Explaining social group disparities in ever tobacco use among women aged 45 years and older in India: a nonlinear decomposition analysis

Abstract

Abstract Background Social group differences in tobacco use among Indian women are well documented; however, most existing studies remain descriptive, reporting associations without explaining disparities in ever tobacco use between Scheduled Caste/Scheduled Tribe (SC/ST) and non-SC/ST women. This study examines the SC/ST–non-SC/ST gap in ever tobacco use among women aged 45 years and older in India by decomposing it into explained and unexplained components. Methods Using Longitudinal Ageing Study in India (LASI, 2017–18) data on women aged ≥ 45 years, we analysed ever tobacco use, defined as ever use of any smoked or smokeless tobacco product. A multivariable logistic regression model was fitted to estimate associations between social group (SC/ST vs. non-SC/ST) and ever tobacco use, adjusting for demographic, socioeconomic, and spatial characteristics. Fairlie decomposition was applied to partition the SC/ST–non-SC/ST gap into explained and unexplained components. Results Ever tobacco use was higher among SC/ST women than among non-SC/ST women (26.4% vs. 16.0%). In adjusted logistic regression models, SC/ST women had significantly higher odds of ever tobacco use compared with non-SC/ST women (adjusted odds ratio [AOR]: 1.30, 95% confidence interval (CI): 1.17–1.43). Fairlie decomposition showed that 75.6% of the SC/ST–non-SC/ST gap in ever tobacco use was explained by differences in observed characteristics. Level of educational attainment (28.8%), occupational history (24.2%), and state/region of residence (40.2%) were the largest contributors to the explained component. Approximately 24.4% of the disparity remained unexplained, indicating that measured socioeconomic and spatial characteristics did not fully account for the SC/ST–non-SC/ST gap. Conclusion The findings show that the SC/ST–non-SC/ST gap in ever use of tobacco is closely linked to social and geographical disadvantage, especially differences in education, work, rural residence, and state context. Tobacco-control efforts should therefore pay greater attention to women aged 45 years and older from socially disadvantaged groups and high-prevalence settings.

The authors' abstract, as published at the source. BMC Public Health, 2026 · DOI ↗