Genmod Work ((top))

, and information criteria (AIC, BIC) to evaluate how well the model describes the data. Workflow for Generating Reports

Using Poisson regression with a log link (PROC GENMOD, SAS), we modeled 30-day readmission counts among 1,200 patients, offset by log(length of stay). Predictors included age, Charlson score, and discharge disposition. The model showed good fit (deviance/df = 1.02). Older age (IRR = 1.03 per year; 95% CI: 1.01–1.05) and higher Charlson score (IRR = 1.21 per point; 1.12–1.31) significantly increased readmission rates. Discharge to home health was protective (IRR = 0.82; 0.71–0.95). No overdispersion detected. Results suggest targeting high‑comorbidity older patients for transitional care. genmod work