P24 Modelling individual-level uncertainty due to missing data in personalised breast cancer risk prediction

Le résumé fourni par la source

Multifactorial breast cancer risk prediction models use a range of predictors to estimate an individual’s chance of developing breast cancer. These can facilitate risk stratification, to identify individuals who are most likely to benefit from screening or preventive options. The BOADICEA breast cancer risk model, implemented in the CanRisk tool (www.canrisk.org), uses genetic, lifestyle, hormonal, family history and anthropometric data to estimate an individual’s breast cancer risk. Risk predictor data are often incomplete, and point estimates calculated when some risk factor data are missing can mask considerable uncertainty. Quantifying this uncertainty is critical for effective risk communication. We used Monte Carlo simulation methods to estimate the distribution of 10-year breast cancer risk for individuals with missing data. Imputation by chained equations with large representative reference datasets was used to sample missing covariates. We developed a framework for estimating the uncertainty distribution, uncertainty intervals, and probability of reclassification, that can be applied to any given individual with missing risk factor data. Results demonstrate that there may be a considerable probability of reclassification after collecting missing data. Methodology presented here can identify situations where it would be most beneficial to collect additional patient information, and enable better informed clinical decision-making.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.