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

Using empirical Bayes methods in biopharmaceutical research.

T A Louis1

  • 1Division of Biostatistics, University of Minnesota, School of Public Health, Minneapolis.

Statistics in Medicine
|June 1, 1991
PubMed
Summary

Empirical Bayes methods enhance biopharmaceutical data analysis by estimating treatment effects and heterogeneity. This approach, using compound sampling models, offers more efficient estimates through shrinkage and aids in ranking, selection, and multiplicity adjustments.

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

Perceived helpfulness of the individual components of a behavioural weight loss program: results from the Hopkins POWER Trial.

Obesity science & practice·2016
Same author

Patient-reported outcomes in the practice-based opportunities for weight reduction (POWER) trial.

Quality of life research : an international journal of quality of life aspects of treatment, care and rehabilitation·2013
Same author

Trends for influenza and pneumonia hospitalization in the older population: age, period, and cohort effects.

Epidemiology and infection·2010
Same author

Geographic prevalence and multilevel determination of community-level factors associated with herpes simplex virus type 2 infection in Chennai, India.

American journal of epidemiology·2008
Same author

State-level adjusted ESRD incident rates: use of observed vs model-predicted category-specific rates.

Kidney international·2006
Same author

National Institutes of Health Consensus Development Conference Statement: adjuvant therapy for breast cancer, November 1-3, 2000.

Journal of the National Cancer Institute·2001

Area of Science:

  • Biostatistics
  • Pharmacometrics
  • Clinical Trial Analysis

Background:

  • Compound sampling models are fundamental to biopharmaceutical data, particularly in multi-centre clinical trials where true treatment effects vary by centre.
  • Observed effects deviate from true effects due to sampling variation, necessitating robust statistical approaches for accurate estimation.

Purpose of the Study:

  • To outline the empirical Bayes approach for analyzing biopharmaceutical data, focusing on its application in clinical trials.
  • To compare parametric and non-parametric prior distributions within the empirical Bayes framework.
  • To discuss the importance of accounting for uncertainty in estimated priors and comparing fixed vs. random effects models.

Main Methods:

  • Utilizes empirical Bayes (variance component) analysis to estimate prior distributions from observed data when the prior is not fully specified.

Related Experiment Videos

  • Employs shrinkage estimation, where clinic-specific effects are adjusted towards a common value for increased efficiency.
  • Covers development and comparison of parametric (e.g., Gaussian) and non-parametric priors.

    • Main Results:

    • Empirical Bayes provides an estimated prior mean representing the typical treatment effect and an estimated prior standard deviation indicating treatment effect heterogeneity.
    • Shrinkage of clinic effects leads to more efficient and stable estimates compared to single-clinic analyses.
    • The model facilitates ranking and selection, multiplicity adjustments, and histogram estimation of clinic-specific effects.

    Conclusions:

    • Empirical Bayes methods offer a structured and efficient framework for analyzing complex biopharmaceutical data, particularly in multi-centre trials.
    • The approach enhances the precision of treatment effect estimates and provides valuable insights into heterogeneity.
    • Key considerations for the use and interpretation of empirical Bayes methods are identified, promoting robust clinical trial analysis.