Background Obesity is a worldwide public health issue. Many factors contribute to this: a holistic approach is required to tackle it. Due to the rapid increase in obesity in many parts of the world, the food environment and its effect on obesity has been an increasing topic of study. Mexico has one of the highest prevalence rates of obesity in the world: 70% of the Mexican population is overweight or obese. Furthermore, the country has gone through a dietary and food retail transition. The aim of this ongoing study is to explore the association between the density of different food outlets and obesity in Mexican adults.

Methods Geographical and food outlet data were obtained from the National Institute of Statistics and Geography in Mexico. For health data I used anthropometric measurements and socio-economic characteristics of adult participants (18+) from the National Health and Nutrition Survey in Mexico 2012. Densities of supermarkets, restaurants chain and non-chain convenience stores and fruit and vegetables stores were calculated per census tract area (CTA). Count of each food outlet type and the population in each CTA were used to calculate densities in ArcGIS. Regressions were undertaken to explore the association of BMI and the density of each food outlet type using complex survey design in STATA14. I classified CTAs as obesogenic or leptogenic according to the density and type of food outlets that predominated. All analyses were adjusted for sex, age, physical activity, socioeconomic status, education and deprivation.

Results No significant associations were found between the density of supermarkets, fruit and vegetable stores, chain and non-chain convenience stores and obesity (BMI ≥ 30 kg/m²). However, the density of restaurants, including fast-food outlets, was significantly and negatively associated with obesity among women (β = −0.54 [95% CI −0.99, 0.08] P = 0.019). In regard to the food environment classification, CTAs that had a higher density of food outlets were also significantly and negatively associated with obesity (β = −1.43 [−2.18 to −0.68] P<0.01).

Conclusion Food environments with a greater healthy food availability and diversity could be associated with lower BMIs. On the other hand, high density of food outlets which offer calorie-dense food, rich in saturated fats and high glycaemic foods and drinks, along with large portions could be a contributing factor to the problem of obesity in Mexico. Further geospatial analyses are ongoing.