Association of dietary patterns and health outcomes by spatial regression analysis of nationally representative survey data from India.

Background: Our aim was to study the regional differences in dietary patterns in India and their association with population-level nutrition-related health indicators such as the prevalence of anemia, overweight, undernutrition, and hyperglycemia.

Objectives: To identify and characterize the dietary patterns from publicly available nationally representative survey data on food consumption conducted by the National Sample Survey Office (NSSO) to study the regional differences in dietary patterns.

Methods: Dietary patterns were identified by factor analysis of per capita food consumption data from the household consumer expenditure survey (2011). Mean factor scores of dietary patterns were calculated for each district separately for urban and rural regions. Ecological association of factor scores with the district-level percentage prevalence of health indicators from the National Family Health Survey-4 (2015-2016) data was done by the Spatial Durbin Model of spatial regression analysis.

Results: Factor analyses revealed four dietary patterns which were similar in terms of the food items that characterized the factors for both rural and urban regions. Direct effects of dietary patterns by spatial regression analyses were observed with several health outcomes after adjusting for differences in socioeconomic development. Prevalence of anemia was positively associated with "Milk and wheat-rich diet" among men in the rural regions but negatively associated with other dietary patterns. Prevalence of overweight and high blood glucose was positively associated with "Rice and meat-rich diet" and "Coconut and seafood rich diet" in the rural regions. "Refined oil and tur dal-rich diet" was positively associated with the prevalence of overweight and hypertension in urban regions and negatively associated with underweight and anemia in men in rural and urban regions.

Conclusions: Spatial regression analyses revealed several important associations between dietary patterns and health outcomes, mostly in rural regions and some in urban regions. These results suggest the role of the major food items consumed in different regions and their impact on health outcomes in India and may have implications in tailoring dietary modifications accordingly.