Understanding geographical variability of women's alcohol prevalence in India using spatial regression models: Evidence from national representative survey

Alcohol consumption among women in India is a complicated phenomenon driven by several determining variables and demonstrating clear regional patterns throughout the country. The study examines district-level disparities in women's alcohol prevalence (WAP) in India using recent NFHS-5 data from 2021. This study used exploratory spatial data analysis (ESDA) techniques like local Moran's I statistics and bivariate local indicators of spatial association (BiLISA) to illustrate spatial patterns of the specified variables and pinpoint probable lagging regions. Besides, global models such as ordinary least squares (OLS), spatial lag models (SLM), and spatial error models (SEM) were used to grasp an overview of explanatory factors and alcohol prevalence at global levels. Further, geographically weighted regression (GWR) was employed locally to capture the determinants' spatial fluctuations. The results indicate significant clustering patterns among the predictors exhibiting the existence of spatial autocorrelation. The analysis also reveals that marginalized populations contribute to higher alcohol consumption among women, as evidenced by the clustering patterns. The study highlights the spatial disparities and factors influencing WAP in India, providing insights for targeted interventions and policy-making at the local level.