Bayesian Analysis of Postoperative Complication Risk Associated With Preoperative Exposure to Fine Particulate Matter: A Single‐Center Cohort Study

BACKGROUND: ) exposure and postoperative complications, highlighting its interpretability and flexibility for clinical environmental epidemiology. METHODS: concentrations in the 7 days prior to surgery. The binary outcome was a composite of postoperative complications: pneumonia, surgical site infection, urinary tract infection, sepsis, stroke, myocardial infarction, or thromboembolic event. A hierarchical Bayesians regression model with weakly informative priors was used adjusting for age, sex, season, neighborhood disadvantage, and the Elixhauser index of comorbidities with census tract as a group (random) effect. We present posterior estimates with credible intervals, highlight model transparency and sensitivity, and discuss contrasts with standard frequentist methods. RESULTS: , the odds of complication rose to over 27% (95% CI: 4%-55%). The results were robust across prior choices and model specifications. We report full posterior distributions and highlight advantages of Bayesian modeling for uncertainty quantification and clinical interpretability. CONCLUSIONS: exposure and postoperative complications, highlighting transparent risk estimation and uncertainty characterization that may inform the design of future multicenter perioperative environmental studies. EDITORIAL COMMENT: Using Bayesian statistical analysis, the authors demonstrate a dose-dependent risk for postoperative complications in patients exposed to air polluted with fine particulate matter with a size of less than 2.5 μm.