Confounding and regression adjustment in difference-in-differences studies.
Confounding and regression adjustment in difference-in-differences studies.
复制标题
DOI：
10.1111/1475-6773.13666
复制
发表时间：
2021-10
影响因子：
3.4
通讯作者：
Hatfield LA
中科院分区：
文献类型：
--
作者：
Zeldow B;Hatfield LA
定义差异性研究中的混杂偏差，并比较旨在纠正因观察到的混杂因素而产生的偏差的基于回归和匹配的估计器。我们模拟了包含不同混淆关系的线性模型的数据：对结果具有时变影响的时不变协变量，对结果具有恒定影响的时变协变量，以及对结果具有时变影响的时变协变量。我们考虑了一个在应用文献中常见的简单设置：治疗是在单个时间点引入的，并且没有观察不到的治疗效果异质性。我们比较了包括简单线性回归模型和匹配技术在内的六种模型规格的治疗效果估计的偏差和均方根误差。提供了仿真代码以供复制。差异差异中的混杂因素是协变量，在治疗组和对照组中，随着时间的推移变化不同，或者对结果有时间变化的影响。当测量这样一个混杂变量时，适当地调整这个混杂因素(即，在与因果模型一致的回归模型中包括混杂因素)可以提供具有最佳SE的无偏估计。然而，当时变的混杂因素受到治疗的影响时，使用差异中的差异来恢复无偏见的因果效应是困难的。差异中的混淆比横断面环境中的混淆更复杂，解决观察到的混淆的技术和直觉不能批量引入。相反，分析师应该首先假设一个因果模型，将协变量与治疗结果联系起来，无论是时变的还是那些对结果有时变影响的协变量。然后，这个因果模型将指导适当的分析模型的规范(例如，使用回归或匹配)，该模型可以产生无偏见的治疗效果估计。我们强调深思熟虑地纳入协变量的重要性，以解决差异研究中混淆的偏差。
To define confounding bias in difference‐in‐difference studies and compare regression‐ and matching‐based estimators designed to correct bias due to observed confounders. We simulated data from linear models that incorporated different confounding relationships: time‐invariant covariates with a time‐varying effect on the outcome, time‐varying covariates with a constant effect on the outcome, and time‐varying covariates with a time‐varying effect on the outcome. We considered a simple setting that is common in the applied literature: treatment is introduced at a single time point and there is no unobserved treatment effect heterogeneity. We compared the bias and root mean squared error of treatment effect estimates from six model specifications, including simple linear regression models and matching techniques. Simulation code is provided for replication. Confounders in difference‐in‐differences are covariates that change differently over time in the treated and comparison group or have a time‐varying effect on the outcome. When such a confounding variable is measured, appropriately adjusting for this confounder (ie, including the confounder in a regression model that is consistent with the causal model) can provide unbiased estimates with optimal SE. However, when a time‐varying confounder is affected by treatment, recovering an unbiased causal effect using difference‐in‐differences is difficult. Confounding in difference‐in‐differences is more complicated than in cross‐sectional settings, from which techniques and intuition to address observed confounding cannot be imported wholesale. Instead, analysts should begin by postulating a causal model that relates covariates, both time‐varying and those with time‐varying effects on the outcome, to treatment. This causal model will then guide the specification of an appropriate analytical model (eg, using regression or matching) that can produce unbiased treatment effect estimates. We emphasize the importance of thoughtful incorporation of covariates to address confounding bias in difference‐in‐difference studies.
登录
查看更多内容
DOI：
10.1007/s10742-014-0123-z
发表时间：
2014-12-01
影响因子：
1.5
作者：
Stuart, Elizabeth A;Huskamp, Haiden A;Duckworth, Kenneth;Simmons, Jeffrey;Song, Zirui;Chernew, Michael;Barry, Colleen L
通讯作者：
Barry, Colleen L
影响因子：
1.5
作者：
O'Neill S;Kreif N;Grieve R;Sutton M;Sekhon JS
通讯作者：
Sekhon JS
影响因子：
3.4
作者：
Karaca-Mandic, Pinar;Norton, Edward C.;Dowd, Bryan
通讯作者：
Dowd, Bryan
影响因子：
120.7
作者：
Desai, Sunita;Hatfield, Laura A.;Mehrotra, Ateev
通讯作者：
Mehrotra, Ateev
影响因子：
3.4
作者：
Rosenthal, Meredith B.;Landrum, Mary Beth;Schneider, Eric C.
通讯作者：
Schneider, Eric C.