EVALUATION OF SAFETY DATA REPORTING IN RANDOMIZED TRIALS

我们建议评估已发表的跨多个医学学科的随机试验中安全性信息的质量和范围，并确定影响这些报告中安全性数据报告的因素。这些试验提供了一个独特的机会，在一个控制的、全面的环境中收集有关新干预措施的毒性概况的信息。这些信息不能从小规模、选择性的早期I/II期试验或零星的上市后经验中替代。随机试验安全性报告的评价是一个被忽视的领域。然而，临床医生不能做出明智的风险-收益决定，没有足够的信息，具体的医疗干预的毒性超出疗效估计。我们制定了评估随机试验中安全性报告的量表，并将其应用于艾滋病毒治疗领域的所有试验。在由于毒性和特定副作用而停药的报告中观察到明显的缺陷。分配给安全结果的日志空间略小于分配给贡献者名称的日志空间。在本项目中，我们将在7个不同临床领域的随机比较试验综合数据库中评估安全性报告的定性和定量充分性；HIV感染、急性心肌梗死的溶栓治疗、老年人高血压的治疗、急性鼻窦炎的抗生素治疗、类风湿关节炎的非甾体药物治疗、幽门螺杆菌的治疗以及重症监护室中胃肠道的选择性去污。对这些领域的安全报告估计进行荟萃分析，以评估特定领域的异质性，并在适当的情况下计算总体汇总估计。该项目预计将发现目前安全数据报告中的不足之处，并展示未来如何改善这种情况，从而促进循证医学在临床实践中的应用潜力。

We propose to evaluate the quality and extent of safety information in published randomized trials across several medical disciplines and to identify factors that affect the reporting of safety data in these reports. Such trials offer a unique opportunity for assembling information on the toxicity profile of new interventions in a controlled, comprehensive setting. This information cannot be substituted from small, selective early phase I/II trials or the sporadic post-marketing experience. The evaluation of safety reporting from randomized trials is a neglected field. However, clinicians cannot make informed risk-benefit decisions without adequate information on the toxicity of specific medical interventions beyond efficacy estimates. We have developed scales for evaluating safety reporting in randomized trials and have applied them to all trials in the field of HIV therapeutics. Glaring deficiencies were observed in the reporting of discontinuations due to toxicity and the reporting of specific side effects. The journal space allocated to safety results was slightly less than that allocated for contributor names. In this project, we will assess the qualitative and quantitative adequacy of safety reporting in comprehensive databases of randomized comparative trials in 7 different clinical fields; HIV infection, thrombolytic therapy in acute myocardial infarction, treatment of hypertension in the elderly, antibiotic therapy for acute sinusitis, non-steroidal agents for rheumatoid arthritis, treatment for Helicobacter pylori, and selective decontamination of the gastrointestinal tract in the intensive care setting. A meta-analysis of the estimates of safety reporting across these fields will be used to assess field-specific heterogeneity and to calculate overall pooled estimates, if appropriate. The project is expected to detect deficiencies in current reporting of safety data and show how the situation can be improved in the future, thus promoting the potential for the application of evidence-based medicine in clinical practice.