Studentized sensitivity analysis for the sample average treatment effect in paired observational studies

TL;DR

This paper introduces a studentized sensitivity analysis method for paired observational studies that accounts for effect heterogeneity, improving robustness of causal inferences against unmeasured confounding.

Contribution

It extends permutation-based sensitivity analysis to handle effect heterogeneity and unknown treatment assignment probabilities in paired observational studies.

Findings

The method effectively assesses sensitivity to unmeasured confounding with heterogeneous effects.

It demonstrates that concerns about constant effect assumptions are often overstated.

The approach generalizes traditional tests to more realistic observational study scenarios.

Abstract

A fundamental limitation of causal inference in observational studies is that perceived evidence for an effect might instead be explained by factors not accounted for in the primary analysis. Methods for assessing the sensitivity of a study's conclusions to unmeasured confounding have been established under the assumption that the treatment effect is constant across all individuals. In the potential presence of unmeasured confounding, it has been argued that certain patterns of effect heterogeneity may conspire with unobserved covariates to render the performed sensitivity analysis inadequate. We present a new method for conducting a sensitivity analysis for the sample average treatment effect in the presence of effect heterogeneity in paired observational studies. Our recommended procedure, called the studentized sensitivity analysis, represents an extension of recent work on studentized permutation tests to the case of observational studies, where randomizations are no longer drawn uniformly. The method naturally extends conventional tests for the sample average treatment effect in paired experiments to the case of unknown, but bounded, probabilities of assignment to treatment. In so doing, we illustrate that concerns about certain sensitivity analyses operating under the presumption of constant effects are largely unwarranted.