Measuring balance and model selection in propensity score methods
Measuring balance and model selection in propensity score methods
复制标题
DOI：
10.1002/pds.2188
复制
发表时间：
2011-11-01
影响因子：
2.6
通讯作者：
Klungel, Olaf H.
中科院分区：
文献类型：
--
作者：
Belitser, Svetlana V.;Martens, Edwin P.;Klungel, Olaf H.
目的倾向评分（PS）方法侧重于平衡组间混杂因素，以估计无偏治疗或暴露效应。然而，在实际测量、报告和使用平衡信息方面缺乏关注，例如在模型选择方面。我们建议在PS方法中使用一种衡量平衡的方法，并描述了几种这样的衡量方法：重叠系数、Kolmogorov-Smirnov距离和Levy距离。方法我们进行了模拟研究，以估计这三个和几个基于平均值的平衡和偏倚措施之间的关联（即真实和估计的治疗效果之间的差异）。结果对于大样本量（n = 2000），偏差与Kolmogorov-Smirnov距离（r = 0.89）、Levy距离（r = 0.89）和绝对标准化平均差（r = 0.90）之间的平均Pearson相关系数相似，而重叠系数（r = -0.42）较低。当样本量减少到400时，基于平均值的平衡测量与偏差有更强的相关性。包含所有混杂变量及其平方和相互作用项的模型比只包含混杂变量主要项的模型产生更小的偏差。我们的结论是，平衡的措施是有用的报告平衡达到倾向得分分析的数量，可以帮助选择最终的PS模型。版权所有：John Wiley & Sons， Ltd。
Purpose Propensity score (PS) methods focus on balancing confounders between groups to estimate an unbiased treatment or exposure effect. However, there is lack of attention in actually measuring, reporting and using the information on balance, for instance for model selection. We propose to use a measure for balance in PS methods and describe several of such measures: the overlapping coefficient, the Kolmogorov-Smirnov distance, and the Levy distance.Methods We performed simulation studies to estimate the association between these three and several mean based measures for balance and bias (i.e., discrepancy between the true and the estimated treatment effect).Results For large sample sizes (n = 2000) the average Pearson's correlation coefficients between bias and Kolmogorov-Smirnov distance (r = 0.89), the Levy distance (r = 0.89) and the absolute standardized mean difference (r = 0.90) were similar, whereas this was lower for the overlapping coefficient (r = -0.42). When sample size decreased to 400, mean based measures of balance had stronger correlations with bias. Models including all confounding variables, their squares and interaction terms resulted in smaller bias than models that included only main terms for confounding variables.Conclusions We conclude that measures for balance are useful for reporting the amount of balance reached in propensity score analysis and can be helpful in selecting the final PS model. Copyright (C) 2011 John Wiley & Sons, Ltd.