Comparing food desert residents with non-food desert residents on grocery shopping behaviours, diet and BMI: results from a propensity score analysis.

<bold>Objective: </bold>To determine whether residence in a US Department of Agriculture-designated food desert is associated with perceived access to healthy foods, grocery shopping behaviours, diet and BMI among a national sample of primary food shoppers.<bold>Design: </bold>Data for the present study came from a self-administered cross-sectional survey administered in 2015. Residential addresses of respondents were geocoded to determine whether their census tract of residence was a designated food desert or not. Inverse probability of treatment-weighted regression was used to assess whether residence in a food desert was associated with dependent variables of interest.<bold>Setting: </bold>USA.<bold>Participants: </bold>Of 4942 adult survey respondents, residential addresses of 75·0 % (n 3705) primary food shoppers were included in the analysis.<bold>Results: </bold>Residence in a food desert (11·1 %, n 411) was not significantly associated with perceived access to healthy foods, most grocery shopping behaviours or dietary behaviour, but was significantly associated with primarily shopping at a superstore or supercentre v. a large grocery store (OR = 1·32; 95 % CI 1·02, 1·71; P = 0·03) and higher BMI (b = 1·14; 95 % CI 0·36, 1·93; P = 0·004).<bold>Conclusions: </bold>Results suggest that food desert residents shop at different food stores and have higher BMI than non-food desert residents. [ABSTRACT FROM AUTHOR]