Risk stratification for overactive bladder (OAB) in type 2 diabetes mellitus (T2DM) remains limited by reliance on symptom-based diagnosis. A retrospective cohort study, published in European Journal of Medical Research, developed and temporally validated a clinical prediction model to stratify OAB risk using routinely available clinical parameters.

The study included 1531 patients with T2DM. The model was derived using multivariable logistic regression with backward selection in a development cohort of 810 patients enrolled between January 2018 and December 2021, and validated in an independent cohort of 721 patients enrolled between January 2022 and February 2025. OAB was defined using the Overactive Bladder Symptom Score (OABSS). Model performance was evaluated using discrimination, calibration, and decision curve analysis (DCA). SHapley Additive exPlanations (SHAP) analysis was used to assess variable contributions.

Seven variables were retained in the final model: diabetic peripheral neuropathy (DPN), glycated hemoglobin (HbA1c), urinary albumin-to-creatinine ratio (UACR) category, neutrophil-to-lymphocyte ratio (NLR), age, triglyceride-glucose (TyG) index, and post-void residual volume (PVR). The model demonstrated strong discrimination, with an area under the curve of 0.880 (95% CI 0.844-0.917) in the development cohort and 0.858 (95% CI 0.814-0.902) in the validation cohort. Calibration was reported as good.

Decision curve analysis supported clinical utility across a wide range of risk thresholds. SHAP analysis identified DPN as the most influential predictor in the model. These findings indicate that a multivariable clinical model incorporating key diabetes-related parameters can stratify OAB risk in patients with T2DM and support risk-based clinical assessment.