Perceived neighborhood environmental attributes associated with adults’ leisure-time physical activity: Findings from Belgium, Australia and the USA

Abstract

The study purpose was to examine the strength, direction and shape of the associations of environmental perceptions with recreational walking and leisure-time moderate-to-vigorous physical activity, using pooled data from four study sites (Baltimore [USA], Seattle [USA], Adelaide [Australia] and Ghent [Belgium]). Moreover, site- and gender-specificity of the associations were examined. In total, 6014 adults (20–65 years, 55.7% women) completed the Neighborhood Environmental Walkability Scale and the International Physical Activity Questionnaire. Both a ‘recreational walking-friendliness’ index and a ‘leisure-time activity friendliness’ index had a positive linear association with recreational walking and leisure-time moderate-to-vigorous physical activity, respectively. The associations were significant in all study sites except Ghent. Present findings were clearly site-specific, imposing possible challenges for built environment recommendations. In Belgium, interventions to promote leisure-time activity may need to target promotion of existing opportunities rather than built environment improvements.

Introduction

A large proportion of the adult populations of developed countries does not engage in sufficient physical activity to gain health benefits (Hallal et al., 2012). Since physical inactivity is associated with elevated risk of major chronic diseases, interventions to promote healthy and active lifestyles need to be improved and implemented (Garber et al., 2011). In that context, ecological models of health behaviors emphasize the importance of taking into account multiple levels of influence when developing interventions. In addition to individual and social attributes, built environment and policy factors are expected to affect physical activity in different domains (e.g., transport or leisure) and different settings (e.g., neighborhoods, parks) (Sallis et al., 2008).

A growing body of evidence shows that objective and perceived built environment factors are positively associated with physical activity in adults (Brownson et al., 2009, Butler et al., 2011, Heath et al., 2006; Wendel-Vos et al., 2007, Wong et al., 2011). However, built environment correlates are behavior-specific and the factors associated with leisure-time physical activity are less understood than those associated with active transportation. Proximity to recreation facilities and aesthetic-related features have been consistently related to leisure-time physical activity (Cerin et al., 2008a, Kondo et al., 2009, Owen et al., 2004, Van Dyck et al., 2011), but the role of other environmental variables, such as residential density, the availability of walking and cycling infrastructures, crime, and traffic safety is less understood (Inoue et al., 2010, Kondo et al., 2009, Saelens and Handy, 2008; Van Dyck et al., 2011).

Most previous studies of associations of built-environment attributes with physical activity have been conducted in one or perhaps two regions and nearly all are from single countries. Furthermore, with the exception of a few recent African and Latin American studies (Gomes et al., 2011, Oyeyemi et al., 2011, Parra et al., 2011), most previous studies have been conducted in the USA, Australia and Europe. Limited within-region, within-country and between-country variability in built environment attributes and physical activity levels can potentially contribute to an underestimation of strength of associations. Findings from multiple regions and countries are needed to provide a wider range of variability, which will better inform national and regional policies shaping communities and to develop international guidelines for improving built environments to stimulate ‘active living’.

A recent study combined data from 11 environmentally-diverse countries to examine perceived environmental correlates of overall physical activity, using common methods, and found several environmental variables to be linearly and positively related to meeting activity guidelines (Sallis et al., 2009a). Associations were stronger than those reported in single-country studies, probably because data from multiple countries provides a wider range of environmental variability. To gain further insight in the strength, shape and directions of the associations between built environment characteristics and health behaviors, we recently examined the relationship between environmental perceptions and sitting time in four study sites in three diverse countries (USA, Australia and Belgium), using common protocols and pooled analyses (Van Dyck et al., 2012). Site- and gender-specific associations between perceived environmental attributes and overall sitting time were found and the relationship between perceived land-use mix and motorized transport was curvilinear.

Based on these promising findings and the added value to be gained by conducting built environment studies in multiple regions/countries using the same methods, we conducted the present analyses on pooled data from the same three countries. We examined the strength, direction and shape of the associations of environmental perceptions with leisure-time activity (recreational walking and leisure-time moderate-to-vigorous physical activity). In addition, we examined whether these associations differed by study site and gender. We chose to focus specifically on leisure-time physical activity, because the built-environment correlates of this type of physical activity remain unclear (Wendel-Vos et al., 2007).