Cenzer Irena, Boscardin W John, Berger Karin
Department of Medicine III, University Hospital, LMU Munich, Munich, Germany.
Division of Geriatrics, University of California, San Francisco, California, USA.
Intractable Rare Dis Res. 2020 May;9(2):79-88. doi: 10.5582/irdr.2020.01016.
Matching is a common method of adjusting for confounding in observational studies. Studies in rare diseases usually include small numbers of exposed subjects, but the performance of matching methods in such cases has not been evaluated thoroughly. In this study, we compare the performance of several matching methods when number of exposed subjects is small. We used Monte Carlo simulations to compare the following methods: Propensity score matching (PSM) with greedy or optimal algorithm, Mahalanobis distance matching, and mixture of PSM and exact matching. We performed the comparisons in datasets with six continuous and six binary variables, with varying effect size on group assignment and outcome. In each case, there were 1,500 unexposed subjects and a varying number of exposed: = 25, 50, 100, 150, 200, 250, or 300. The probability of outcome in unexposed subjects was set to 5% (rare), 20% (common), or 50% (frequent). We compared the methods based on the bias of estimate of risk difference, coverage of 95% confidence intervals for risk difference, and balance of covariates. We observed a difference in performance of matching methods in very small samples ( = 25-50) and in moderately small samples ( = 100-300). Our study showed that PSM performs better than other matching methods when number of exposed subjects is small, but the matching algorithm and the matching ratio should be considered carefully. We recommend using PSM with optimal algorithm and one-to-five matching ratio in very small samples, and PSM matching with any algorithm and one-to-one matching in moderately small samples.
匹配是观察性研究中调整混杂因素的常用方法。罕见病研究通常纳入的暴露对象数量较少，但此类情况下匹配方法的性能尚未得到充分评估。在本研究中，我们比较了暴露对象数量较少时几种匹配方法的性能。我们使用蒙特卡洛模拟来比较以下方法：采用贪婪或最优算法的倾向得分匹配（PSM）、马氏距离匹配以及PSM与精确匹配的混合方法。我们在具有六个连续变量和六个二元变量的数据集上进行比较，组分配和结局的效应大小各不相同。在每种情况下，有1500名未暴露对象以及数量不等的暴露对象：n = 25、50、100、150、200、250或300。未暴露对象的结局概率设定为5%（罕见）、20%（常见）或50%（频繁）。我们基于风险差异估计的偏差、风险差异95%置信区间的覆盖范围以及协变量的平衡来比较这些方法。我们观察到在极小样本（n = 25 - 50）和中等小样本（n = 100 - 300）中匹配方法的性能存在差异。我们的研究表明，当暴露对象数量较少时，PSM的性能优于其他匹配方法，但应仔细考虑匹配算法和匹配比例。我们建议在极小样本中使用具有最优算法和1:5匹配比例的PSM，在中等小样本中使用具有任何算法和1:1匹配比例的PSM匹配。
