Files
Abstract
Causal inference methods play a vital role in health sciences research, requiring robust methodologies for treatment evaluation and clinical decision-making. Propensity Score Matching (PSM) and Structural Causal Models (SCMs) are key approaches, both grounded in the potential outcome framework but differing in their foundations — PSM follows Rubin's Causal Model, while SCMs follows the Structural Theory of Causation. Despite their shared objectives, these methods diverge significantly in methodologies and interpretations. We systematically study the methods with a specific clinical application.