Haas David M, Kirkpatrick Page, McIntosh Jennifer J, Caldwell Deborah M
Department of Obstetrics and Gynecology, Indiana University School of Medicine, Indianapolis, IN, USA.
J Matern Fetal Neonatal Med. 2012 Sep;25(9):1646-52. doi: 10.3109/14767058.2011.650249. Epub 2012 Jan 23.
To assess the quality of tocolysis randomized controlled trials (RCTs) and to determine trial factors contributing to better quality evidence.
The Cochrane Central Register of Controlled Trials, MEDLINE, MEDLINE In-Process, EMBASE and CINAHL were searched for terms "preterm labor," "tocolytic" or "obstetric labor, premature" up to 1 August 2009.Data regarding study design, characteristics, number of participants and outcomes reported were extracted by at least two review authors. Study quality was assigned using the Cochrane Collaboration Handbook methodology and categories. Trends for quality over time, the impact of study size and the individual drugs compared were analyzed for impact on overall quality of trials.
Of the 3197 titles initially identified, 89 RCTs of tocolytic therapy were reviewed. Of the six quality areas, 10 (11.2%) trials satisfied all areas, while only one trial (1.1%) met one area. The mean number of adequate areas was 4.1 ± 1.2. Overall, 52 (58.4%) of the trials achieved high-quality categorization. Controlling for multiple trial factors, the trial continent and decade were significant predictors of overall trial quality.
The majority of tocolysis RCTs are of high quality. Larger trials, more recent trials and placebo-controlled trials were associated with higher quality scores.
评估宫缩抑制剂随机对照试验（RCTs）的质量，并确定有助于获得更高质量证据的试验因素。
检索Cochrane对照试验中央注册库、MEDLINE、MEDLINE在研数据库、EMBASE和CINAHL，检索截至2009年8月1日的“早产”“宫缩抑制剂”或“早产临产”等术语。至少两名综述作者提取有关研究设计、特征、参与者数量和报告结果的数据。使用Cochrane协作网手册方法和类别对研究质量进行评定。分析质量随时间的趋势、研究规模的影响以及所比较的各药物对试验总体质量的影响。
在最初识别的3197篇标题中，对89项宫缩抑制剂治疗的随机对照试验进行了综述。在六个质量领域中，10项试验（11.2%）满足所有领域，而只有1项试验（1.1%）满足一个领域。充分领域的平均数量为4.1±1.2。总体而言，52项试验（58.4%）获得了高质量评定。在控制多个试验因素后，试验所在洲和年代是试验总体质量的显著预测因素。
大多数宫缩抑制剂随机对照试验质量较高。规模较大的试验、较新的试验以及安慰剂对照试验的质量得分较高。
