Schill W, Jöckel K H
Bremen Institute for Prevention Research and Social Medicine (BIPS), Germany.
Eur J Clin Nutr. 1993 Oct;47 Suppl 2:S34-41.
In case-control studies, one often finds that covariates are missing or measured with error in the entire sample, whereas complete or exact covariate information is available only in a subsample. This paper discusses the analysis of case-control studies under a double-sampling scheme, where at the first stage covariates are measured with error or missing, and at the second stage are validated in a subsample. The method proposed combines the risk information from both samples by assuming that (1) the disease incidence model is logistic, (2) the partial or proxy information takes on finitely many values, and (3) the error is non-differential. The estimator is obtained by jointly fitting logistic models to the first and second stage data, a variance formula is presented. Parameters can be estimated by use of standard packages for dose-response data. Data from an ongoing case-control study on lung cancer serve as an example.
在病例对照研究中，人们常常发现协变量在整个样本中缺失或存在测量误差，而完整或精确的协变量信息仅在一个子样本中可用。本文讨论了在双重抽样方案下病例对照研究的分析方法，即在第一阶段协变量存在测量误差或缺失，而在第二阶段在一个子样本中进行验证。所提出的方法通过假设（1）疾病发病率模型为逻辑模型，（2）部分或替代信息取有限多个值，以及（3）误差是非差异性的，来结合两个样本的风险信息。通过将逻辑模型联合拟合到第一阶段和第二阶段数据来获得估计量，并给出了方差公式。参数可以通过使用剂量反应数据的标准软件包进行估计。以一项正在进行的肺癌病例对照研究的数据为例。
