Frontiers in Endocrinology · Sep 30, 2026

Socioeconomic disadvantage, air pollution, and metabolic health among women: analyses of UK Biobank and US data

By Ma Y, Xie M, Liang M et al.

BACKGROUND: Socioeconomic disadvantage and ambient air pollution are each associated with adverse liver and metabolic health. We examined their joint associations with liver-related and metabolic outcomes among middle-aged and older women. METHODS: This multi-database observational study included UK Biobank women and US women aged 40 years or older. UK exposures were Townsend deprivation and a residential 2010 PM2.5-NO₂ score; outcomes were calculated Fatty Liver Index (FLI), FLI-defined probable fatty liver, and MRI liver fat. The contextual UK main model jointly included air pollution, Townsend deprivation, age, and collapsed ethnicity; extended sensitivity models additionally included smoking, alcohol frequency, education, and walking pace. US exposures were state poverty and PM2.5, and the primary outcome was age-standardized obesity prevalence. RESULTS: Among 231,525 UK women, mean FLI was 38.19 and 25.79% had FLI ≥60. In the contextual UK main model, Townsend deprivation was associated with higher FLI [β = 1.114, 95% confidence interval (CI) 1.069 to 1.159] and higher odds of FLI ≥60 [odds ratio (OR) 1.080, 95% CI 1.076 to 1.084]. The positive age-adjusted pollution associations were attenuated after Townsend adjustment (FLI β = -0.068, 95% CI -0.137 to 0.001; FLI ≥60 OR 0.999, 95% CI 0.993 to 1.004). Across 46 US jurisdictions, each one standard deviation higher poverty level was associated with 1.38 percentage points higher female obesity prevalence after PM2.5 and regional adjustment (95% CI 0.58 to 2.17). CONCLUSION: Socioeconomic disadvantage was consistently associated with poorer liver-related and metabolic health among middle-aged and older women. In contrast, the UK pollution-FLI association was not supported after concurrent adjustment for area deprivation. Social conditions and environmental exposures should be considered jointly when assessing metabolic health in women.

doi.org/10.3389/fendo.2026.1948066Open accessPubMed 42876562Populations