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

Statistical approaches to development and validation of predictive instruments

U E Ruttimann1

  • 1Laboratory of Clinical Studies, National Institute on Alcohol Abuse and Alcoholism, National Institutes of Health, Bethesda, MD.

Critical Care Clinics
|January 1, 1994
PubMed
Summary

This study outlines prediction models for intensive care unit (ICU) outcomes, focusing on multivariate logistic regression. It details methods for validating predictors and assessing model accuracy using receiver operating characteristic (ROC) analysis.

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

A pyramid approach to subpixel registration based on intensity.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2008
Same author

Certainty and mortality prediction in critically ill children.

Journal of medical ethics·2004
Same author

Relevance of diagnostic diversity and patient volumes for quality and length of stay in pediatric intensive care units.

Pediatric critical care medicine : a journal of the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies·2003
Same author

Long-stay patients in the pediatric intensive care unit.

Critical care medicine·2001
Same author

The impact of prematurity: a perspective of pediatric intensive care units.

Critical care medicine·2001
Same author

Comparison of three techniques for internal jugular vein cannulation in infants.

Paediatric anaesthesia·2000

Area of Science:

  • Medical Informatics
  • Biostatistics
  • Critical Care Medicine

Background:

  • Developing accurate prediction models for intensive care unit (ICU) outcomes is crucial for patient management.
  • Multivariate logistic regression is a common statistical technique employed in clinical prediction modeling.
  • Validating predictors and assessing model performance are essential steps in ensuring reliable predictions.

Purpose of the Study:

  • To outline concepts for developing prediction models for ICU outcomes.
  • To emphasize the application of multivariate logistic regression in this context.
  • To review methods for predictor validation and model performance assessment.

Main Methods:

  • Discussion of experimental study designs for predictor validation.

Related Experiment Videos

  • Explanation of statistical procedures for assessing agreement between predicted and observed mortality risks.
  • Detailed review of receiver operating characteristic (ROC) analysis for evaluating predictor accuracy.

    • Main Results:

    • Multivariate logistic regression is presented as a key method for ICU outcome prediction.
    • Two statistical procedures are discussed for evaluating the concordance of predicted vs. observed mortality.
    • ROC analysis is highlighted as a primary tool for graphical and statistical comparison of predictor accuracy.

    Conclusions:

    • The study provides a framework for developing and validating ICU outcome prediction models.
    • Effective use of statistical methods, including logistic regression and ROC analysis, is vital for accurate patient classification.
    • These methods aid in distinguishing between survivors and nonsurvivors in critical care settings.