Is Bootstrapping Sufficient for Validating a Risk Model for Selection of Participants for a Lung Cancer Screening Program?
Menée à partir de données portant sur 502 321 participants (âge : de 40 à 70 ans ; durée de suivi : 1 469 518 personnes-années), cette étude évalue la performance d'un modèle mathématique, incorporant notamment des variables de facteurs de risque (âge, sexe, statut tabagique, antécédents médicaux, ...) et un indicateur de la capacité respiratoire (volume expiratoire forcé durant la première seconde), pour prédire le risque de développer un cancer du poumon (738 cas)
Risk prediction models are powerful tools that use multivariable regression to combine predictors or predisposing factors to estimate the probability or risk of the presence or future occurrence of clinical outcomes such as lung cancer.Several lung cancer risk prediction models have been developed. Such models are usually constructed in data sets with information from a well-defined population with similar characteristics. Discrimination is a measure of how well a model can separate diseased from nondiseased individuals and is most often measured using the area under the receiver operating characteristic curve or concordance c-statistic, although other methods and metrics of performance of prediction models have been published.
Journal of Clinical Oncology , éditorial en libre accès, 2016