Characterising food environment exposure at home, at work, and along commuting journeys using data on adults in the UK

Thomas Burgoine,Pablo Monsivais

doi:10.1186/1479-5868-10-85

Abstract

BackgroundSocio-ecological models of behaviour suggest that dietary behaviours are potentially shaped by exposure to the food environment (‘foodscape’). Research on associations between the foodscape and diet and health has largely focussed on foodscapes around the home, despite recognition that non-home environments are likely to be important in a more complete assessment of foodscape exposure. This paper characterises and describes foodscape exposure of different types, at home, at work, and along commuting routes for a sample of working adults in Cambridgeshire, UK.MethodsHome and work locations, and transport habits for 2,696 adults aged 29–60 were drawn from the Fenland Study, UK. Food outlet locations were obtained from local councils and classified by type - we focus on convenience stores, restaurants, supermarkets and takeaway food outlets. Density of and proximity to food outlets was characterised at home and work. Commuting routes were modelled based on the shortest street network distance between home and work, with exposure (counts of food outlets) that accounted for travel mode and frequency. We describe these three domains of food environment exposure using descriptive and inferential statistics.ResultsFor all types of food outlet, we found very different foodscapes around homes and workplaces (with overall outlet exposure at work 125% higher), as well as a potentially substantial exposure contribution from commuting routes. On average, work and commuting environments each contributed to foodscape exposure at least equally to residential neighbourhoods, which only accounted for roughly 30% of total exposure. Furthermore, for participants with highest overall exposure to takeaway food outlets, workplaces accounted for most of the exposure. Levels of relative exposure between home, work and commuting environments were poorly correlated.ConclusionsRelying solely on residential neighbourhood characterisation greatly underestimated total foodscape exposure in this sample, with levels of home exposure unrelated to levels of away from home exposure. Such mis-estimation is likely to be expressed in analyses as attenuated parameter estimates, suggesting a minimal ‘environmental’ contribution to outcomes of interest. Future work should aim to assess exposure more completely through characterising environments beyond the residential neighbourhood, where behaviours related to food consumption are likely to occur.

Highlights

Socio-ecological models of behaviour suggest that dietary behaviours are potentially shaped by exposure to the food environment (‘foodscape’)
One study found that 49% of participants in Seattle had greater supermarket exposure when outside of the home neighbourhood [17], whilst another observed that fast food outlet exposure around workplaces was more than two-fold higher than around residences [18]; similar results have been found elsewhere [19,20]
Our aims were: 1) to detail the data sources and methods used to derive estimates of foodscape exposure, which extend beyond the residential address, and include travel mode and frequency; 2) to describe and test for significant differences in the distributions of food outlets by type in each domain, and to examine domain specific contributions to total foodscape exposure, and; 3) to assess whether relative levels of food environment exposure across the three domains were similar on a per-person basis

Summary

Introduction

Socio-ecological models of behaviour suggest that dietary behaviours are potentially shaped by exposure to the food environment (‘foodscape’). Research on associations between the foodscape and diet and health has largely focussed on foodscapes around the home, despite recognition that non-home environments are likely to be important in a more complete assessment of foodscape exposure. The significance of non-home environments to diet and health has been recognised [11,12,13], with the assessment of these environments likely to be especially important for understanding dietary behaviours in working age adults [14,15]. Considering supermarkets as ‘enablers’ of a healthy diet [21,22], and the relative unhealthiness of fast foods consumed away from the home [23], these non-home exposures may be important for food intake and health. Associations between relatively low home food outlet exposures and diet may be substantially confounded by relatively high exposures in nonhome domains

Objectives

Methods

Results

Discussion

Conclusion