JMIR Publications

ABSTRACT

Background:

Introduction Benign prostatic hyperplasia (BPH) is one of the most common urological disorders among elderly men, with prevalence increasing markedly with age. Epidemiological data indicate that over 50% of men older than 50 years and up to 80% of those over 80 years are affected (1-3). Clinically, BPH is primarily characterized by lower urinary tract symptoms (LUTS) involving storage, voiding, and post-micturition phases. Imaging typically reveals prostatic enlargement, and pathology is marked by hyperplasia of stromal and glandular tissues. As the disease progresses, bladder outlet obstruction can worsen, resulting in bladder dysfunction and even upper urinary tract impairment (4-6). Importantly, BPH progression is often insidious and irreversible: some patients experience rapid worsening of LUTS, while others develop long-term detrusor dysfunction due to chronic obstruction, leading to complications such as overactive bladder, urinary retention, and recurrent urinary tract infections (4, 5). These complications not only diminish quality of life but also impose a substantial individual and societal disease burden, underscoring the need to better understand the natural course and clinical outcomes of BPH (1, 7). The clinical outcomes of BPH progression are frequently severe, and management remains challenging. A proportion of patients with moderate-to-severe disease ultimately require surgical intervention, most commonly transurethral resection of the prostate (TURP) or various laser procedures (8, 9). However, the presence of comorbidities can complicate perioperative risk, prolong recovery, and make optimal timing of surgery difficult to determine. Moreover, although prior studies have examined predictors of BPH progression and treatment needs, reliable individualized prediction tools remain lacking. In clinical practice, baseline characteristics alone are insufficient to accurately anticipate symptom deterioration or surgical necessity, leading either to delayed treatment with irreversible harm or to overtreatment with unnecessary risk(10). Existing research often focuses on single predictors or short-term outcomes, with limited consideration of multifactorial interactions and long-term disease trajectories, particularly in relation to comorbid conditions (9). Overall, current evidence on BPH remains insufficient in systematically assessing disease progression, risk factors, and long-term clinical outcomes. Large-scale, long-term cohort studies encompassing the full continuum of “progression–outcome–prediction” are still scarce (11). To address this gap, the present study will leverage nearly two decades of multicenter follow-up data from the PLA General Hospital medical consortium to establish a large retrospective cohort. The study will: (1) characterize the trajectories of disease progression following BPH diagnosis and identify associated risk factors; (2) investigate six key outcome events—surgical intervention, urinary retention, inguinal hernia, chronic kidney disease, urothelial carcinoma(12-16), and death—including their incidence, cumulative rates, and median time to occurrence; and (3) develop risk prediction models for BPH-related outcomes. Findings from this research are expected to provide robust evidence to support prognostic evaluation and individualized risk stratification in the clinical management of BPH.

Objective:

Objectives This study aims to analyze disease trajectories, risk factors, and key outcomes of BPH, and to develop predictive models using multicenter retrospective data from the PLA General Hospital network.

Methods:

Methods and Analysis This study will utilize multicenter data from the PLA General Hospital consortium (the First, Third, Fourth, Fifth, Sixth, Seventh, and Eighth Medical Centers) (2000–2021) in 15 October 2025, including patients with a confirmed diagnosis of BPH and follow-up ≥180 days. The date of first diagnosis will serve as baseline. Patients with prostate cancer, urethral stricture, neurogenic bladder, or missing key clinical data will be excluded. Outcomes include: (1) BPH-related surgery (TURP, bipolar or laser procedures); (2) urinary retention; (3) inguinal hernia; (4) chronic kidney disease (CKD, eGFR <60 mL/min/1.73m²); (5) urothelial carcinoma; and (6) all-cause mortality. Data management: Standardized case report forms (CRFs) will be extracted and securely transferred with encryption. Logical checks and random audits will ensure accuracy. Missing data will be primarily handled by multiple imputation (MI), with sensitivity analyses conducted using last observation carried forward (LOCF). Sample size estimation: Based on Cox regression and Fine–Gray competing risk models, at least 300–490 events are required to analyze ~25 covariates. Assuming a 10% incidence of surgery at 5 years, ~3,334 patients will be needed. For rarer outcomes such as CKD (3% incidence), ~10,000 patients are required. Final sample size will be refined according to empirical event rates. Statistical analysis: Baseline characteristics will be summarized as mean ± standard deviation (SD), median with interquartile range (IQR), or counts with percentages. Kaplan–Meier (KM) curves will estimate cumulative incidence and median time to outcomes. Cox proportional hazards models and Fine–Gray models will evaluate risk factors, with results reported as hazard ratios (HRs) and 95% confidence intervals (CIs). Prediction models (e.g., nomograms) will be constructed from multivariable results. Internal validation will use bootstrapping, and external validation will be performed by cross-validation between the two cohorts. Model performance will be assessed using concordance index (C-index), receiver operating characteristic (ROC) curves, area under the curve (AUC), calibration plots, decision curves,and reclassification metrics (net reclassification improvement [NRI], integrated discrimination improvement [IDI]). Statistical significance will be defined as two-sided p<0.05.

Results:

Discussion This study aims to establish a large-scale, long-term, multicenter retrospective cohort to systematically elucidate the full disease continuum of BPH—from progression to outcomes to prediction—and to address the current knowledge gaps regarding the natural history and prognostic determinants of BPH in the Chinese population. Limitations of existing research Research on the clinical course and outcomes of BPH has several limitations. First, although existing evidence suggests that frailty phenotype and larger prostate volume are established risk factors for disease progression (17, 18), robust quantitative data on mid- to long-term risks of key outcomes, such as surgical intervention and acute urinary retention, remain limited. Second, most studies have relatively short follow-up durations, lacking long-term risk estimates over ten years, particularly for elderly and frail populations (19-21). Third, many prior studies suffer from small sample sizes or single-center designs, limiting statistical power for infrequent outcomes such as chronic kidney disease (11, 22). More importantly, inadequate adjustment for competing risks in some analyses may lead to overestimation of disease-specific outcomes (23, 24). These limitations highlight the urgent need for large-sample, long-term, multicenter studies employing robust statistical frameworks. Strengths of the present study Compared with previous studies, the present work offers several methodological advantages: Extended follow-up duration – Leveraging nearly two decades of clinical data, this study can delineate long-term trajectories of BPH outcomes, overcoming the constraints of short- to mid-term studies. Multicenter, large-scale cohort – Inclusion of patients across various disease stages and treatment pathways enhances representativeness and external validity, while sample size estimation ensures sufficient power to detect low-incidence outcomes. Comprehensive outcome assessment – Beyond conventional endpoints such as surgery and urinary retention, the study incorporates inguinal hernia, chronic kidney disease, urothelial carcinoma, and all-cause mortality, providing a more holistic evaluation of disease burden. Rigorous statistical methodology – Depending on outcome characteristics, both Cox proportional hazards and Fine–Gray competing risk models will be employed, with multivariable adjustments to minimize confounding and maximize robustness. Predictive modeling framework – Building on identified risk factors, the study will construct visual predictive tools (e.g., nomograms), with internal validation and cross-validation across cohorts, and quantify predictive performance using C-statistics. This addresses the lack of reliable prognostic tools for BPH outcomes. Expected contributions This study is expected to generate several important insights. First, it may identify novel risk factors beyond age and prostate volume, including metabolic parameters, comorbidity profiles, and initial treatment strategies, thereby improving early risk stratification. Second, it will provide authoritative, large-scale estimates of cumulative incidence and median time-to-event for major outcomes in the Chinese population, offering a robust evidence base for prognosis and healthcare planning. Limitations and future perspectives Nonetheless, inherent limitations of a retrospective design must be acknowledged (25-29). Data quality is constrained by completeness and consistency of historical medical records, with potential for measurement bias. Residual confounding cannot be fully excluded despite multivariable adjustment. The study population is derived from hospital-based cohorts, which may limit generalizability to the community setting. Finally, outcomes with small event counts may yield imprecise risk estimates; such findings should be considered exploratory and require validation in future prospective studies. Importantly, the established cohort will serve as a valuable research platform for subsequent subgroup analyses, enabling investigation of heterogeneous disease trajectories across age strata and comorbidity clusters. Furthermore, these data provide a foundation for predictive model development and validation, with the potential to be integrated into clinical decision-support systems for individualized risk assessment.

Conclusions:

Conclusion Despite the limitations inherent to retrospective studies, the strengths of this multicenter, long-term design and rigorous methodology position this study to make significant contributions to the understanding of BPH natural history. Its findings are expected to provide high-quality evidence for clinical management, risk stratification, and precision intervention, ultimately improving patient outcomes and informing healthcare policy. Acknowledgments First and foremost, we would like to express our gratitude to the patients who will be included in the cohort for their contribution of clinical data to be used in the subsequent statistical analysis of this study. Additionally, we acknowledge the Big Data Center of PLA General Hospital for providing platform support for the data retrieval process of this study.