High statistical heterogeneity is more frequent in meta-analysis of continuous than binary outcomes
High statistical heterogeneity is more frequent in meta-analysis of continuous than binary outcomes
复制标题
DOI：
10.1016/j.jclinepi.2015.09.005
复制
发表时间：
2016-02-01
影响因子：
7.2
通讯作者：
Guyatt, Gordon H.
中科院分区：
文献类型：
--
作者：
Alba, Ana C.;Alexander, Paul E.;Guyatt, Gordon H.
目的：我们比较二元结局和连续结局meta分析的异质性分布。研究设计和设置：我们在MEDLINE和Cochrane数据库中检索了2012年发表的随机试验的meta分析引文，这些试验报告了二元或连续结果的异质性。两名审稿人独立进行资格筛选和数据提取。我们在二元和连续结果的荟萃分析中评估了I-2的分布，并探讨了解释分布差异的假设。结果：经过全文筛选，我们选择了671项荟萃分析，评估了557项二元结果和352项连续结果。由I-2评估的异质性证明在连续结果中高于二元结果：报告I-2为0%的连续和二元结果的比例分别为34%和52%，报告I-2为60-100%的比例为39%和14%。在连续但非二元结果中，I-2随着meta分析中纳入的研究数量的增加而增加。精确度和样本量的增加并不能解释在大量研究的连续结果荟萃分析中发现的较大I-2。结论：评价连续结局的荟萃分析显示I-2明显高于二元结局的荟萃分析。结果表明，不同的标准解释I-2在连续和二进制结果可能是合适的。(C) 2016 Elsevier Inc.版权所有。
Objectives: We compared the distribution of heterogeneity in meta-analyses of binary and continuous outcomes.Study Design and Setting: We searched citations in MEDLINE and Cochrane databases for meta-analyses of randomized trials published in 2012 that reported a measure of heterogeneity of either binary or continuous outcomes. Two reviewers independently performed eligibility screening and data abstraction. We evaluated the distribution of I-2 in meta-analyses of binary and continuous outcomes and explored hypotheses explaining the difference in distributions.Results: After full-text screening, we selected 671 meta-analyses evaluating 557 binary and 352 continuous outcomes. Heterogeneity as assessed by I-2 proved higher in continuous than in binary outcomes: the proportion of continuous and binary outcomes reporting an I-2 of 0% was 34% vs. 52%, respectively, and reporting an I-2 of 60-100% was 39% vs. 14%. In continuous but not binary outcomes, I-2 increased with larger number of studies included in a meta-analysis. Increased precision and sample size do not explain the larger I-2 found in meta analyses of continuous outcomes with a larger number of studies.Conclusions: Meta-analyses evaluating continuous outcomes showed substantially higher I-2 than meta-analyses of binary outcomes. Results suggest differing standards for interpreting I-2 in continuous vs. binary outcomes may be appropriate. (C) 2016 Elsevier Inc. All rights reserved.