Built environment, walking and health inequalities in urban Scotland
Background: Many adults do not take recommended amounts of physical activity (PA). This is associated with adverse health outcomes such as obesity, overweight, diabetes and heart disease. Moreover, physical inactivity is socially patterned. People with lower socioeconomic status or who live in more deprived areas do less PA which may in turn contribute to inequalities in health outcomes. Identifying the causes and possible pathways for increasing PA and addressing health inequalities is a pressing national and international priority. There is increasing evidence that features of the built environment (BE) can support physical activities such as walking. The built environment may also ameliorate health inequalities by providing a supportive context for walking across diverse sections of the population. However, there is little evidence relating to the UK and Scottish context or about inequalities in these associations for different groups such as people with different demographic characteristics or people living in areas with different levels of deprivation. This study aimed to fill this knowledge gap, examining associations between built environments and walking in urban Scotland. It considered individual and spatial inequalities in these relationships. Methods: This study had a quantitative cross-sectional design. Geographical Information Systems (GIS) was used to create neighbourhood level BE measures of Area Walking Potential (AWP) across urban Scotland. These were destination accessibility, street connectivity, residential density and walkability (a composite measure of the former three measures). An examination of the distribution of AWP across Scotland and in relation to area deprivation was made. The measures were then appended to individual level walking data for adults aged 19+ years from the 2010 Scottish Health Survey. Regression analysis tested for associations between the AWP measures with four different walking outcomes: any walking, frequency of walking, achieving 30 minutes of walking per day and total minutes walked in the previous week. Individual and area level confounders were controlled for. Associations were examined using two sizes of neighbourhood area: 500m and 1000m zones around residential centres. Interactions with individual demographic, socioeconomic, household characteristics and area deprivation were evaluated. Results: There was modest evidence of positive associations between AWP and walking. After controlling for covariates, destination accessibility showed the strongest associations with frequency of walking. There were limited associations for street connectivity and walkability and no associations between residential density and walking. Positive associations remained for some groups less likely to walk, such as older adults. However, there were also interaction effects showing inequalities in associations between AWP and walking. In particular, people with lower educational attainment were less influenced by AWP. The spatial analysis showed areas with lowest deprivation had lowest AWP although people in more deprived areas walked less overall. Conclusions: There is some evidence that the BE supports some types of walking in Scotland. The BE may also enhance walking opportunities for certain groups who generally walk less, and therefore could potentially reduce inequalities in health outcomes. However, the socioeconomic inequalities in outcomes suggest multifaceted approaches to increasing walking are more likely to reach all sections of the population. The evidence that there are geographic inequalities in levels of AWP can be used to inform geographically targeted interventions aimed at improving walking environments. This research has generated original evidence in the Scottish context, highlighting the importance of context specific research.