Investigating differences in treatment effect estimates between propensity score matching and weighting: a demonstration using STAR*D trial data.
Investigating differences in treatment effect estimates between propensity score matching and weighting: a demonstration using STAR*D trial data.
复制标题
DOI：
10.1002/pds.3396
复制
发表时间：
2013-02
影响因子：
2.6
通讯作者：
Stürmer T
中科院分区：
文献类型：
--
作者：
Ellis AR;Dusetzina SB;Hansen RA;Gaynes BN;Farley JF;Stürmer T
倾向评分（PS）实施的选择影响治疗效果估计，不仅因为不同的方法估计不同的数量，而且因为不同的估计以不同的方式响应的现象，如治疗效果的异质性和有限的可用性的潜在匹配。使用有效性数据，我们描述了匹配和加权估计的敏感性分析的经验教训。使用2001-2004年抑郁症治疗有效性试验“缓解抑郁症的序贯治疗替代方案”的子样本数据（N= 1，292），我们实施了PS匹配和加权来估计治疗效果，并进行了多重敏感性分析。匹配和加权平衡了两个协变量，但产生了不同的样本和治疗效应估计值（匹配RR 1.00，95% CI：0.75-1.34;加权RR 1.28，95% CI：0.97-1.69）。在敏感性分析中，随着从加权分析中排除PS分布两端的观察值数量的增加，加权估计值接近匹配估计值（排除治疗组第5百分位数以下和未治疗组第95百分位数以上的所有观察值后，加权RR 1.04，95% CI 0.77-1.39）。治疗似乎仅在最高和最低PS分层中获益。由于不完全匹配、加权估计值对极端观察结果的敏感性以及可能的治疗效应异质性，匹配和加权估计值存在差异。PS分析需要确定关注的人群和治疗效果，选择适当的实施方法，并进行和报告敏感性分析。加权估计尤其应包括与有影响的观测有关的敏感性分析，例如与预测相反的观测。
The choice of propensity score (PS) implementation influences treatment effect estimates not only because different methods estimate different quantities, but also because different estimators respond in different ways to phenomena such as treatment effect heterogeneity and limited availability of potential matches. Using effectiveness data, we describe lessons learned from sensitivity analyses with matched and weighted estimates. With subsample data (N=1,292) from Sequenced Treatment Alternatives to Relieve Depression, a 2001–2004 effectiveness trial of depression treatments, we implemented PS matching and weighting to estimate the treatment effect in the treated and conducted multiple sensitivity analyses. Matching and weighting both balanced covariates but yielded different samples and treatment effect estimates (matched RR 1.00, 95% CI:0.75–1.34; weighted RR 1.28, 95% CI:0.97–1.69). In sensitivity analyses, as increasing numbers of observations at both ends of the PS distribution were excluded from the weighted analysis, weighted estimates approached the matched estimate (weighted RR 1.04, 95% CI 0.77–1.39 after excluding all observations below the 5th percentile of the treated and above the 95th percentile of the untreated). Treatment appeared to have benefits only in the highest and lowest PS strata. Matched and weighted estimates differed due to incomplete matching, sensitivity of weighted estimates to extreme observations, and possibly treatment effect heterogeneity. PS analysis requires identifying the population and treatment effect of interest, selecting an appropriate implementation method, and conducting and reporting sensitivity analyses. Weighted estimation especially should include sensitivity analyses relating to influential observations, such as those treated contrary to prediction.
登录
查看更多内容
影响因子：
2.9
作者：
Gaynes, Bradley N.;Dusetzina, Stacie B.;Stuermer, Til
通讯作者：
Stuermer, Til
影响因子：
1.9
作者：
ROSENBAUM, PR;RUBIN, DB
通讯作者：
RUBIN, DB
影响因子：
0.9
作者：
Freedman, David A.;Berk, Richard A.
通讯作者：
Berk, Richard A.
影响因子：
7
作者：
McCaffrey, DF;Ridgeway, G;Morral, AR
通讯作者：
Morral, AR
影响因子：
--
作者：
Judd, LL;Akiskal, HS;Keller, MB
通讯作者：
Keller, MB