Frameworks for Estimating Causal Effects in Observational Settings: Comparing Confounder Adjustment and Instrumental Variables

TL;DR

This paper compares confounder adjustment and instrumental variable methods for estimating causal effects in observational health studies, emphasizing principles for when assumptions are violated and demonstrating with a case study on donepezil for MCI.

Contribution

It formalizes general principles and heuristics for causal inference using confounders and IVs, especially under assumption violations, and applies these to a real-world case study.

Findings

Confounder and IV methods yield different estimates in the MCI case study.

Flexible methods like TMLE and DML can adapt to complex settings.

Comparison highlights the strengths and limitations of each approach.

Abstract

To estimate causal effects, analysts performing observational studies in health settings utilize several strategies to mitigate bias due to confounding by indication. There are two broad classes of approaches for these purposes: use of confounders and instrumental variables (IVs). Because such approaches are largely characterized by untestable assumptions, analysts must operate under an indefinite paradigm that these methods will work imperfectly. In this tutorial, we formalize a set of general principles and heuristics for estimating causal effects in the two approaches when the assumptions are potentially violated. This crucially requires reframing the process of observational studies as hypothesizing potential scenarios where the estimates from one approach are less inconsistent than the other. While most of our discussion of methodology centers around the linear setting, we touch upon complexities in non-linear settings and flexible procedures such as target minimum loss-based estimation (TMLE) and double machine learning (DML). To demonstrate the application of our principles, we investigate the use of donepezil off-label for mild cognitive impairment (MCI). We compare and contrast results from confounder and IV methods, traditional and flexible, within our analysis and to a similar observational study and clinical trial.