Study objective-Prevent is a public health model for estimating the effect on mortality of changes in exposure to risk factors.
When the model is tested by simulating a development that has already taken place, the results may differ considerably from the actual situation.
The purpose of this study is to test the Prevent model by applying it to a synthetic cohort in which the development is unaffected by concealed factors.
Design-A micro-simulation model was developed to create basic data for Prevent and give « exact » results as to changes in risk factor prevalences and mortality.
The estimates of Prevent simulations were compared with the « exact » results.
Main results-Modelling one risk factor related to a cause specific mortality gave fairly similar results by the two methods.
Including two risk factors Prevent tends slightly to overestimate the health benefits of prevention.
Conclusions-The differences between the « exact » mortality and the Prevent estimates will be small for realistic scenarios and Prevent provide reasonable estimates of the health benefits of prevention.
Mots-clés Pascal : Prévention, Action, Analyse statistique, Mortalité, Facteur risque, Evolution, Epidémiologie, Méthodologie, Evaluation, Modèle, Homme
Mots-clés Pascal anglais : Prevention, Action, Statistical analysis, Mortality, Risk factor, Evolution, Epidemiology, Methodology, Evaluation, Models, Human
Notice produite par :
Inist-CNRS - Institut de l'Information Scientifique et Technique
Cote : 99-0227661
Code Inist : 002B30A01A1. Création : 16/11/1999.