On the C‐statistics for evaluating overall adequacy of risk prediction procedures with censored survival data

Uno, Hajime; Cai, Tianxi; Pencina, Michael J.; D’Agostino, Ralph B.; Wei, L. J.

doi:10.1002/sim.4154

Cited by 1,154 publications

(997 citation statements)

References 34 publications

Supporting

Mentioning

983

Contrasting

Unclassified

Order By: Relevance

“…To determine which model best predicted inpatient mortality, receiver operating characteristic (ROC) curves were plotted and the regression models were compared on the basis of the area under the ROC curve (AUC) and its 95% CI [20,54]. The AUC quantifies the ability of our models to assign a high probability of mortality to the patients who died [42].…”

Section: Methodsmentioning

confidence: 99%

The Elixhauser Comorbidity Method Outperforms the Charlson Index in Predicting Inpatient Death After Orthopaedic Surgery

Menendez

Neuhaus

Dijk

et al. 2014

Clinical Orthopaedics &Amp; Related Research

442

292

View full text Add to dashboard Cite

Background Scores derived from comorbidities can help with risk adjustment of quality and safety data. The Charlson and Elixhauser comorbidity measures are wellknown risk adjustment models, yet the optimal score for orthopaedic patients remains unclear. Questions/purposes We determined whether there was a difference in the accuracy of the Charlson and Elixhauser comorbidity-based measures in predicting (1) in-hospital mortality after major orthopaedic surgery, (2) in-hospital adverse events, and (3) nonroutine discharge. Methods Among an estimated 14,007,813 patients undergoing orthopaedic surgery identified in the National Hospital Discharge Survey (1990Survey ( -2007, 0.80% died in the hospital. The association of each Charlson comorbidity measure and Elixhauser comorbidity measure with mortality was assessed in bivariate analysis. Two main multivariable logistic regression models were constructed, with in-hospital mortality as the dependent variable and one of the two comorbidity-based measures (and age, sex, and year of surgery) as independent variables. A base model that included only age, sex, and year of surgery also was evaluated. The discriminative ability of the models was quantified using the area under the receiver operating characteristic curve (AUC). The AUC quantifies the ability of our models to assign a high probability of mortality to patients who die. Values range from 0.50 to 1.0, with 0.50 indicating no ability to discriminate and 1.0 indicating perfect discrimination. Results Elixhauser comorbidity adjustment provided a better prediction of in-hospital case mortality (AUC, 0.86; 95% CI, 0.86-0.86) compared with the Charlson model (AUC, 0.83; 95% CI, and to the base model with no comorbidities (AUC, 0.81; 95% CI, 0.81-0.81). In terms of relative improvement in predictive performance, the Elixhauser measure performed 60% better than the Charlson score in predicting mortality. The Elixhauser model discriminated inpatient morbidity better than the Charlson measure, but the discriminative ability of the model was poor and the difference in the absolute improvement in predictive power between the two models (AUC, 0.01) is of dubious clinical importance. Both comorbidity models exhibited the same degree of discrimination for estimating nonroutine discharge (AUC, 0.81; 95% CI, 0.81-0.82 for both models). Conclusions Provider-specific outcomes, particularly inpatient mortality, may be evaluated differently depending on the comorbidity risk adjustment model selected. Future research assessing and comparing the performance of the

show abstract

Section: Methodsmentioning

confidence: 99%

The Elixhauser Comorbidity Method Outperforms the Charlson Index in Predicting Inpatient Death After Orthopaedic Surgery

Menendez

Neuhaus

Dijk

et al. 2014

Clinical Orthopaedics &Amp; Related Research

442

292

View full text Add to dashboard Cite

show abstract

“…The variables included in Cox regression models were those that were statistically significant in the univariate analysis and with minimal overlap in representation. Discrimination of serum FGF21 levels in predicting incident CHD was analyzed using C ‐statistics 22. The incremental value of serum FGF21 levels with reference to a baseline clinical model of conventional cardiovascular risk factors in predicting incident CHD was assessed by the integrated discrimination improvement and the category‐free net reclassification index.…”

Section: Methodsmentioning

confidence: 99%

Role of Circulating Fibroblast Growth Factor 21 Measurement in Primary Prevention of Coronary Heart Disease Among Chinese Patients With Type 2 Diabetes Mellitus

Lee

Woo

Chow

et al. 2017

JAHA

View full text Add to dashboard Cite

BackgroundFibroblast growth factor 21 (FGF21) has demonstrated beneficial effects on lipid and carbohydrate metabolism. In cross‐sectional studies, an association of raised circulating FGF21 levels with coronary heart disease (CHD) was found in some but not all studies. Here we investigated prospectively whether baseline serum FGF21 levels could predict incident CHD in subjects with type 2 diabetes mellitus and no known cardiovascular diseases.Methods and ResultsBaseline serum FGF21 levels were measured in 3528 Chinese subjects with type 2 diabetes mellitus recruited from the Hong Kong West Diabetes Registry. The role of baseline serum FGF21 levels in predicting incident CHD over a median follow‐up of 3.8 years was analyzed using Cox regression analysis. Among 3528 recruited subjects without known cardiovascular diseases, 147 (4.2%) developed CHD over a mean follow‐up of 4 years. Baseline serum log‐transformed FGF21 levels were significantly higher in those who had incident CHD than those who did not (222.7 pg/mL [92.8–438.4] versus 151.1 pg/mL [75.6–274.6]; P<0.001). On multivariable Cox regression analysis, baseline serum FGF21 levels, using an optimal cutoff of 206.22 pg/mL derived from our study, independently predicted incident CHD (hazard ratio, 1.55; 95% CI, 1.10–2.19; P=0.013) and significantly improved net reclassification index and integrated discrimination improvement after adjustment for conventional cardiovascular risk factors.ConclusionsWe have demonstrated, for the first time, that serum FGF21 level is an independent predictor of incident CHD and might be usefully utilized as a biomarker for identifying type 2 diabetes mellitus subjects with raised CHD risk, for primary prevention.

show abstract

“…One example of such global measure is the C-statistic of the prediction score (Harrell Jr et al 1996;Korn and Simon 1990;Pencina and D'Agostino 2004). When the event time T † is subject to right censoring which may have finite support [0, τ], one may consider a truncated C-statistic, as considered in Heagerty and Zheng (2005) and Uno et al (2011a). It is straightforward to extend C τ to our subgroup-specific C-statistic and construct an IPW kernel estimator for C τ (s) as for other accuracy measures.…”

Section: Resultsmentioning

confidence: 99%

Subgroup Specific Incremental Value of New Markers for Risk Prediction

Zhou¹,

Zheng²,

Cai³

2013

Risk Assessment and Evaluation of Predictions

View full text Add to dashboard Cite

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.

show abstract

On the C‐statistics for evaluating overall adequacy of risk prediction procedures with censored survival data

Cited by 1,154 publications

References 34 publications

The Elixhauser Comorbidity Method Outperforms the Charlson Index in Predicting Inpatient Death After Orthopaedic Surgery

The Elixhauser Comorbidity Method Outperforms the Charlson Index in Predicting Inpatient Death After Orthopaedic Surgery

Role of Circulating Fibroblast Growth Factor 21 Measurement in Primary Prevention of Coronary Heart Disease Among Chinese Patients With Type 2 Diabetes Mellitus

Subgroup Specific Incremental Value of New Markers for Risk Prediction

Contact Info

Product

Resources

About