Methods for Large-scale Single Mediator Hypothesis Testing: Possible Choices and Comparisons

TL;DR

This paper reviews and compares methods for large-scale mediation hypothesis testing, introduces a new Sobel-comp method with a corrected distribution, and provides practical guidelines based on extensive simulations and real data application.

Contribution

It develops the Sobel-comp method with a corrected mixture distribution and offers comprehensive comparisons and practical guidelines for mediation testing.

Findings

Mixture distribution methods best control false positives and maximize true positives.

Sobel-comp outperforms existing methods in simulations.

Guidelines for method selection in practice are provided.

Abstract

Mediation hypothesis testing for a large number of mediators is challenging due to the composite structure of the null hypothesis, H0:alpha*beta=0 (alpha: effect of the exposure on the mediator after adjusting for confounders; beta: effect of the mediator on the outcome after adjusting for exposure and confounders). In this paper, we reviewed three classes of methods for multiple mediation hypothesis testing. In addition to these existing methods, we developed the Sobel-comp method, which uses a corrected mixture reference distribution for Sobel's test statistic. We performed extensive simulation studies to compare all six methods in terms of the false positive rates under the null hypothesis and the true positive rates under the alternative hypothesis. We found that the class of methods which uses a mixture reference distribution could best maintain the false positive rates at the nominal level under the null hypothesis and had the greatest true positive rates under the alternative hypothesis. We applied all methods to study the mediation mechanism of DNA methylation sites in the pathway from adult socioeconomic status to glycated hemoglobin level using data from the Multi-Ethnic Study of Atherosclerosis (MESA). We also provide guidelines for choosing the optimal mediation hypothesis testing method in practice.