Abstract / Summary
Abstract Background Urticaria is common and women carry a substantial share of chronic disease burden, yet population evidence on menopausal hormone therapy and later urticaria is limited. We estimated the association between baseline hormone therapy history and subsequent recorded urticaria. Methods This prospective cohort study included 266,560 UK Biobank women with known baseline hormone therapy history and no pre-baseline urticaria record. The outcome was first post-baseline recorded International Classification of Diseases, Tenth Revision code L50 from linked health records; the code did not identify chronic spontaneous urticaria. The primary estimand came from a multivariable Cox proportional hazards model. Urticaria and alopecia areata were prespecified co-primary outcomes with Benjamini–Hochberg correction. Sensitivity analyses examined temporality, record source, competing death, menopause, reproductive history, cross-field exposure consistency, propensity-score weighting, and unmeasured confounding. Results During 3.46 million person-years, 2,404 recorded urticaria events occurred. Crude incidence was 0.764 versus 0.654 per 1,000 person-years among women with versus without hormone therapy history. The adjusted hazard ratio was 1.156 (95% confidence interval, 1.051–1.270; P = 0.00270; co-primary q = 0.00541). Estimates were similar after a 2-year lag (1.133; 1.015–1.264), overlap weighting with measured-covariate balance (1.150; 1.045–1.267), restriction to postmenopausal women (1.141; 1.023–1.273), and expanded reproductive adjustment (1.146; 1.043–1.260). Conventional stabilized inverse-probability weighting failed covariate balance and was not treated as causal support. The alopecia areata co-primary estimate was imprecise (1.170; 0.832–1.647). Conclusions Baseline menopausal hormone therapy history was associated with a modestly higher rate of subsequent recorded urticaria. The small absolute rate difference and susceptibility to residual confounding limit immediate clinical implications. The findings warrant replication and do not establish causation or a hormonal mechanism.