Optimizing precision in stepped-wedge designs via machine learning and quadratic inference functions

2026-02-10

AI Summary

This study addresses the limitations of existing analytical methods for stepped wedge designs, which rely on parametric models and pre-specified correlation structures, thereby constraining estimation accuracy. To overcome these constraints, the authors propose a novel approach that integrates machine learning with quadratic inference functions (QIF), enabling flexible covariate adjustment and adaptive learning of the correlation structure to enhance the efficiency of causal average treatment effect estimation. The method accommodates treatment effect heterogeneity and retains asymptotic optimality even under model misspecification, while guaranteeing L₂ convergence and asymptotic normality. Extensive simulations and two empirical applications demonstrate that the proposed estimator consistently achieves efficiency no worse than that of the independence working correlation structure and attains the minimum possible asymptotic variance.

Technology Category

• Machine Learning: Causal Learning
• Reasoning under Uncertainty: Causality
• Intelligent Robots: State Estimation

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

Stepped-wedge designs are increasingly used in randomized experiments to accommodate logistical and ethical constraints by staggering treatment roll-out over time. Despite their popularity, existing analytical methods largely rely on parametric models with linear covariate adjustment and prespecified correlation structures, which may limit achievable precision in practice. We propose a new class of estimators for the causal average treatment effect in stepped-wedge designs that optimizes precision through flexible, machine-learning-based covariate adjustment to capture complex outcome-covariate relationships, together with quadratic inference functions to adaptively learn the correlation structure. We establish consistency and asymptotic normality under mild conditions requiring only $L_2$ convergence of nuisance estimators, even under model misspecification, and characterize when the estimator attains the minimal asymptotic variance. Moreover, we prove that the proposed estimator never reduces efficiency relative to an independence working correlation. The proposed method further accommodates treatment-effect heterogeneity across both exposure duration and calendar time. Finally, we demonstrate our methods through simulation studies and reanalyses of two empirical studies that differ substantially in research area and key design parameters.

Problem

• stepped-wedge design
• precision optimization
• covariate adjustment
• correlation structure
• causal inference

Innovation

• stepped-wedge design
• machine learning
• quadratic inference functions
• covariate adjustment
• treatment-effect heterogeneity