Subgroup Specific Incremental Value of New Markers for Risk Prediction

Abstract

In many clinical applications, understanding when measurement of new markers is necessary to provide added accuracy to existing prediction tools could lead to more cost effective disease management. Many statistical tools for evaluating the incremental value (IncV) of the novel markers over the routine clinical risk factors have been developed in recent years. However, most existing literature focuses primarily on global assessment. Since the IncVs of new markers often vary across subgroups, it would be of great interest to identify subgroups for which the new markers are most/least useful in improving risk prediction. In this paper we provide novel statistical procedures for systematically identifying potential traditional-marker based subgroups in whom it might be beneficial to apply a new model with measurements of both the novel and traditional markers. We consider various conditional time-dependent accuracy parameters for censored failure time outcome to assess the subgroup-specific IncVs. We provide non-parametric kernel-based estimation procedures to calculate the proposed parameters. Simultaneous interval estimation procedures are provided to account for sampling variation and adjust for multiple testing. Simulation studies suggest that our proposed procedures work well in finite samples. The proposed procedures are applied to the Framingham Offspring Study to examine the added value of an inflammation marker, C-reactive protein, on top of the traditional Framingham risk score for predicting 10-year risk of cardiovascular disease.

Nonparametric estimation of time‐dependent ROC curves conditional on a continuous covariate

“…As pointed out in the introduction, the estimators for the FPR, TPR, and AUC in , , and have been already suggested in (p. 148). It should be noted, however, that both proposals differ on the computation of the inverse probability of censoring weights.…”

“…It should be noted, however, that both proposals differ on the computation of the inverse probability of censoring weights. In , the authors propose to estimate them either based on the Kaplan‐Meier estimator (when C is independent of both T and ( X , Y )) or through a Cox regression model. Accordingly, in contrast to that approach, our method is fully nonparametric in the covariate‐dependent censoring scenario.…”

“…Yet, to date, it has received little attention in the time‐dependent framework. Important references in this area are Cai et al , , Song and Zhou , Zheng et al , , Zheng et al , , and Zhou et al , . With the exception of the latter, estimation in all cases rely on (semi) parametric assumptions.…”

“…For instance, the proposals by Song and Zhou , Zheng et al , , and Zheng et al , are based on Cox‐type regression models . Special attention deserves the work by Zhou et al , . In that paper, the authors provide a nonparametric estimator for the time‐dependent ROC curve that is similar to one of the proposals we present here.…”

Reclassification calibration test for censored survival data: performance and comparison to goodness-of-fit criteria

“…When the true model is unknown, maximum likelihood estimation guarantees only calibration in the large. It does not guarantee for example a good discrimination and calibration in subgroups as noted by Zhou et al [8].…”