In this large representative sample of adults we observed a substantial risk factor burden indicating that almost one quarter of the population are highly susceptible to cardiometabolic morbidity.
Our findings showed strong and consistent negative associations between the objectively measured walkability of the immediate (800 m) and extended (1,600 m) residential neighborhood and the prevalence of obesity. For type-2 diabetes mellitus, strong negative associations were also observed at the 800 m neighborhood buffer and this was only significant in men. Furthermore, no consistent associations between walkability and hypertension and hypercholesterolaemia were observed.
A number of studies have investigated the associations between characteristics of the built environment and obesity . However, the majority of this research has been conducted in the United States and overall there have been mixed findings [20, 30, 35, 36]. In contrast to our findings, a recent study from the same region that also used a walkability index to measure the built environment (Western Australia) found no association between walkability and body mass index . This might be explained by differences in the study population (the current study consists of a large population representative sample vs. the previous study consisted of home owners who moved into new housing developments). Moreover, differences in the outcome classification might also explain this apparently contradictory finding.
Overall, associations between neighborhood walkability and CMRF appeared to be stronger and more consistent in the immediate (800 m) compared with the extended (1,600 m) neighborhood, as well as in men compared with women. The observed sex difference may, in part, be due to a more walkable or more pedestrian friendly environment having a larger influence on the physical activity behavior and in turn weight status and risk of type-2 diabetes mellitus in men than women. Mediation analyses provided some confirmation of this as the association between walkability and obesity was partly explained by physical activity and sedentary behavior. However, a considerable amount of the association between walkability and obesity was not explained by physical activity and sedentary behavior in our study. One possible explanation for this observation is, that a more context-specific measure of physical activity, such as neighborhood level physical activity (vs. a general measure of total physical activity as used in our study) may more adequately and, to a greater extent, explain the association between neighborhood walkability and obesity . Similarly, a measure of sitting time at home as compared to total sitting time, which is largely affected by the work environment, for the measurement of sedentary behavior may explain observed associations. Previous research that reported the proportion of physical activity within and outside the residential neighborhood offers some support to this hypothesis . Also, objectively and thus more accurately measured physical activity and sedentary behavior may be able to explain observed associations more adequately. However, this study was restricted to using a self-report measure of total physical activity and sedentary behavior as captured by the HWSS survey. The fact that neighborhood walkability was less strongly associated with obesity and type-2 diabetes mellitus in women raises the question as to whether the influence of specific built environment factors differs by sex. For instance, other factors of the built environment such as the availability and accessibility of neighborhood destinations (e.g., food outlets, parks and recreation venues) and perceived and real safety may have a greater influence on obesity and other CMRF in women than in men [39, 40]. Future research should explore the relationship between other built environment factors and CMRF separately for men and women and consider using neighborhood level measures of physical activity. Moreover, the more proximate neighborhood appears to be more important for cardiometabolic outcomes and this should be further investigated in future studies.
Our findings with regard to type-2 diabetes mellitus, hypertension, and hypercholesterolaemia are only partly consistent with previous research. To date, the majority of studies have investigated the association between environmental factors such as noise exposure and hypertension or dyslipidaemia . Among the small number of published studies investigating the influence of the built environment only one used a walkability index [21–23, 41]. Generally, however, studies found that a supportive built environment (e.g. high walkability, high population density) was positively associated with blood pressure levels or hypertension prevalence [21–23, 41]. In contrast, we found no significant relationship between walkability and hypertension. A recent systematic review identified only two studies investigating the association between the built environment and type-2 diabetes mellitus but none of these studies measured walkability. Studies found that housing conditions and more resourceful neighborhoods (suitability of the environment for physical activity and availability of healthy foods) was associated with a lower incidence of type-2 diabetes mellitus [24, 25]. In support of prior findings, our study found that walkability was negatively associated with type-2 diabetes mellitus in men only and at the 800m neighborhood buffer. This finding is plausible and somewhat compatible with our findings that particularly in men the immediate neighborhood (800 m buffer) was more strongly and more consistently associated with obesity. Body weight and obesity, respectively, are among the strongest risk factors for type-2 diabetes mellitus [42–44]. Hypothetically, the stronger association between the walkability and obesity in men and in the immediate neighborhood could explain the observed association between walkability and type-2 diabetes mellitus. These findings warrant further investigation and confirmation in future studies. Finally, our study appears to be the first to investigate the association between the built environment and hypercholesterolaemia , however no significant association was observed.
A number of factors may explain the variation between our findings and that of previous research. For instance, our study had a representative sample of the adult population aged 25 and older. Many previous studies used a different methodology and were not based on representative adult population samples but specifically targeted certain population groups, such as middle aged people [22, 23] or populations with specific ethnic origins . Furthermore, as previously indicated, we used walkability as a composite measure of the built environment whereas many other studies have focused on single environmental characteristics (e.g. neighborhood conditions, population density) [22, 24]. Another potential factor could be the choice of neighborhood buffers used. We investigated the immediate and more extended neighborhood (800 m and 1,600 m neighborhood buffer) with associations being somewhat stronger at the 800 m neighborhood buffer. However, in contrast to our findings other studies have also reported associations between the built environment and type-2 diabetes mellitus and hypertension at a neighborhood buffer of 1,600 m [25, 41].
A major strength of the current study is that it investigated multiple CMRF within the same study population. In addition, our study was based on a representative population sample with a high proportion of participants granting permission for data linkage. Furthermore, walkability was measured using objective attributes of the built environment. Nevertheless, when interpreting our findings a number of limitations should be considered. First, despite their frequent use, limitations of composite measures of the built environment (walkability index) have been highlighted . Among others, these include the inability to identify specific relevant environmental components, the use of entropy scores that can be similar for different walking environments, and the fact that a wide range of land uses is usually not captured in existing walkability indices. On the other hand, using a walkability index could have the advantage of providing better estimates compared to individual and frequently inter-related environmental components . In addition, focusing on a number of relevant environmental variables is likely necessary and could have greater potential to improve pedestrian friendliness of a neighborhood , Second, our study used self-reported outcome measures which could introduce measurement error. Future studies investigating multiple CMRF should include objective measures of CMRF’s. Third, although we used a composite measure of the built environment other potentially important built environment or neighborhood attributes (e.g., presence of food outlet destinations) were not considered and may differ by sex (e.g., neighborhood safety and crime). Fourth, this study had a cross-sectional design, which limits the ability to draw causal inferences. It should also be acknowledged that we were not able to identify factors related to the choice of residential neighborhood because the neighborhood and health data were linked by an independent government department. Adjustment for residential preference was therefore not possible within our study. However, we have longitudinal evidence that the influence of neighborhood selection might not be large . Nevertheless, future studies should attempt to consider this aspect when investigating associations between the neighborhood environment and health related outcomes.
There is growing recognition that the global burden of preventable chronic disease is so great, that if uncurbed it could cripple health systems and undermine social and economic development . Combating chronic diseases is therefore said to require a whole-of-society multi-sector approach, including consideration being given to city planning that support more active lifestyles such as daily walking and cycling . While the associations observed in this study are relatively modest, small shifts across whole populations can have a large impact on the burden of disease. Thus, these results highlight the importance of engaging city planners to create walkable pedestrian-friendly neighborhoods as one key strategy in tackling key CMRF.