EUVIMEDEUVIMED entdecken
Open Access

Causal Inference Methods for Combining Randomized Trials and Observational Studies: A Review

Bénédicte Colnet, Imke Mayer, Guanhua Chen, Awa Dieng, Ruohong Li, Gaël Varoquaux, Jean-Philippe Vert, Julie Josse, Shu Yang

Statistical Science · 01.02.2024 · 39 (1) · 165-191

DOI: 10.1214/23-sts889 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Statistical Science, erschienen am 01.02.2024. Verfasst von Bénédicte Colnet und 8 weiteren Autor:innen. Beteiligte Einrichtung: INRIA Saclay, Palaiseau, France. Die Arbeit wurde bisher 94-mal zitiert (Crossref) und nennt 169 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NIA NIH HHS, FDA HHS.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

With increasing data availability, causal effects can be evaluated across different data sets, both randomized controlled trials (RCTs) and observational studies. RCTs isolate the effect of the treatment from that of unwanted (confounding) co-occurring effects but they may suffer from unrepresentativeness, and thus lack external validity. On the other hand, large observational samples are often more representative of the target population but can conflate confounding effects with the treatment of interest. In this paper, we review the growing literature on methods for causal inference on combined RCTs and observational studies, striving for the best of both worlds. We first discuss identification and estimation methods that improve generalizability of RCTs using the representativeness of observational data. Classical estimators include weighting, difference between conditional outcome models and doubly robust estimators. We then discuss methods that combine RCTs and observational data to either ensure unconfoundedness of the observational analysis or to improve (conditional) average treatment effect estimation. We also connect and contrast works developed in both the potential outcomes literature and the structural causal model literature. Finally, we compare the main methods using a simulation study and real world data to analyze the effect of tranexamic acid on the mortality rate in major trauma patients. A review of available codes and new implementations is also provided.

Schlagwörter

Stichwörter der Autor:innen

Causal effect generalizationS-admissibilitydata integrationdouble robustnessheterogeneous datatransportability

Autor:innen und Institutionen

  • INRIA Saclay, Palaiseau, France.
  • Owkin, London, UK.
  • Department of Biostatistics and Medical Informatics, University of Wisconsin-Madison, Madison, WI 53726, USA.
  • Google DeepMind, Montreal, Canada.
  • Microsoft, Kirkland, WA 98033, USA.
  • INRIA Saclay, Paries, France.

Förderung

  • NIA NIH HHS
  • FDA HHS

Zitieren

APA

Colnet, B., Mayer, I., Chen, G., Dieng, A., Li, R., Varoquaux, G., Vert, J. P., Josse, J., & Yang, S. (2024). Causal Inference Methods for Combining Randomized Trials and Observational Studies: A Review. Statistical Science, 39(1), 165-191. https://doi.org/10.1214/23-sts889

Vancouver

Colnet B, Mayer I, Chen G, Dieng A, Li R, Varoquaux G, et al. Causal Inference Methods for Combining Randomized Trials and Observational Studies: A Review. Statistical Science. 2024;39(1):165-191. doi:10.1214/23-sts889

BibTeX
@article{Colnet2024,
  author = {Colnet, Bénédicte and Mayer, Imke and Chen, Guanhua and Dieng, Awa and Li, Ruohong and Varoquaux, Gaël and Vert, Jean-Philippe and Josse, Julie and Yang, Shu},
  title = {Causal Inference Methods for Combining Randomized Trials and Observational Studies: A Review},
  journal = {Statistical Science},
  year = {2024},
  volume = {39},
  number = {1},
  pages = {165--191},
  doi = {10.1214/23-sts889}
}
Kostenlos starten →