Understanding the impact of urban exposure on obesity among middle and old-age migrants in India

Abstract

Rural-to-urban migration is associated with elevated obesity in Western settings. However, whether migration to urban areas ages has any impact on obesity in India is inconclusive and scarce. We, therefore, assessed the impact of migration on obesity among rural-to-urban migrants and compared it with their rural counterparts. This study utilized the first wave of Longitudinal Ageing Study in India. BMI (BMI ≥ 25 kg/m2) and waist circumference (>102 cm and >88 cm for men and women, respectively) were employed to assess overall obesity and abdominal obesity. To fulfil the study objective, this study employed logistic and quantile regression techniques. The study found that individuals migrating from rural to urban areas are significantly more likely to develop obesity than rural stayers. Moreover, within the group of rural-urban migrants, prolonged urban residence was a strong and cumulative predictor for obesity. The risk of obesity was 1.91 times higher (those who lived 5 or fewer years in urban areas), 2.05 times higher (for 6-10 years), and 2.40 times higher (for more than 10 years) compared to their rural counterparts. This study identified migration and prolonged urban exposure as crucial risk factors for the development of obesity among middle-aged and older adults in India.

Fig 1. Distribution of BMI score stratified by migration status among the middle aged and older adults (aged 45+) in India, LASI wave 1 (2017−18).

Fig 2. Sex stratified weighted prevalence of obesity by residence status (non-migrants) and duration of residence in urban areas (among migrants) aged 45+ in India, LASI Wave 1 (2017−18).

Fig 3. Multivariable quantile regression predicting obesity at the 0.10–0.90 quantile.

Note: Obesity estimates from quantile regression are represented by the solid red line and their 95% confidence intervals are shown by the shaded red region. The solid black line is the effect size estimate from conventional linear regression analysis with its 95% confidence intervals shown by the black dashed lines.