, 2004); and fourth, the significance of the neighborhood in preserving a sense of identity for older adults (Rowles, 1983; Rubinstein & Parmelee, 1992). Furthermore, in order to understand issues of physical health in older adults, neighborhood environments appear to be key elements (Mahmood et al., 2012). Therefore, the current study focuses upon the relationships between three social environmental dimensions (interpersonal relationships, place attachment, and formal community engagement) and older adults’ walking for transportation.

The physical and social environment are no separate entities, they are closely interlinked and exert independent as well as joint effects on human behavior (Alfonzo, 2005; Stokols, 1996). Concerning the relationship between the physical environment and older adults’ walking for transportation, several studies have observed higher levels of walking for transportation in areas with easy access to destinations (e.g. shops, services, etc.) (Frank, Kerr, Rosenberg, & King, 2010; King et al., 2011; Salvador, Reis, & Florindo, 2010; Van Cauwenberg, Clarys, et al., 2012). For example, Frank et al. (2010) reported residents of high-walkable neighborhoods (many destinations within walking distance) to be twice as likely to walk for transportation compared to residents of low-walkable neighborhoods (few destinations within walking distance). For other physical environmental factors, such as presence and quality of walking facilities, traffic- and crime-related safety, and aesthetics, a systematic review concluded that current evidence is inconclusive (Van Cauwenberg et al., 2011). Although the positive effects of the social environment on older adults’ well-being (Shiovitz-Ezra & Litwin, 2012), physical health (Ahnquist, Wamala, & Lindstrom, 2012), and levels of mortality and morbidity (Berkman, Glass, Brissette, & Seeman, 2000) have long been recognized, research is only just beginning to unravel its relationship with older adults’ walking behaviors. A study in American older adults reported individual-level (but not neighborhood-level) social cohesion to be positively related to total walking (Mendes de Leon et al., 2009). However, another US study found that neighborhood-level social cohesion did explain differences in total walking between neighborhoods (Fisher, Li, Michael, & Cleveland, 2004). Studying correlates of recreational walking in Dutch older adults, Kamphuis et al. (2009) observed a negative relationship for social cohesion, but a positive relationship for size of social network. No relationships were found for feeling at home in the neighborhood and social disorganization. In a longitudinal study in older Hispanics living in the US, Brown et al. (2011) found that a positive perceived neighborhood climate was related to more total walking one year later. In another longitudinal study, Brown et al. (2008) reported that the presence of physical environmental factors that are believed to promote visual and social contacts among residents (such as stoops, front porches, windows.) positively predicted physical functioning three years later. The inconsistent findings in the above described studies point to the need for further research into the relationships between social environmental factors and older adults’ walking behaviors. Furthermore, although one of the core principles of social ecological models is the domain-specificity of correlates (Alfonzo, 2005; Sallis et al., 2006), none of the above studies focused specifically upon older adults’ walking for transportation. Physical and social environments are not only hypothesized to influence health behaviors, they are also interrelated and influence each other (Eicher & Kawachi, 2011; Frumkin, Frank, & Jackson, 2004; Stokols, 1996). Leyden (2003) found that adults living in high-walkable Irish neighborhoods reported higher levels of knowing their neighbors, political participation, trust in other people, and social participation compared to participants living in low-walkable neighborhoods. These findings held when walkability was defined by a subjective rating of the researcher as well as by participants’ perceptions. However, Wood et al. (2008) found the objectively measured number of destinations to be negatively related, but the perceived adequacy of local facilities positively related to social capital. Furthermore, they found the objectively measured presence of nearby bus stops to be related to lower levels, whereas nearby shops were related to higher levels of social capital. In Japanese older adults, no significant relationships were found between an objective walkability-index and any of six social capital indices (i.e. general trust, norms of reciprocity, place attachment,

J. Van Cauwenberg et al. / Social Science & Medicine 104 (2014) 23e30

two types of formal community engagement, and meeting friends) (Hanibuchi et al., 2012). In contrary, perceived accessibility to key resources (e.g. good quality and affordable food, leisure activities of interest, etc.) was positively related to social participation in Canadian older adults (Richard, Gauvin, Gosselin, & Laforest, 2009). To summarize, there appears to be a complex interplay between individuals, their surrounding physical and social environments, and their walking (for transportation) behavior. More research is needed to disentangle these complex relationships in order to inform policy makers and urban planners to design neighborhoods that optimally promote walking for transportation and health in older adults. The purpose of this paper is to address some of the current shortcomings of the literature. A first major gap concerns the inconsistent findings in the few previous studies examining the relationship between perceived social environmental characteristics and older adults’ walking behavior. Second, previous studies have not focused specifically upon relationships with older adults’ walking for transportation. A third limitation concerns the lack of consensus around the conceptualization and operationalization of the social environment, with most studies focusing on a composite measure, rather than the underlying dimensions. Finally, few studies have examined individual, physical and social environmental characteristics simultaneously. In response to these lacunae, this study investigates the relationship between the perceived social environment and older adults’ walking for transportation, adjusting for individual and perceived physical environmental factors. Relationships with factors pertaining to the three social environmental dimensions (interpersonal relationships, place attachment and formal community engagement) will be examined. Methods Procedures and participants The data for the present research were derived from the Belgian Ageing studies (BAS). Detailed information on the study protocol has been described previously (De Donder, De Witte, Buffel, Dury, & Verte, 2012; Van Cauwenberg, Clarys, et al., 2012; Verté, De Witte, & De Donder, 2007). This research project monitors issues of quality of life among community-dwelling older adults in order to support the implementation of evidenced-based policies. Data were collected by means of a structured questionnaire among 67,563 adults, aged 60 years and older, in 142 of the 308 municipalities in the Dutch-speaking part of Belgium (Flanders) between 2004 and 2010. Municipalities were not selected randomly. The provinces in Flanders organized an open call to offer local policy-makers and municipalities the possibility to participate in the research. Thereupon they could freely decide to participate. The participating 142 municipalities did not differ in average yearly income/inhabitant (16,452 euros) from the average of all 308 Flemish municipalities (16,323 euros), but were more densely populated (572 vs. 457 inhabitants/km2). In each of the participating municipalities, BAS applied a random sampling strategy stratified for age and gender to recruit persons aged 60 years or older. Consequently, the final sample was not representative at a national level, but every sample was representative for the specific municipality. For the current study, participants aged 1 reflected higher odds of daily walking for transportation in the anticipated direction (e.g. higher odds for more frequent contacts with neighbors). Model parameter estimates were obtained via Markov Chain Monte Carlo (MCMC) procedures applying an orthogonal parameterization (Browne, 2012). First, the separate relationships between the ten perceived social environmental variables and walking for transportation were analyzed and odds ratios with corresponding Bayesian confidence intervals were reported. Those variables that proved to be significant in the first step, were entered simultaneously into model 1: a mixed (multiple) logistic regression analysis. Next, the non-significant perceived social environmental variables were deleted and the demographic (model 2) and perceived physical environmental (model 3) covariates were added, respectively. Additionally, model 3 was optimized by allowing random slopes. Models were compared using the Deviance Information Criterion (DIC) (Browne, 2012). Since MLwiN uses listwise deletion, participants with missing data on any variable in the model were not included in the analysis. This resulted in the inclusion of 24,875 participants in the analyses for model 3. Participants included in the analysis of model 3 were less likely to be female (51.5 vs. 59.5% females), slightly younger (73.8 vs. 74.9 years), better educated (12.8 vs. 7.2% had tertiary education), and walked more (38.4 vs. 32.9% walked for transportation daily) compared to those that were not included. Significance level was determined at 0.05. Results Descriptive statistics Participants had a mean age of 74.3 (6.6) years and 55.6% of them were female. 35.7% of the older adults reported to walk for transportation daily. Table 1 presents the descriptive statistics for the other individual, perceived physical and social environmental characteristics. Relationships between perceived social environmental factors and daily walking for transportation Results of the analysis for the perceived social environmental factors separately and the three models are presented in Table 2. The following two variables were found to be unrelated to daily walking for transportation: only older adults (O.R. ¼ 0.98; 95% C.I. ¼ 0.94e1.03) and too many youngsters living in the neighborhood (O.R. ¼ 1.09; 95% C.I. ¼ 0.99e1.19). Satisfaction with contacts with neighbors was significantly positively related to daily walking for transportation when analyzed separately (O.R. ¼ 1.38; 95% C.I. ¼ 1.30e1.46), but was no longer significant in model 1 (O.R. ¼ 0.96; 95% C.I. ¼ 0.89e1.03). Similarly, a significant positive relationship was found for neighborhood satisfaction (O.R. ¼ 1.30; 95% C.I. ¼ 1.24e1.36), but this was no longer significant in model 2 (O.R. ¼ 1.02; 95% C.I. ¼ 0.95e1.10). Representing the three dimensions of social environment, six factors were significantly related to daily walking for transportation

J. Van Cauwenberg et al. / Social Science & Medicine 104 (2014) 23e30

for transportation daily compared to participants that never have contacts with their neighbors (O.R. ¼ 1.98; 95% C.I. ¼ 1.68e2.33). Participants having contacts weekly or more were also 70% more likely to walk for transportation daily compared to participants with less than weekly contacts (O.R. ¼ 1.70; 95% C.I. ¼ 1.58e1.83). There was no significant difference in walking for transportation between participants that have less than weekly contact and participants that never have contact with their neighbors (O.R. ¼ 1.17; 95% C.I. ¼ 0.99e1.38). Neighbors’ social support was significantly positively related to daily walking for transportation. Participants with neighbors’ social support were 12% more likely to walk for transportation daily compared to those without neighbors’ social support (O.R. ¼ 1.12; 95% C.I. ¼ 1.05e1.18). A significant negative relationship was observed for not too many immigrants living in the neighborhood. Participants who perceived too many immigrants were 63% more likely to walk for transportation daily than those who reported not too many immigrants living in their neighborhood (O.R. ¼ 0.61; 95% C.I. ¼ 0.54e 0.70). Neighborhood involvement was significantly positively related to daily walking for transportation. Participants who felt involved in their neighborhood were 12% more likely to walk for transportation daily (O.R. ¼ 1.12; 95% C.I. ¼ 1.05e1.19). A significant positive relationship was observed for participation. The likelihood of daily walking for transportation increased with 2% for each additional activity in which a participant participated (O.R. ¼ 1.02; 95% C.I. ¼ 1.01e1.04). Volunteering was also significantly positively related to daily walking for transportation. Those who engaged in volunteering were 11% more likely to walk for transportation daily (O.R. ¼ 1.11; 95% C.I. ¼ 1.02e1.20).

Table 1 Descriptive statistics of the participants. Individual variables Age (M  SD)a Gender (% female) Marital status (%) Widowed Never married/ divorced Married/co-habiting Functional limitations (M  SD)a Educational level (%) No/primary education Lower secondary Higher secondary Higher

Physical environmental variables 74.3  6.6 Perceived distance to destinations (%) 55.6 Large 22.3 Medium 23.4 26.1 Low 54.3 6.2 Environmental index 4.8  1.6 a (M  SD) 67.7 Social environmental variables 2.5  2.6 Interpersonal relationships

46.2 27.7 16.1 10.0

Income level (%)

Relationships between the perceived neighborhood social environment and walking for transportation among older adults.

Ecological models state that physical activity (PA) behaviors can be explained by the interplay between individuals and their surrounding physical and...
311KB Sizes 0 Downloads 3 Views