Evaluation of Methods Adjusting for Unmeasured Confounding Using Large Healthcare Databases: An Empirical Study Concerning Drugs Inducing Prematurity
Clinical and Translational Science · 2025
DOI: 10.1111/cts.70394 ↗Kurz gefasst
Fachartikel in der Zeitschrift Clinical and Translational Science, 2025. Verfasst von Chi-Hong Duong und 7 weiteren Autor:innen. Beteiligte Einrichtung: High-Dimensional Biostatistics for Drug Safety and Genomics, CESP, Université Paris-Saclay, UVSQ, Université Paris-Sud, Inserm, Villejuif, France. Hauptthemen laut MEDLINE: Databases, Factual, Pharmacoepidemiology, Premature Birth. Die Arbeit nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).
Abstract (Originalsprache)
With the growing availability of large healthcare databases for clinical science, mitigating unmeasured confounding has emerged as a major issue in pharmacoepidemiologic studies. Extensions of causal inference methods to high‐dimensional settings could help address this problem, but studies comparing their performance in real‐world databases are still lacking. This study aims to compare the ability to reduce the measured and indirectly measured confounding of three causal inference methods adapted to a real‐world high‐dimensional database using a machine learning LASSO algorithm: G‐computation (GC), Targeted Maximum Likelihood estimation (TMLE) and Propensity Score with overlap or stabilized inverse probability treatment weighting. This large‐scale empirical study was based on the French National Healthcare Claims Database (SNDS), consisting of 2,172,702 pregnancies 22 weeks of gestation over the period 2011–2014. We used a set of 42 negative and 13 positive reference drugs related to prematurity risk. For each reference drug, the logarithm of the odds ratio for prematurity and its 95% confidence interval were estimated using each method. The proportions of false positive and true positive associations were calculated and compared between the methods. All methods yielded fewer false positives than a crude model based on a minimal set of adjusted covariates. TMLE produced the lowest proportion of false positives (45.2%), followed by GC (47.6%). GC yielded the highest proportion of true positives (92.3%). Our results confirm the interest of causal inference methods exploiting the wealth of data in healthcare databases, especially GC in terms of performance and ease of implementation.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- High-Dimensional Biostatistics for Drug Safety and Genomics, CESP, Université Paris-Saclay, UVSQ, Université Paris-Sud, Inserm, Villejuif, France.
- Faculty of Medicine and Midwifery, ETHICS EA 7446 Lille Catholic University, Lille, France.
- Biostatistics and Bioinformatics (DIM), Dijon University Hospital, Dijon, France.
- Faculty of Medicine, University of Bourgogne-Franche-Comté, Dijon, France.
- Clinical Epidemiology/Clinical Trials Unit, Clinical Investigation Center 1432, Dijon, France.
Zitieren
APA
Duong, C. H., Escolano, S., Demailly, R., Thiebaut, A., Cottenet, J., Quantin, C., Tubert-Bitter, P., & Ahmed, I. (2025). Evaluation of Methods Adjusting for Unmeasured Confounding Using Large Healthcare Databases: An Empirical Study Concerning Drugs Inducing Prematurity. Clinical and Translational Science. https://doi.org/10.1111/cts.70394
Vancouver
Duong CH, Escolano S, Demailly R, Thiebaut A, Cottenet J, Quantin C, et al. Evaluation of Methods Adjusting for Unmeasured Confounding Using Large Healthcare Databases: An Empirical Study Concerning Drugs Inducing Prematurity. Clinical and Translational Science. 2025. doi:10.1111/cts.70394
BibTeX
@article{Duong2025,
author = {Duong, Chi-Hong and Escolano, Sylvie and Demailly, Romain and Thiebaut, Anne and Cottenet, Jonathan and Quantin, Catherine and Tubert-Bitter, Pascale and Ahmed, Ismaïl},
title = {Evaluation of Methods Adjusting for Unmeasured Confounding Using Large Healthcare Databases: An Empirical Study Concerning Drugs Inducing Prematurity},
journal = {Clinical and Translational Science},
year = {2025},
doi = {10.1111/cts.70394}
}
EUVIMED