期刊
PHARMACOEPIDEMIOLOGY AND DRUG SAFETY
卷 30, 期 11, 页码 1471-1485出版社
WILEY
DOI: 10.1002/pds.5338
关键词
extrapolation; generalizability; overlap; propensity score; trimming; weighting
资金
- National Cancer Institute [R21CA227613, R01CA172973]
- National Institutes of Health
The positivity assumption in causal analysis of observational data requires that all treatments of interest be observed in every patient subgroup, and violations of this assumption can lead to challenges in estimation and inference. This paper emphasizes the importance of the positivity assumption, discusses the implications of violations, and provides insight into appropriate methods for different types of violations. Additionally, a case study using electronic health record-derived data is used to demonstrate alternative approaches and relevant considerations for addressing positivity violations.
In the causal analysis of observational data, the positivity assumption requires that all treatments of interest be observed in every patient subgroup. Violations of this assumption are indicated by nonoverlap in the data in the sense that patients with certain covariate combinations are not observed to receive a treatment of interest, which may arise from contraindications to treatment or small sample size. In this paper, we emphasize the importance and implications of this often-overlooked assumption. Further, we elaborate on the challenges nonoverlap poses to estimation and inference and discuss previously proposed methods. We distinguish between structural and practical violations and provide insight into which methods are appropriate for each. To demonstrate alternative approaches and relevant considerations (including how overlap is defined and the target population to which results may be generalized) when addressing positivity violations, we employ an electronic health record-derived data set to assess the effects of metformin on colon cancer recurrence among diabetic patients.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据