EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Combining experimental and observational data through a power likelihood

Xi Lin iD, Jens Magelund Tarp, Robin J Evans

Biometrics · 2025

DOI: 10.1093/biomtc/ujaf008 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Biometrics, 2025. Verfasst von Xi Lin und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Statistics, University of Oxford, Oxford, OX1 3LB, United Kingdom. Hauptthemen laut MEDLINE: Observational Studies as Topic, Randomized Controlled Trials as Topic, Biometry. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 32 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Randomized controlled trials are the gold standard for causal inference and play a pivotal role in modern evidence-based medicine. However, the sample sizes they use are often too limited to provide adequate power for drawing causal conclusions. In contrast, observational data are becoming increasingly accessible in large volumes but can be subject to bias as a result of hidden confounding. Given these complementary features, we propose a power likelihood approach to augmenting randomized controlled trials with observational data to improve the efficiency of treatment effect estimation. We provide a data-adaptive procedure for maximizing the expected log predictive density (ELPD) to select the learning rate that best regulates the information from the observational data. We validate our method through a simulation study that shows increased power while maintaining an approximate nominal coverage rate. Finally, we apply our method in a real-world data fusion study augmenting the PIONEER 6 clinical trial with a US health claims dataset, demonstrating the effectiveness of our method and providing detailed guidance on how to address practical considerations in its application.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Observational Studies as TopicRandomized Controlled Trials as TopicBiometryHumansLikelihood FunctionsComputer SimulationSample SizeCausalityData Interpretation, StatisticalModels, Statistical

Stichwörter der Autor:innen

Bayesian analysiscausal inferenceclinical trialsdata fusionefficiency gainexternal controls

Autor:innen und Institutionen

  • Department of Statistics, University of Oxford, Oxford, OX1 3LB, United Kingdom.
  • Innovative Medical Evidence Generation, Novo Nordisk, Søborg 2860, Denmark.

ORCID-Profile: Xi Lin

Zitieren

APA

Lin, X., Tarp, J. M., & Evans, R. J. (2025). Combining experimental and observational data through a power likelihood. Biometrics. https://doi.org/10.1093/biomtc/ujaf008

Vancouver

Lin X, Tarp JM, Evans RJ. Combining experimental and observational data through a power likelihood. Biometrics. 2025. doi:10.1093/biomtc/ujaf008

BibTeX
@article{Lin2025,
  author = {Lin, Xi and Tarp, Jens Magelund and Evans, Robin J},
  title = {Combining experimental and observational data through a power likelihood},
  journal = {Biometrics},
  year = {2025},
  doi = {10.1093/biomtc/ujaf008}
}
Kostenlos starten →