李文婷 方鹏骞

患者盲法在疼痛相关随机对照试验中重要性的meta分析

李文婷 方鹏骞

目的 通过对疼痛相关随机对照试验的meta分析，探讨患者盲法对结果偏倚的重要性，以期对完善我国临床试验中方法学方面的质量管理有所裨益。方法 检索MEDLINE、 EMBASE和the Cochrane Databases of Systematic Reviews这三大数据库，并筛选出有关慢性疼痛随机对照试验的meta分析。对于每个meta分析，使用随机效应模型计算下列干预的均差：①对患者采用盲法及不进行盲法；②分配隐匿和不进行分配隐匿。之后再合并各个meta分析的标准均差。 结果 最终有17篇meta分析纳入研究，包括了93个随机对照试验，其中57个汇报了不充分盲法，36个为充分盲法。合并的标准均差为0.09（95% CI ：－0.00～0.19），说明对患者进行盲法的试验结果不佳。 结论 慢性疼痛临床试验结果会受到对患者是否采用了盲法的影响。

疼痛；患者盲法；随机对照试验；meta分析

【Abstract】Objective The paper discusses about the importance of patient blind method to bias of the results by carrying out meta analysis of pain related randomized controlled trial in order to assist in improving quality management with respect to methodology in clinical trials in China.Method Three databases including MEDLINE，EMBASE and the Cochrane Databases of Systematic Reviews are researched and meta analysis of chronic pain related randomized controlled trial is selected out.For each meta analysis，mean difference of the following interventions is calculated by using random effect model:blind method is carried out for the patients or not；allocation concealment and non-allocation concealment.After that，standard mean difference of each meta analysis is combined.Result 17 papers of meta analysis are incorporated into the research including 93 randomized controlled trials，57 of which report inadequate blind method and 36 use adequate blind method.Combined standard mean difference is 0.09 （95% CI:-0.09-0/19），which means tests of blind method for patients are not good.Conclusion Whether blind method is adopted for the patients will have an effect on the result of chronic pain clinical trial.

【Key words】Pain，Blind method for patient；Randomized controlled trial；Meta analysis.

目前，纳入随机对照试验的meta分析被视为评价临床试验的最佳证据，但其也存在着结果偏倚的可能。若随机对照试验在方法学方面存在质量管理缺陷则会影响到结果，从而影响到纳入这些研究的meta分析结果。任何相关的偏倚都会严重影响到meta分析的结果和结论，尤其是主观结果［1］。慢性疼痛，一个常见的患者主观报告结果，在新药注册及资源分配方面发挥了及其重要的作用。患者盲法在随机对照临床试验中具有非常重要的作用并影响结果的偏倚［2］。为了探索患者盲法在疼痛相关临床随机对照试验中的重要性及其对结果偏倚的影响，本文对疼痛相关随机对照试验进行了meta分析。……