Odds are the sign is right

TL;DR

This paper proposes a new sensitivity analysis condition based on odds ratios to evaluate the impact of unmeasured confounders, demonstrating its reliability through simulations and an application to Zika virus data.

Contribution

It introduces an odds ratio-based condition for sensitivity analysis that is applicable to confounding variables and compares favorably with existing methods.

Findings

The odds ratio condition performs reliably in simulations.

It is applicable when the unmeasured covariate is a confounder.

The method is demonstrated with Zika virus and birth defects data.

Abstract

This article introduces a new condition based on odds ratios for sensitivity analysis. The analysis involves the average effect of a treatment or exposure on a response or outcome with estimates adjusted for and conditional on a single, unmeasured, dichotomous covariate. Results of statistical simulations are displayed to show that the odds ratio condition is as reliable as other commonly used conditions for sensitivity analysis. Other conditions utilize quantities reflective of a mediating covariate. The odds ratio condition can be applied when the covariate is a confounding variable. As an example application we use the odds ratio condition to analyze and interpret a positive association observed between Zika virus infection and birth defects.