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

Validation of a melanoma prognostic model

D J Margolis1, A C Halpern, T Rebbeck

  • 1Department of Dermatology, University of Pennsylvania School of Medicine, Philadelphia, USA. Margolis@cceb.med.upenn.edu

Archives of Dermatology
|January 6, 1999
PubMed
Summary

A 4-variable melanoma prognostic model showed good external validity in new patient groups. A simpler 1-variable model using tumor thickness offers similar accuracy for melanoma prognosis.

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

3D Whole-body skin imaging for automated melanoma detection.

Journal of the European Academy of Dermatology and Venereology : JEADV·2023
Same author

The long-term evolution of melanocytic nevi among high-risk adults.

Journal of the European Academy of Dermatology and Venereology : JEADV·2022
Same author

Atopic dermatitis: filaggrin and skin barrier dysfunction.

The British journal of dermatology·2022
Same author

Skin cancer risk among testicular germ-cell cancer survivors: a systematic review and meta-analysis.

Journal of the European Academy of Dermatology and Venereology : JEADV·2022
Same author

Accuracy of commercially available smartphone applications for the detection of melanoma.

The British journal of dermatology·2021
Same author

Human leukocyte antigen class-I variation is associated with atopic dermatitis: A case-control study.

Human immunology·2021

Area of Science:

  • Oncology
  • Biostatistics
  • Epidemiology

Background:

  • A previously published 4-variable prognostic model for melanoma utilized clinically accessible data (patient age, sex, tumor location, thickness).
  • Prognostic models require validation in independent populations to confirm their generalizability and heuristic value.
  • External validation is crucial for assessing the reliability of predictive models in diverse clinical settings.

Purpose of the Study:

  • To perform an external validation of a 4-variable melanoma prognostic model.
  • To compare the prognostic accuracy of the 4-variable model against a simplified 1-variable model (tumor thickness).
  • To assess the generalizability of these melanoma prognostic models.

Main Methods:

  • Utilized a population-based cohort of individuals diagnosed with melanoma for external validation.

Related Experiment Videos

  • Evaluated both a 4-variable model and a 1-variable model (tumor thickness) using logistic regression.
  • Assessed model performance using the c statistic and Brier score to measure predictive accuracy.

    • Main Results:

    • The 4-variable model demonstrated strong external validity with c statistics of 0.86 and 0.81 in two independent datasets.
    • The 1-variable model (tumor thickness) also showed robust performance, with c statistics of 0.83 and 0.79.
    • Brier scores indicated comparable accuracy between the 4-variable and 1-variable models across both datasets, with no significant differences.

    Conclusions:

    • Both the 4-variable and 1-variable melanoma prognostic models are generalizable to different populations.
    • The 1-variable model, focusing solely on tumor thickness, provides a simpler yet highly accurate alternative for melanoma prognosis.
    • Tumor thickness alone is a powerful predictor of melanoma outcomes, offering a practical tool for clinical use with minimal loss of accuracy.