JCMS


Critical Appraisal
Prognostic Models: Evidence-Based
Approach to Predicting Disease Outcome

Critical Appraisal Review covers various topics in epidemiology and related fields that are useful to the practicing physician in understanding and using published studies on skin disease topics. Readers may suggest future topics for this series by writing to the Editorial Office.

We use our clinical experience to inform patients that the prognosis is "guarded" or "good." However, we now have the tools to do better with prognostic models for melanoma and other conditions. If appropriate, we can give our melanoma patients a number that describes their odds of cure. Here is a brief primer to assist the clinician in interpreting studies that claim to provide prognostic models.

Martin Weinstock


Jonathan Kantor and David J. Margolis

Abstract

Background: A thorough understanding of the techniques underlying prognostic modelling, the potential shortcomings of prognostic models, and their applicability to dermatology are important for both clinicians and researchers.

Objective: This article distinguishes between prognostic and explanatory (causal) models, discusses some of the techniques used in developing a prognostic model, and addresses the importance of model generalizability.

Conclusion: Prognostic models may be useful to the clinician, but must be used with care. It is important to critically appraise prognostic models and to assure that they are relevant to the population of interest. The ability to critically appraise prognostic models is predicated on a thorough understanding of the techniques used in their development and evaluation.

Sommaire

Antécédents : Il importe, tant pour les cliniciens que pour les chercheurs, de bien comprendre les techniques qui sous-tendent la modélisation pronostique, les limites possibles des modèles pronostiques et leur applicabilité à la dermatologie.

Objectif : Distinguer entre modèles pronostiques et modèles explicatifs (causals); discuter de certaines des techniques employées pour élaborer un modèle pronostique et évoquer les limites de la généralisation des modèles.

Conclusion : Certes, les modèles pronostiques sont utiles au clinicien, mais il faut les utiliser avec prudence. Il importe d'en faire une évaluation critique et de s'assurer qu'ils conviennent à la population à l'étude. Or, la capacité d'évaluer les modèles pronostiques d'un oeil critique repose sur une compréhension approfondie des techniques qui président à leur élaboration et à leur évaluation.


Received 8/27/98. Accepted for publication 10/16/98.

University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania, USA

Reprint requests: Dr. David J. Margolis, Departments of Dermatology and Biostatistics and Epidemiology, Room 815 Blockley Hall; 423 Guardian Drive, University of Pennsylvania School of Medicine, Philadelphia, PA USA 19104

Full text available in the print edition / Pour le texte intégral veuillez consulter la version imprimée.


JCMS 3(3) Contents