An estimated-score approach for dealing with missing covariate data in matched case-control studies

摘要

Purpose:To propose a robust method to handle the problem of missing covariate data in matched case-control studies, using an estimated conditional score approach when the missingness mechanism does not depend on the disease status.Summary:Due to retrospective nature of epidemiological studies for estimating the effect of exposure variables, information on a covariate could be missing for some subjects. A article uses an empirical procedure, within the conditional logistic likelihood framework, to estimate the odds of the disease for the subjects with missing covariate values. The asymptotic behavior of the proposed estimator is studied. Simulation results are reported to assess the finite sample performance. Two matched case-control studies are included as application examples. (18 refs.)