Quantifying and comparing dynamic predictive accuracy of joint models for longitudinal marker and time-to-event in presence of censoring and competing risks

TY - JOUR

T1 - Quantifying and comparing dynamic predictive accuracy of joint models for longitudinal marker and time-to-event in presence of censoring and competing risks

AU - Blanche, Paul

AU - Proust-Lima, Cécile

AU - Loubère, Lucie

AU - Berr, Claudine

AU - Dartigues, Jean-François

AU - Jacqmin-Gadda, Hélène

N1 - © 2014, The International Biometric Society.

PY - 2015/3/1

Y1 - 2015/3/1

N2 - Thanks to the growing interest in personalized medicine, joint modeling of longitudinal marker and time-to-event data has recently started to be used to derive dynamic individual risk predictions. Individual predictions are called dynamic because they are updated when information on the subject's health profile grows with time. We focus in this work on statistical methods for quantifying and comparing dynamic predictive accuracy of this kind of prognostic models, accounting for right censoring and possibly competing events. Dynamic area under the ROC curve (AUC) and Brier Score (BS) are used to quantify predictive accuracy. Nonparametric inverse probability of censoring weighting is used to estimate dynamic curves of AUC and BS as functions of the time at which predictions are made. Asymptotic results are established and both pointwise confidence intervals and simultaneous confidence bands are derived. Tests are also proposed to compare the dynamic prediction accuracy curves of two prognostic models. The finite sample behavior of the inference procedures is assessed via simulations. We apply the proposed methodology to compare various prediction models using repeated measures of two psychometric tests to predict dementia in the elderly, accounting for the competing risk of death. Models are estimated on the French Paquid cohort and predictive accuracies are evaluated and compared on the French Three-City cohort.

AB - Thanks to the growing interest in personalized medicine, joint modeling of longitudinal marker and time-to-event data has recently started to be used to derive dynamic individual risk predictions. Individual predictions are called dynamic because they are updated when information on the subject's health profile grows with time. We focus in this work on statistical methods for quantifying and comparing dynamic predictive accuracy of this kind of prognostic models, accounting for right censoring and possibly competing events. Dynamic area under the ROC curve (AUC) and Brier Score (BS) are used to quantify predictive accuracy. Nonparametric inverse probability of censoring weighting is used to estimate dynamic curves of AUC and BS as functions of the time at which predictions are made. Asymptotic results are established and both pointwise confidence intervals and simultaneous confidence bands are derived. Tests are also proposed to compare the dynamic prediction accuracy curves of two prognostic models. The finite sample behavior of the inference procedures is assessed via simulations. We apply the proposed methodology to compare various prediction models using repeated measures of two psychometric tests to predict dementia in the elderly, accounting for the competing risk of death. Models are estimated on the French Paquid cohort and predictive accuracies are evaluated and compared on the French Three-City cohort.

KW - Cohort Studies

KW - Computer Simulation

KW - Dementia

KW - Humans

KW - Models, Statistical

KW - Prevalence

KW - Psychometrics

KW - Reproducibility of Results

KW - Risk Assessment

KW - Sensitivity and Specificity

KW - Survival Analysis

U2 - 10.1111/biom.12232

DO - 10.1111/biom.12232

M3 - Journal article

C2 - 25311240

SN - 0006-341X

VL - 71

SP - 102

EP - 113

JO - Biometrics

JF - Biometrics

IS - 1

ER -