Transfer Learning for Meta-analysis Under Covariate Shift

TL;DR

This paper introduces a covariate shift-aware transfer learning framework for meta-analysis, leveraging placebo data as high-fidelity labels to improve treatment effect estimation across diverse populations.

Contribution

It proposes a novel placebo-anchored transport method that calibrates baseline risk and estimates heterogeneous treatment effects under covariate shift, with theoretical and empirical validation.

Findings

Outperforms baseline methods in synthetic and semi-synthetic experiments.

Improves treatment effect estimation accuracy at small target sample sizes.

Maintains strong ranking performance even when target data is limited.

Abstract

Randomized controlled trials often do not represent the populations where decisions are made, and covariate shift across studies can invalidate standard IPD meta-analysis and transport estimators. We propose a placebo-anchored transport framework that treats source-trial outcomes as abundant proxy signals and target-trial placebo outcomes as scarce, high-fidelity gold labels to calibrate baseline risk. A low-complexity (sparse) correction anchors proxy outcome models to the target population, and the anchored models are embedded in a cross-fitted doubly robust learner, yielding a Neyman-orthogonal, target-site doubly robust estimator for patient-level heterogeneous treatment effects when target treated outcomes are available. We distinguish two regimes: in connected targets (with a treated arm), the method yields target-identified effect estimates; in disconnected targets (placebo-only), it reduces to a principled screen--then--transport procedure under explicit working-model transport assumptions. Experiments on synthetic data and a semi-synthetic IHDP benchmark evaluate pointwise CATE accuracy, ATE error, ranking quality for targeting, decision-theoretic policy regret, and calibration. Across connected settings, the proposed method is best or near-best and improves substantially over proxy-only, target-only, and transport baselines at small target sample sizes; in disconnected settings, it retains strong ranking performance for targeting while pointwise accuracy depends on the strength of the working transport condition.