Evaluation of safety reporting in interventional trials of benign prostatic hyperplasia: a registry-publication comparison study
摘要
To assess the concordance of adverse event reporting between clinical trial registries and their corresponding publications for interventions treating benign prostatic hyperplasia.
Materials and MethodsWe conducted a registry-to-publication comparison study to evaluate the completeness and concordance of adverse event reporting in benign prostatic hyperplasia. Studies were identified via searches of ClinicalTrials.gov studies with results posted from September 2009 to December 2024. We identified corresponding publications via ClinicalTrials.gov, PubMed, Google Scholar, pharmaceutical company portals, and Cochrane reviews. Data extraction was performed independently and in duplicate by masked reviewers. A quantitative analysis used descriptive statistics, Bland-Altman plots, funnel plots, and regression to assess concordance and identify predictors of incomplete reporting. Due to the study's focus on concordance, traditional risk of bias assessment was not conducted, however robust systematic review methods were employed.
ResultsOf 36 included benign prostatic hyperplasia trials, registry reporting of adverse events was more complete than in publications. Post-Final Rule, overall adverse events and death reporting improved in registries but declined or remained inconsistent in manuscripts. Over 70% of trials showed discrepancies in serious adverse events counts, with most publications underreporting events compared to ClinicalTrials.gov.
ConclusionsIn benign prostatic hyperplasia trials, adverse event data were more complete in ClinicalTrials.gov than in corresponding publications. Serious harm reporting was inconsistent, and post-Final Rule improvements in registry data were not mirrored in print. These gaps risk distorting clinical interpretation and highlight the need for aligned reporting standards to ensure transparent, patient-relevant safety communication.
Trial registrationRegistered on the Open Science Framework.