From Sample Average Treatment Effect to Population Average Treatment Effect on the Treated: Combining Experimental with Observational Studies to Estimate Population Treatment Effects

Erin Hartman, Richard Grieve, Roland R. Ramsahai, Jasjeet S. Sekhon

Journal of the Royal Statistical Society Series A (Statistics in Society) · 2015 · 153 citations · 53 references

DOIFull text

Open access

Concepts

TL;DR

RCTs provide unbiased sample average treatment effects but may not yield unbiased population average treatment effects. The study derives the assumptions needed to identify population average treatment effects from RCTs and proposes designs that combine non‑randomized data to adjust RCT estimates. The authors develop placebo tests based on the identifying assumptions and introduce new designs that merge RCT and observational data, illustrated by a cost‑effectiveness analysis of pulmonary artery catheterization. The combined approach was applied to a cost‑effectiveness analysis of pulmonary artery catheterization, demonstrating its feasibility.

Abstract

Summary Randomized controlled trials (RCTs) can provide unbiased estimates of sample average treatment effects. However, a common concern is that RCTs may fail to provide unbiased estimates of population average treatment effects. We derive the assumptions that are required to identify population average treatment effects from RCTs. We provide placebo tests, which formally follow from the identifying assumptions and can assess whether they hold. We offer new research designs for estimating population effects that use non-randomized studies to adjust the RCT data. This approach is considered in a cost-effectiveness analysis of a clinical intervention: pulmonary artery catheterization.

References

53