Learning Causally Predictable Outcomes from Psychiatric Longitudinal Data

2025-06-19

AI Summary

In psychiatric longitudinal data, symptom heterogeneity and latent-variable confounding frequently invalidate conventional causal effect estimation—particularly because fixed outcome definitions often violate the no-unmeasured-confounding assumption. To address this, we propose a dynamic outcome optimization framework that learns non-negative, clinically interpretable symptom-weighted aggregations to jointly maximize treatment effect persistence and backdoor invariance, accompanied by an empirical test for outcome-level unconfoundedness. Our method integrates time-limited early direct-effect modeling, non-negative weight optimization, and causal robustness regularization to mitigate latent confounding. Evaluated on multi-center cohorts of major depressive disorder and schizophrenia, the framework consistently outperforms state-of-the-art approaches, yielding stable, clinically meaningful composite-outcome causal effects with improved identifiability and interpretability.

Technology Category

• Machine Learning: Causal Learning
• Search and Optimization: Mixed Discrete/Continuous Search
• Constraint Satisfaction and Optimization: Constraint Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• User Modeling, Personalization and Recommendation: Studies of user behavior, including longitudinal effects of personalized systems

Abstract

Causal inference in longitudinal biomedical data remains a central challenge, especially in psychiatry, where symptom heterogeneity and latent confounding frequently undermine classical estimators. Most existing methods for treatment effect estimation presuppose a fixed outcome variable and address confounding through observed covariate adjustment. However, the assumption of unconfoundedness may not hold for a fixed outcome in practice. To address this foundational limitation, we directly optimize the outcome definition to maximize causal identifiability. Our DEBIAS (Durable Effects with Backdoor-Invariant Aggregated Symptoms) algorithm learns non-negative, clinically interpretable weights for outcome aggregation, maximizing durable treatment effects and empirically minimizing both observed and latent confounding by leveraging the time-limited direct effects of prior treatments in psychiatric longitudinal data. The algorithm also furnishes an empirically verifiable test for outcome unconfoundedness. DEBIAS consistently outperforms state-of-the-art methods in recovering causal effects for clinically interpretable composite outcomes across comprehensive experiments in depression and schizophrenia.

Problem

• Improving causal inference in psychiatric longitudinal data
• Optimizing outcome definition for better causal identifiability
• Addressing observed and latent confounding in treatment effects

Innovation

• Optimizes outcome definition for causal identifiability
• Learns interpretable weights to minimize confounding
• Leverages time-limited effects in psychiatric data