Variance estimation for logistic regression in case-cohort studies

TL;DR

This paper identifies biases in the standard robust variance estimator for logistic regression in case-cohort studies and proposes a bootstrap-based alternative that improves the accuracy of confidence intervals and P-values.

Contribution

It introduces a bootstrap-based variance estimator to correct biases in the traditional robust variance method for case-cohort logistic regression analysis.

Findings

Bootstrap method yields more precise confidence intervals.

Bootstrap maintains adequate coverage probabilities.

Robust variance estimator can be biased and lead to inaccurate inferences.

Abstract

The logistic regression analysis proposed by Schouten et al. (Stat Med. 1993;12:1733-1745) has been a standard method in current statistical analysis of case-cohort studies, and it enables effective estimation of risk ratio from selected subsamples. Schouten et al. (1993) also proposed the standard error estimate of the risk ratio estimator can be calculated by the robust variance estimator. In this article, however, we show that the robust variance estimator does not account for the duplications of case and subcohort samples and generally has certain bias, i.e., inaccurate confidence intervals and P-values are possibly obtained. To address the invalid statistical inference problem, we provide an alternative bootstrap-based valid variance estimator. Through simulation studies, the bootstrap method consistently provided more precise confidence intervals compared with those provided by the robust variance method, while retaining adequate coverage probabilities. The conventional robust variance estimator has certain bias, and inadequate conclusions might be deduced. The bootstrap method would be an alternative effective approach in practice to provide accurate evidence.