• Who we are
    • About us
    • Our values
    • Environmental, social & governance
    • Therapeutic areas
  • What we do
    • Consulting (Acsel Health)
    • HEOR & market access
    • Scientific communications
    • Patient engagement
  • Insights
  • News & Events
  • Join us
    • Careers
    • Reasons to join
  • Contact us
  • Menu Menu

Publication Library / Publications

A simulation study of sample size demonstrated the importance of the number of events per variable to develop prediction models in clustered data

Objective

This study aims to investigate the influence of the amount of clustering [intraclass correlation (ICC) = 0%, 5%, or 20%], the number of events per variable (EPV) or candidate predictor (EPV = 5, 10, 20, or 50), and backward variable selection on the performance of prediction models.

Study design and setting

Researchers frequently combine data from several centers to develop clinical prediction models. In our simulation study, we developed models from clustered training data using multilevel logistic regression and validated them in external data.

Conclusion

We recommend at least 10 EPV to fit prediction models in clustered data using logistic regression. Up to 50 EPV may be needed when variable selection is performed.

Authors L Wynants, W Bouwmeester, K G Moons, M Moerbeek, D Timmerman, S Van Huffel, B Van Calster, Y Vergouwe
Journal Journal of Clinical Epidemiology
Therapeutic Area Other
Center of Excellence Real-world Evidence & Data Analytics
Year 2015
Read full article

Services

  • Consulting
  • HEOR & market access
  • Scientific communications
  • Creative communications
  • Patient engagement

Company

  • About Us
  • Our values
  • Environmental, social & governance
  • Our commitment to rare disease
  • Careers
  • Reasons to join
  • News & insights
  • Events
  • Locations & contact

Legal and Governance

  • Terms of use
  • Privacy notice
  • Cookie policy
  • IT security measures
  • Modern slavery statement
  • Disclosure UK – ABPI
  • Looking for OpenHealth Company?
  • Legal statements & documents
  • Global ethical business conduct code
  • Suppliers
footer-logo-mark
  • Twitter
  • Linkedin
  • Instagram
  • Facebook

© Copyright OPEN Health 2025. All rights reserved. OPEN Health is a registered trademark.

backtotop-arrow
Scroll to top