Overview

Abstract (English)

This paper examines spatial pattern of adult obesity in Canada from 2000/01 to 2009/10, and analyzes factors associated with obesity and body mass index (BMI). The analysis is based on a panel data set constructed at the health region (HR) level using five confidential Canadian Community Health Surveys. We found that obesity rates are not distributed randomly across HRs in Canada (Moran’s I, 2000/01: 0.501 and 2009/10: 0.659 [p < 0.001]). Therefore, accounting for the spatial pattern of obesity was necessary in order to obtain reliable parameter estimates from regression analyses. We estimated the random- and fixed-effects specifications for the spatial autoregressive and spatial error models. The Hausman specification tests suggested that the fixed-effects models were preferred to that of the random-effects models. The results showed that the proportions of immigrants and homeowners are inversely related to BMI and obesity, while household income, fruits & vegetables consumption, physical activity and smoking are negatively related to BMI. These findings highlight the importance of socioeconomic and behavioural factors affecting adult obesity in Canada even after spatial heterogeneity was accounted for. Thus, a comprehensive and multi-faceted strategy is required to combat adult obesity in Canada.

Abstract (French)

Please note that abstracts only appear in the language of the publication and might not have a translation.