Optimal Designs of Two-Phase Case-Control Studies for General Predictor Effects

TL;DR

This paper introduces an optimal two-phase sampling design for case-control studies that improves the estimation of predictor effects, including both local and non-local effects, using pseudo conditional likelihood estimators.

Contribution

It extends existing two-phase designs to optimize estimation of predictor effects beyond local effects, applicable to case-control studies with costly predictors.

Findings

Simulation studies confirm improved estimation efficiency.

Application to COVID-19 data demonstrates practical effectiveness.

Significant reduction in estimation variance.

Abstract

Under two-phase designs, the outcome and several covariates and confounders are measured in the first phase, and a new predictor of interest, which may be costly to collect, can be measured on a subsample in the second phase, without incurring the costs of recruiting subjects. By using the information gathered in the first phase, the second-phase subsample can be selected to enhance the efficiency of testing and estimating the effect of the new predictor on the outcome. Past studies have focused on optimal two-phase sampling schemes for statistical inference on local ($\beta = o(1)$) effects of the predictor of interest. In this study, we propose an extension of the two-phase designs that employs an optimal sampling scheme for estimating predictor effects with pseudo conditional likelihood estimators in case-control studies. This approach is applicable to both local and non-local effects. We demonstrate the effectiveness of the proposed sampling scheme through simulation studies and analysis of data from 170 patients hospitalized for treatment of COVID-19. The results show a significant improvement in the estimation of the parameter of interest.