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

Probability imputation revisited for prognostic factor studies

M Schemper1, G Heinze

  • 1Department of Medical Computer Sciences, Vienna University, Austria.

Statistics in Medicine
|January 15, 1997
PubMed
Summary
This summary is machine-generated.

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

Impact of high-volume, intermediate-volume and low-volume bowel preparation on colonoscopy quality and patient satisfaction: An observational study.

United European gastroenterology journal·2019
Same author

The long-term recurrence risk of patients with unprovoked venous thromboembolism: an observational cohort study.

Journal of thrombosis and haemostasis : JTH·2016
Same author

Statistical controversies in clinical research: the importance of importance.

Annals of oncology : official journal of the European Society for Medical Oncology·2016
Same author

Prognostic role of pretreatment plasma fibrinogen in patients with solid tumors: A systematic review and meta-analysis.

Cancer treatment reviews·2015
Same author

The INTERPRET-DD study of diabetes and depression: a protocol.

Diabetic medicine : a journal of the British Diabetic Association·2015
Same author

Parameters influencing FVIII pharmacokinetics in patients with severe and moderate haemophilia A.

Haemophilia : the official journal of the World Federation of Hemophilia·2015
Same journal

Interpretable Bayesian Modeling for Multireader Multicase Studies: Addressing Overdispersion and Limited Sample Size in Diagnostic Enhancement Evaluation.

Statistics in medicine·2026
Same journal

Adaptive Sequential Multiple Hypotheses Testing for Concomitant Vaccine Safety Surveillance.

Statistics in medicine·2026
Same journal

Novel Distance Regression for Repeated Outcomes With Missing Data: Applications to Longitudinal and Crossover Studies of Microbiome Beta-Diversity.

Statistics in medicine·2026
Same journal

Optimal Weighted Tests for Replication Studies and the 'Two-Trials Rule' With Multiple Hypotheses.

Statistics in medicine·2026
Same journal

Identifiable Copula-Double-Cox Models: A Fully Parametric Framework for Dependent Right-Censored Survival Data.

Statistics in medicine·2026
Same journal

Moving From Individualized Risk-Based Prevention to Benefit-Based Prevention: Estimating Individualized Life-Years Gained From Prevention Services as a Basis for Eligibility.

Statistics in medicine·2026
See all related articles

Missing covariate values in prognostic factor studies can be handled using conditional probability imputation (PIT). While not universally recommended, PIT remains a satisfactory and practical method for many studies, comparable to more complex techniques.

Area of Science:

  • Biostatistics
  • Epidemiology
  • Medical Informatics

Background:

  • Missing covariate values frequently hinder prognostic factor studies using regression models.
  • Conditional probability imputation (PIT) was proposed in 1990 as an efficient method for handling missing data.
  • Recent studies suggest model-based methods may be superior to PIT, questioning its universal applicability.

Purpose of the Study:

  • To evaluate the continued appropriateness and performance of conditional probability imputation (PIT) in prognostic factor studies.
  • To compare PIT with multiple imputation in the context of prognostic factor analysis.
  • To discuss practical aspects of PIT application, including comparability of marginal and partial effects.

Main Methods:

  • Review and empirical evaluation of conditional probability imputation (PIT) for handling missing covariate data.

Related Experiment Videos

  • Comparison of PIT performance against complete case analysis, omission of covariates, and multiple imputation.
  • Application and discussion of PIT using a prostate cancer dataset.

    • Main Results:

    • Conditional probability imputation (PIT) demonstrates satisfactory performance in typical prognostic factor studies.
    • PIT outperforms complete case and omission of covariates strategies.
    • Comparisons with multiple imputation did not reveal a significant advantage for the more complex technique.

    Conclusions:

    • Conditional probability imputation (PIT) remains an appropriate and attractive method for analyzing prognostic factor studies.
    • PIT allows for direct comparison of marginal and partial effects analyses.
    • Despite some limitations, PIT offers a practical and effective solution for missing data in this context.