Abstract
BACKGROUND: Ursodeoxycholic acid (UDCA), used for treating cholestasis, was reported to protect against severe COVID-19 outcomes. We aimed to assess this potential effect using nationwide Swedish register data. METHOD: We included all SARS-CoV-2 test-positive subjects aged ≥ 18 years during 2020–2023, with previous diagnosis of potentially UDCA-treated diseases, including primary biliary cholangitis, liver cirrhosis, autoimmune hepatitis, or cholangitis. Subjects who had filled a UDCA prescription within 6 months before testing positive were considered exposed. Subjects were followed from test-positivity to the earliest of analyzed outcome (COVID-19-related hospitalization, 30-day all-cause mortality, or COVID-19-related death), one year of follow-up, emigration, death, or end of 2023. Control for potential confounding (vaccination status, COVID-19 waves, prior comorbidities, and sociodemographics) was done by adjustment, propensity score weighting (PSW), and doubly robust analysis. Cox regression was used to estimate hazard ratios (HR) with 95% confidence intervals (95%CI). RESULTS: The study cohort consisted of 833 UDCA-exposed and 3638 non-exposed individuals. The adjusted Cox analysis showed potentially decreased risks (non-significant) among UDCA-exposed for 30-day mortality (HR 0.62, 95%CI 0.34–1.13) and COVID-19 death (0.70, 0.37–1.32), but a smaller non-significant increased risk for hospitalization (1.19, 0.92–1.53). PSW achieved an adequate balance between exposed and non-exposed and showed similar results for 30-day mortality (0.78, 0.40–1.53), COVID-19 death (0.86, 0.43–1.71), and hospitalization (1.15, 0.81–1.61). The doubly robust method showed similar results as PSW. CONCLUSION: This study did not provide evidence supporting protective effects for severe COVID-19 outcomes. However, the results are limited by the small cohort of patients treated with UDCA. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12889-026-26908-1.