Environmental Correlates of Physical Activity

PURPOSE: Understanding environmental correlates of physical activity can inform policy changes. Each country has a limited range of environmental variation, so associations are likely to be underestimated. As part of the International Physical Activity Prevalence Study, 10 countries used the same set of 7 self-report environmental variables and the International Physical Activity Questionnaire (short version), allowing correlational analyses with pooled data representing the wide variation seen around the world. METHODS: The countries were Belgium, Brazil, Colombia, China (Hong Kong), Japan, Lithuania, New Zealand, Norway, Sweden, and USA, with a total sample exceeding 26,000 adults. Samples were roughly representative, and seasons of data collection were comparable. Physical activity was dichotomized based on meeting accepted guidelines for walking, moderate and/or vigorous activity. Environmental attributes were rated on a four-point Likert scale and were scored “agree” vs. “disagree”. Logistic regression analyses were conducted to evaluate the associations between level of physical activity and environmental attributes, adjusted for age, sex, and country. RESULTS: Three of 7 environmental variables were significantly related to meeting physical activity guidelines. Odds ratios were 1.16 (95% CI = 1.07, 1.25) for shops within easy walking distance, 1.14(95% CI = 1.06, 1.23) for bicycle facilities in neighborhood, and 1.17(95% CI = 1.09, 1.26) for free or low cost recreational facilities in the neighborhood. Nonsignificant associations were observed between physical activity level and residential density, presence of sidewalks, presence of transit stop near home, and perceived crime. CONCLUSION: Total self-reported physical activity was related to environmental features expected to facilitate active transportation and active recreation in a pooled sample from 10 countries around the world. RESULTS: provide a rationale for improving bicycling and recreational facilities and for promoting policies that favor mixed use development patterns. Environmental correlates are shown to have global applicability, so more detailed studies of the built environment are justified.