Pitre Tyler, Kirsh Sarah, Jassal Tanvir, Anderson Mason, Padoan Adelia, Xiang Alexander, Mah Jasmine, Zeraatkar Dena
Department of Medicine McMaster University Hamilton Ontario Canada.
Department of Anesthesia McMaster University Hamilton Ontario Canada.
Cochrane Evid Synth Methods. 2023 Jun 20;1(4):e12015. doi: 10.1002/cesm.12015. eCollection 2023 Jun.
Blinding-the concealment of the arm to which participants have been randomized-is an important consideration for assessing the risk of bias of randomized trials. A growing body of evidence has, however, yielded inconsistent results on whether trials without blinding produce biased findings.
To conduct a systematic review and meta-analysis of the evidence addressing whether trials with and without blinding produce different results.
We searched MEDLINE, EMBASE, Cochrane Reviews, JBI EBP, and Web of Science, from inception to May 2022, for studies comparing the results of trials with and without blinding. Pairs of reviewers, working independently and in duplicate, reviewed search results for eligible studies and extracted data. We pooled the results of studies comparing trials with and without blinding of patients, healthcare providers/investigators, and outcome assessors/adjudicators using frequentist random-effects meta-analyses. We coded study results such that a ratio of odds ratio < 1 and difference in standardized mean difference < 0 indicate that trials without blinding overestimate the beneficial effects of treatments.
We identified 47 eligible studies. For dichotomous outcomes, we found low certainty evidence that trials without blinding of patients and healthcare providers and trials without blinding of patients may slightly overestimate the beneficial effects of treatments. We found moderate certainty evidence that trials without blinding of outcome assessors overestimate the beneficial effects of treatments. For continuous outcomes, we found low certainty evidence that trials without blinding of patients and healthcare providers may overestimate the beneficial effects of treatments. We found moderate certainty evidence that trials without blinding of outcome assessors/adjudicators probably overestimate the beneficial effects of treatments.
Our systematic review and meta-analysis suggest that blinding may influence trial results in select situations-although the findings are of low certainty and the magnitude of effect is modest. In the absence of high-certainty evidence suggesting that trials with and without blinding produce similar results, investigators should be cautious about interpreting the results of trials without blinding.
设盲——对参与者被随机分配至的组别进行隐藏——是评估随机试验偏倚风险时的一个重要考量因素。然而，越来越多的证据对于未设盲的试验是否会产生有偏倚的结果给出了不一致的结论。
对有关设盲和未设盲的试验是否产生不同结果的证据进行系统评价和荟萃分析。
我们检索了MEDLINE、EMBASE、Cochrane系统评价、JBI循证卫生保健数据库和科学网，检索时间从各数据库建库至2022年5月，以查找比较设盲和未设盲试验结果的研究。由两组评审员独立且重复地对检索结果进行评审，以筛选出符合条件的研究并提取数据。我们使用频率学派随机效应荟萃分析对比较患者、医疗保健提供者/研究者以及结局评估者/裁决者设盲和未设盲试验结果的研究结果进行汇总。我们对研究结果进行编码，使得优势比的比值<1以及标准化均数差的差值<0表明未设盲的试验高估了治疗的有益效果。
我们确定了47项符合条件的研究。对于二分结局，我们发现低确定性证据表明，患者和医疗保健提供者未设盲的试验以及患者未设盲的试验可能会略微高估治疗的有益效果。我们发现中度确定性证据表明，结局评估者未设盲的试验高估了治疗的有益效果。对于连续性结局，我们发现低确定性证据表明，患者和医疗保健提供者未设盲的试验可能会高估治疗的有益效果。我们发现中度确定性证据表明，结局评估者/裁决者未设盲的试验可能高估了治疗的有益效果。
我们的系统评价和荟萃分析表明，设盲可能在某些情况下影响试验结果——尽管研究结果的确定性较低且效应量较小。在缺乏高确定性证据表明设盲和未设盲的试验产生相似结果的情况下，研究人员在解释未设盲试验的结果时应谨慎。
