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

Simplifying a prognostic model: a simulation study based on clinical data.

Gareth Ambler1, Anthony R Brady, Patrick Royston

  • 1Department of Statistical Science, University College, 1-19 Torrington Place, London WC1E 7HB, UK. g.ambler@ucl.ac.uk

Statistics in Medicine
|December 17, 2002
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

Assessment of Adverse Events Using the Therapy-Disability-Neurology (TDN) Grading System in a Cohort of Aneurysmal Subarachnoid Hemorrhage Patients: A Single-Center Retrospective Cohort Study.

Brain sciences·2026
Same author

Patient-reported non-motor outcomes after endovascular thrombectomy and intravenous thrombolysis: an observational study.

European stroke journal·2026
Same author

The use of variable selection in clinical prediction modelling for binary outcomes: a systematic review.

Journal of clinical epidemiology·2026
Same author

Author Response: MRI-Based Prediction of Macrovascular Causes of Intracerebral Hemorrhage: The MACRO Score.

Neurology·2026
Same author

Association of glomerular hyperfiltration with mortality in stroke: an analysis using pooled individual patient data.

European stroke journal·2026
Same author

Systematic review, meta-analysis, and meta-regression of the association of serial leukocyte counts in critically ill patients and mortality.

Annals of intensive care·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
Same journal

A Mixture of Distributed Lag Non-Linear Models to Account for Spatially Heterogeneous Exposure-Lag-Response Associations.

Statistics in medicine·2026
Same journal

Practical Considerations for Gaussian Process Modeling for Causal Inference in Quasi-Experimental Studies With Panel Data.

Statistics in medicine·2026
Same journal

Covariate Adjustment for Wilcoxon Two Sample Statistic and Test.

Statistics in medicine·2026
See all related articles

Simplified prognostic models can perform as well as or better than complex full models. Researchers can use methods like Akaike information criterion to find simpler, effective clinical prediction tools.

Area of Science:

  • Biostatistics
  • Clinical Epidemiology
  • Health Informatics

Background:

  • Prognostic models predict clinical outcomes for diseases.
  • Full models, using all prespecified predictors, are advocated to avoid bias but can be complex.
  • Parsimonious models are often preferred in practice for their simplicity.

Purpose of the Study:

  • To investigate the impact of model simplification on prognostic model performance.
  • To compare different methods for estimating full models and simplifying them.
  • To evaluate if simplified models can retain essential prognostic information.

Main Methods:

  • Compared three methods for estimating full prognostic models, including penalized estimation and Tibshirani's lasso.
  • Evaluated two methods for simplifying full models: backwards elimination and a novel 'stepdown' approach.

Related Experiment Videos

  • Utilized simulation studies with two medical datasets to assess performance measures like mean square error and prognostic classification.

    • Main Results:

    • Simplified prognostic models can achieve performance comparable to, or even exceeding, full models.
    • Methods for estimating full models and subsequent simplification were analyzed.
    • Optimizing the Akaike information criterion (AIC) proved effective for selecting the optimal degree of model simplification.

    Conclusions:

    • Model simplification is a viable strategy for developing effective prognostic tools.
    • Parsimonious prognostic models can maintain high predictive accuracy.
    • The Akaike information criterion is a suitable metric for guiding the simplification process in prognostic model development.