Application of single-level and multi-level modeling approach to examine geographic and socioeconomic variation in underweight, overweight and obesity in Nepal: findings from NDHS 2016.

Nepal's dual burden of undernutrition and over nutrition warrants further exploration of the population level differences in nutritional status. The study aimed to explore, for the first time in Nepal, potential geographic and socioeconomic variation in underweight and overweight and/or obesity prevalence in the country, adjusted for cluster and sample weight. Data came from 14,937 participants, including 6,172 men and 8,765 women, 15 years or older who participated in the 2016 Nepal Demography and Health Survey (NDHS). Single-level and multilevel multi-nominal logistic regression models and Lorenz curves were used to explore the inequalities in weight status. Urban residents had higher odds of being overweight and/or obese (OR: 1.89, 95% CI: 1.62-2.20) and lower odds of being underweight (OR: 0.81, 95% CI: 0.70-0.93) than rural residents. Participants from Provinces 2, and 7 were less likely to be overweight/obese and more likely to be underweight (referent: province-1). Participants from higher wealth quintile households were associated with higher odds of being overweight and/or obese (P-trend < 0.001) and lower odds of being underweight (P-trend < 0.001). Urban females at the highest wealth quintile were more vulnerable to overweight and/or obesity as 49% of them were overweight and/or obese and nearly 39% at the lowest wealth quintile were underweight.