Recursive pattitioning is a nonparametric technique that produces a classification tree in which subjects are assigned to mutually exclusive subsets according to a set of predictor variables.
We examined the utility of recursive partitioning as a supplement to logistic regression for the multivariable analysis of data from case-control studies, demonstrating the technique using data from women enrolled in a population-based study of subarachnoid hemorrhage.
The classification tree produced by recursive partitioning consisted of three main risk subgroups : (1) elderly women who had long-standing hypertension and who were not smokers, (2) middle-aged women who were cigarette smokers and frequent binge drinkers, and (3) women in whom risk variables indicate relative estrogen deficiency (i.e., postmenopausal status, no recent exposure to hormone replacement therapy, cigarette smoking).
As a supplemental method, recursive partitioning not only identifies subgroups with varying ricks, but also may uncover interactions between variables that may he overlooked in the traditional application of logistic regression to case-control data.
Mots-clés Pascal : Hémorragie, Sousarachnoïdien, Analyse multivariable, Epidémiologie, Facteur risque, Méthodologie, Homme, Etude cas témoin, Etats Unis, Amérique du Nord, Amérique, Système nerveux pathologie, Encéphale pathologie, Cérébrovasculaire pathologie, Appareil circulatoire pathologie, Vaisseau sanguin pathologie, Système nerveux central pathologie
Mots-clés Pascal anglais : Hemorrhage, Subarachnoid, Multivariate analysis, Epidemiology, Risk factor, Methodology, Human, Case control study, United States, North America, America, Nervous system diseases, Cerebral disorder, Cerebrovascular disease, Cardiovascular disease, Vascular disease, Central nervous system disease
Notice produite par :
Inist-CNRS - Institut de l'Information Scientifique et Technique
Cote : 98-0223888
Code Inist : 002B30A01A1. Création : 11/09/1998.