Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Causal inference methods for small non-randomized studies: Methods and an application in COVID-19.

Sarah Friedrich1, Tim Friede1

  • 1Department of Medical Statistics, University Medical Center Göttingen, Humboldtallee 32, 37073 Göttingen, Germany.

Contemporary Clinical Trials
|November 14, 2020
PubMed
Summary

Analyzing non-randomized studies is crucial for rapid pandemic response. This study evaluates propensity score methods and doubly robust estimators for analyzing early clinical trial data, offering robust alternatives to traditional approaches.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

HERZCHECK: Early Detection of Subclinical Preheart Failure Using Mobile Cardiac Magnetic Resonance and Telemedicine in Rural and Underressourced Regions.

Circulation. Heart failure·2026
Same author

Design, simulate, refine: simulation-guided clinical trials for accelerated drug development.

Nature reviews. Drug discovery·2026
Same author

Short-Term Functional Trajectories After Surgery in Older Adults: National Patterns of Loss and Recovery in 436,471 Patients.

Annals of surgery open : perspectives of surgical history, education, and clinical approaches·2026
Same author

Availability of mechanical circulatory support (MCS) and hospital survival in ST-segment elevation myocardial infarction related cardiogenic shock (STEMI-CS).

European heart journal. Acute cardiovascular care·2026
Same author

Effects of fasudil on disease spreading in ALS - A MUNIX-based post-hoc analysis of the ROCK-ALS trial.

Neurotherapeutics : the journal of the American Society for Experimental NeuroTherapeutics·2026
Same author

Stratification of children with myocarditis using radiomics signatures in LGE cardiovascular MRI.

Computer methods and programs in biomedicine·2026

Area of Science:

  • Biostatistics
  • Epidemiology
  • Clinical Trial Design

Background:

  • Pandemic research, like for SARS-CoV-2, requires faster development cycles for vaccines and treatments.
  • Early clinical trials often have small sample sizes and limited designs, risking overinterpretation of findings.
  • Non-randomized studies present unique challenges due to treatment-selection bias.

Purpose of the Study:

  • To unify and evaluate alternative analytical approaches for non-randomized studies.
  • To investigate the performance of propensity score (PS) methods and their extensions in small sample settings.
  • To assess the utility of doubly robust estimators in the context of early clinical trial data analysis.

Main Methods:

  • Simulation study evaluating propensity score (PS) methods.
Keywords:
COVID-19Causal inferencePropensity scoreSmall samples

Related Experiment Videos

  • Comparison of standard PS methods with g-computation and doubly robust estimators.
  • Focus on small sample settings characteristic of early-phase clinical trials.

    • Main Results:

    • Propensity score methods can help mitigate treatment-selection bias in non-randomized studies.
    • Doubly robust estimators offer advantages in specific analytical scenarios.
    • The performance of these methods was assessed in simulated small sample settings.

    Conclusions:

    • Propensity score-based methods and doubly robust estimators provide valuable tools for analyzing non-randomized studies in pandemic situations.
    • These methods can improve the reliability of findings from early clinical trials.
    • The study provides R code for implementing these simulation-based analyses.