Interpretable machine learning for rare events: an inverse probability weighted application to understanding glucagon-like peptide-1 receptor agonists use and incident alcohol-related disorder
Alcohol and Alcoholism · 13.05.2026 · 61 (4)
DOI: 10.1093/alcalc/agag045 ↗Kurz gefasst
Fachartikel in der Zeitschrift Alcohol and Alcoholism, online erschienen am 20.06.2026. Verfasst von Rinoj Gautam und 4 weiteren Autor:innen. Beteiligte Einrichtung: College of Pharmacy, UNT Health, Fort Worth, 3500 Camp Bowie Blvd, Fort Worth, TX 76107, United States. Hauptthemen laut MEDLINE: Glucagon-Like Peptide-1 Receptor Agonists, Machine Learning, Diabetes Mellitus, Type 2, Alcohol-Related Disorders. Die Arbeit nennt 41 Literaturangaben. Gefördert von: National Institute on Minority Health and Health Disparities through the Texas Center for Health Disparities (NIMHD), National Institutes of Health/Artificial Intelligence/Machine Learning Consortium to Advance Health Equity and Researcher Diversity.
Abstract (Originalsprache)
Objective
Large administrative datasets enable investigation of rare outcomes, but accurate individual-level prediction remains challenging. Emerging evidence suggests that glucagon-like peptide-1 receptor agonists (GLP-1 RAs), used to treat diabetes and obesity, may reduce alcohol craving and intake. We examined predictors of incident alcohol-related disorder (ARD) using inverse probability–weighted (IPW) machine learning models, focusing on GLP-1 RA use.
Methods
We conducted a retrospective cohort study using the MarketScan Multi-State Medicaid Database (N = 96 460). Type 2 diabetic adults aged 18–64 years with GLP-1 RA prescription coverage (2022–2023) and no ARD in 2022 were included. Baseline variables included demographics, social risk factors (Z-codes), comorbidities, insurance characteristics, and GLP-1 RA use. Propensity scores were used to derive stabilized IPW weights to adjust for confounding in GLP-1 RA exposure. An Extreme Gradient Boosting model with IPW was trained using a 70/30 train–test split with five-fold cross-validation and probability calibration. Performance was evaluated using Area Under the Receiver Operating Characteristics Curve (AUROC), recall, and precision. Model interpretability was assessed using SHapley Additive exPlanations (SHAP).
Results
GLP-1 RAs were used by 21.7% of individuals; 2.1% (N = 2004) developed incident ARD. The model demonstrated modest discrimination (AUROC = 0.67). Recall and precision were 0.28 and 0.05, respectively. SHAP analysis ranked GLP-1 RA use as the sixth most influential predictor and showed an association with lower predicted ARD risk.
Conclusion
Predicting rare outcomes remains difficult, particularly for identifying positive cases. Interpretable machine learning identified GLP-1 RA use as an important feature associated with lower predicted ARD risk, warranting further investigation.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- College of Pharmacy, UNT Health, Fort Worth, 3500 Camp Bowie Blvd, Fort Worth, TX 76107, United States.
- Department of Emergency Medicine, JPS Health Network, 1500 S. Main St., Fort Worth, TX 76104, United States.
- Health Outcomes Research and AI Innovation, Department of Pharmacy Practice, School of Pharmacy, Temple University, 1801 N. Broad St., Philadelphia, PA 19140, United States.
ORCID-Profile: Hao Wang
Förderung
- National Institute on Minority Health and Health Disparities through the Texas Center for Health Disparities (NIMHD)
- National Institutes of Health/Artificial Intelligence/Machine Learning Consortium to Advance Health Equity and Researcher Diversity
Zitieren
APA
Gautam, R., Wang, H., Zhou, B., Neba, R., & Sambamoorthi, U. (2026). Interpretable machine learning for rare events: an inverse probability weighted application to understanding glucagon-like peptide-1 receptor agonists use and incident alcohol-related disorder. Alcohol and Alcoholism, 61(4). https://doi.org/10.1093/alcalc/agag045
Vancouver
Gautam R, Wang H, Zhou B, Neba R, Sambamoorthi U. Interpretable machine learning for rare events: an inverse probability weighted application to understanding glucagon-like peptide-1 receptor agonists use and incident alcohol-related disorder. Alcohol and Alcoholism. 2026;61(4). doi:10.1093/alcalc/agag045
BibTeX
@article{Gautam2026,
author = {Gautam, Rinoj and Wang, Hao and Zhou, Bo and Neba, Rolake and Sambamoorthi, Usha},
title = {Interpretable machine learning for rare events: an inverse probability weighted application to understanding glucagon-like peptide-1 receptor agonists use and incident alcohol-related disorder},
journal = {Alcohol and Alcoholism},
year = {2026},
volume = {61},
number = {4},
doi = {10.1093/alcalc/agag045}
}
EUVIMED