Examining the Association between Estimated Prevalence and Diagnostic Test Accuracy using Directed Acyclic Graphs

TL;DR

This paper uses directed acyclic graphs and simulations to explore how biases in diagnostic studies can create a correlation between disease prevalence estimates and test accuracy, highlighting the importance of bias assessment.

Contribution

It systematically introduces DAGs to model bias structures in diagnostic meta-analyses and demonstrates how these biases can produce prevalence-accuracy associations.

Findings

Association arises without a perfect reference test or due to confounding factors.

Proper statistical adjustments can eliminate the observed association.

Bias exploration is crucial for accurate interpretation of diagnostic meta-analyses.

Abstract

There have been reports of correlation between estimates of prevalence and test accuracy across studies included in diagnostic meta-analyses. It has been hypothesized that this unexpected association arises because of certain biases commonly found in diagnostic accuracy studies. A theoretical explanation has not been studied systematically. In this work, we introduce directed acyclic graphs to illustrate common structures of bias in diagnostic test accuracy studies and to define the resulting data-generating mechanism behind a diagnostic meta-analysis. Using simulation studies, we examine how these common biases can produce a correlation between estimates of prevalence and index test accuracy and what factors influence its magnitude and direction. We found that an association arises either in the absence of a perfect reference test or in the presence of a covariate that simultaneously causes spectrum effect and is associated with the prevalence (confounding). We also show that the association between prevalence and accuracy can be removed by appropriate statistical methods. In the risk of bias evaluation in diagnostic meta-analyses, an observed association between estimates of prevalence and accuracy should be explored to understand its source and to adjust for latent or observed variables if possible.