Abstract
BACKGROUND: Benign prostatic hyperplasia (BPH) is highly prevalent among aging men and may lead to progressive lower urinary tract symptoms, surgical intervention, and serious outcomes such as acute urinary retention and renal impairment. In routine clinical practice, the long-term spectrum, timing, and determinants of multiple key outcome events after BPH diagnosis remain insufficiently characterized. This evidence gap is particularly relevant in China, where care pathways, follow-up patterns, and comorbidity profiles may differ from those reported in other populations. A comprehensive real-world description of long-term outcome patterns and determinants is needed to inform risk-stratified follow-up intensity and intervention timing, where current evidence remains limited. OBJECTIVE: This study aims to characterize BPH progression trajectories and associated risk factors, quantify long-term risks and time-to-event profiles of major clinical outcomes, and develop and validate prognostic models using multicenter retrospective data from the People's Liberation Army General Hospital consortium. METHODS: We will establish a multicenter retrospective cohort using routinely collected clinical data from the People's Liberation Army General Hospital consortium (the First, Third, Fourth, Fifth, Sixth, Seventh, and Eighth Medical Centers) spanning 2001 to 2021. The index date is defined as the first recorded BPH diagnosis, and we will include patients with confirmed BPH for whom long-term follow-up data are available. To approximate incident (first-diagnosis) BPH, we applied a 2-year washout period before the index date and cleaned the cohort during assembly by excluding patients with any BPH-related records within this window. Prespecified outcomes include BPH-related surgery, urinary retention, inguinal hernia, chronic kidney disease, urothelial carcinoma, and all-cause mortality. Cox proportional hazard regression will be applied for all-cause mortality, and Fine-Gray competing risk models will be used for nonmortality end points where death may act as a competing event. Model development and validation will be performed across centers, with performance assessed using discrimination, calibration, and clinical utility metrics. RESULTS: By mid-January 2026, data collection for the multicenter retrospective dataset (2001-2021) had been completed, and data cleaning and harmonization had commenced. As this is a retrospective electronic health record-based study, no participant recruitment is involved; the final analytic sample size will be confirmed after harmonization and cleaning. The baseline characteristic table for cohort 1 (First Medical Center) has been completed as part of data quality control. Completion of data harmonization and the prespecified analyses (including model development and validation) is anticipated by May 2026, followed by dissemination and translation to support routine clinical decision-making. CONCLUSIONS: This protocol describes a large, multicenter, real-world cohort study designed to provide robust long-term evidence on BPH progression and clinically important outcomes in China and establish interpretable and validated prognostic models to inform risk-stratified follow-up and intervention decisions.