实用性随机对照试验中的过程评估:系统综述
该系统综述通过两阶段系统检索,从2015年发表的31项实用性随机对照试验(RCT)中识别出17项独立的过程评估研究,并分析其频率、特征、标签、价值及可及性。研究发现过程评估标签含义不明确,但每项试验结果论文均报告过程数据(中位5个组件),过程评估虽有多方面价值但可见性差。
要点
- 过程评估在实用性RCT中常见但标签使用不一致,仅3/17项明确标注为过程评估。
- 每项试验结果论文均报告过程数据,中位5个过程评估组件。
- 过程评估可见性差:13/17项未在试验结果论文中提及,12/16篇期刊文章未出现在试验注册库。
核心结论
- 在综述的实用性RCT样本中,过程评估标签的含义不确定,其范围和意义需进一步研究和澄清。
- 过程评估虽以多种方式增加价值,但可见性往往较差。
研究设计与样本
- 采用两阶段系统检索:先识别2015年发表的实用性RCT主要结局论文索引样本(31项),再识别所有相关出版物。
- 基于医学研究委员会过程评估框架的操作性定义,识别单独报告的过程评估和试验结果论文中报告的过程数据。
- 提取并分析定量和定性数据以回答综述目标。
主要发现
- 从31项实用性RCT索引样本中识别出17项独立的过程评估研究,特征多样,仅3项标注为「过程评估」。
- 每项试验结果论文均报告过程数据,每项试验中位5个不同的过程评估组件。
- 报告了与实际收集过程数据、招募参与者参与过程评估及卫生服务研究法规相关的障碍和促进因素。
- 综合了过程评估对干预、试验和更广泛知识的广泛益处。
- 可见性差:13/17项过程评估未在试验结果论文中提及,12/16篇过程评估期刊文章未出现在试验注册库。
临床怎么用
- 在实用性RCT中应明确规划并报告过程评估,提高其可见性和可及性。
- 建议将过程评估与试验结果论文及试验注册库关联,以增强其效用。
局限性
- 样本仅基于2015年发表的实用性RCT索引样本,可能不涵盖其他年份或非实用性试验。
- 过程评估的操作性定义基于MRC框架,可能影响识别范围。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
BACKGROUND: Process evaluations are increasingly conducted within pragmatic randomised controlled trials (RCTs) of health services interventions and provide vital information to enhance understanding of RCT findings. However, issues pertaining to process evaluation in this specific context have been little discussed. We aimed to describe the frequency, characteristics, labelling, value, practical conduct issues, and accessibility of published process evaluations within pragmatic RCTs in health services research. METHODS: We used a 2-phase systematic search process to (1) identify an index sample of journal articles reporting primary outcome results of pragmatic RCTs published in 2015 and then (2) identify all associated publications. We used an operational definition of process evaluation based on the Medical Research Council's process evaluation framework to identify both process evaluations reported separately and process data reported in the trial results papers. We extracted and analysed quantitative and qualitative data to answer review objectives. RESULTS: From an index sample of 31 pragmatic RCTs, we identified 17 separate process evaluation studies. These had varied characteristics and only three were labelled 'process evaluation'. Each of the 31 trial results papers also reported process data, with a median of five different process evaluation components per trial. Reported barriers and facilitators related to real-world collection of process data, recruitment of participants to process evaluations, and health services research regulations. We synthesised a wide range of reported benefits of process evaluations to interventions, trials, and wider knowledge. Visibility was often poor, with 13/17 process evaluations not mentioned in the trial results paper and 12/16 process evaluation journal articles not appearing in the trial registry. CONCLUSIONS: In our sample of reviewed pragmatic RCTs, the meaning of the label 'process evaluation' appears uncertain, and the scope and significance of the term warrant further research and clarification. Although there were many ways in which the process evaluations added value, they often had poor visibility. Our findings suggest approaches that could enhance the planning and utility of process evaluations in the context of pragmatic RCTs. TRIAL REGISTRATION: Not applicable for PROSPERO registration.
引用
- French C, Pinnock H, Forbes G, et al. Process evaluation within pragmatic randomised controlled trials: what is it, why is it done, and can we find it?-a systematic review. Trials. 2020. PMID: 33168067. doi:10.1186/s13063-020-04762-9
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。