Random-effects meta-analysis of combined outcomes based on reconstructions of individual patient data.
Random-effects meta-analysis of combined outcomes based on reconstructions of individual patient data.
复制标题
基于个体患者数据重建的综合结果随机效应荟萃分析。
DOI：
10.1002/jrsm.1406
复制
发表时间：
2020-09
影响因子：
9.8
通讯作者：
Wang R
中科院分区：
文献类型：
--
作者：
Song Y;Sun F;Redline S;Wang R
临床试验的荟萃分析通常一次关注一个结果。然而，治疗决策取决于对结果的总体评估，平衡各个领域的益处和潜在风险。这就需要对包含来自不同领域的信息的组合结果进行荟萃分析方法。当所有研究中都有个体患者数据 (IPD) 时，可以计算每个个体的综合结果，并应用标准荟萃分析方法。然而，IPD 通常很难获得。我们提出了一种估计组合结果的总体治疗效果的方法，该方法首先根据可用的汇总统计数据重建伪 IPD，然后汇总多个重建数据集的估计值。我们专注于由两个连续的原始结果构建的组合结果。重建步骤需要指定这两个原始结果的联合分布，包括通常未知的相关性。对于以线性方式组合的结果，这种相关性的错误指定会影响所得治疗效果估计量的效率，但不会影响一致性。对于其他组合结果，需要准确估计相关性以确保治疗效果估计的一致性。为此，我们提出了几种方法来估计不同数据可用性场景下的这种相关性。我们通过模拟研究评估所提出方法的性能，并将其应用于两个例子：（1）二肽基肽酶 4 抑制剂与对照治疗 2 型糖尿病的荟萃分析； (2) 气道正压通气治疗与阻塞性睡眠呼吸暂停患者降压控制的荟萃分析。
Meta-analyses of clinical trials typically focus on one outcome at a time. However, treatment decision-making depends on an overall assessment of outcomes balancing benefit in various domains and potential risks. This calls for meta-analysis methods for combined outcomes that encompass information from different domains. When individual patient data (IPD) are available from all studies, combined outcomes can be calculated for each individual and standard meta-analysis methods would apply. However, IPD are usually difficult to obtain. We propose a method to estimate the overall treatment effect for combined outcomes based on first reconstructing pseudo IPD from available summary statistics and then pooling estimates from multiple reconstructed datasets. We focus on combined outcomes constructed from two continuous original outcomes. The reconstruction step requires the specification of the joint distribution of these two original outcomes, including the correlation which is often unknown. For outcomes that are combined in a linear fashion, misspecifications of this correlation affect efficiency, but not consistency, of the resulting treatment effect estimator. For other combined outcomes, an accurate estimate of the correlation is necessary to ensure the consistency of treatment effect estimates. To this end, we propose several ways to estimate this correlation under different data availability scenarios. We evaluate the performance of the proposed methods through simulation studies and apply these to two examples: (1) a meta-analysis of dipeptidyl peptidase-4 inhibitors versus control on treating type 2 diabetes; (2) a meta-analysis of positive airway pressure therapy versus control on lowering blood pressure among patients with obstructive sleep apnea.
登录
查看更多内容
影响因子：
2
作者：
Chen, Yong;Hong, Chuan;Riley, Richard D.
通讯作者：
Riley, Richard D.
影响因子：
2.2
作者：
DerSimonian, Rebecca;Kacker, Raghu
通讯作者：
Kacker, Raghu
影响因子：
37.8
作者：
Becker, HF;Jerrentrup, A;Peter, JH
通讯作者：
Peter, JH
影响因子：
3
作者：
Elixhauser, A;Steiner, C;Coffey, RN
通讯作者：
Coffey, RN
DOI：
10.1016/0197-2456(86)90046-2
发表时间：
1986-09-01
期刊：
CONTROLLED CLINICAL TRIALS
影响因子：
--
作者：
DERSIMONIAN, R;LAIRD, N
通讯作者：
LAIRD, N