Copas' method is sensitive to different mechanisms of publication bias

TL;DR

This paper evaluates Copas' method for correcting publication bias in meta-analyses, revealing its sensitivity to various bias mechanisms and questioning its robustness and practical usefulness.

Contribution

The study demonstrates through simulations that Copas' method is not robust against multiple publication bias mechanisms, highlighting limitations in its applicability.

Findings

Copas' method is sensitive to different publication bias mechanisms.

It performs poorly under realistic bias scenarios.

The results question the practical usefulness of Copas' method.

Abstract

Copas' method corrects a pooled estimate from an aggregated data meta-analysis for publication bias. Its performance has been studied for one particular mechanism of publication bias. We show through simulations that Copas' method is not robust against other realistic mechanisms. This questions the usefulness of Copas' method, since publication bias mechanisms are typically unknown in practice.