The c-index is not proper for the evaluation of t-year predicted risks

Research output: Contribution to journal › Journal article › Research › peer-review

Standard

The c-index is not proper for the evaluation of t-year predicted risks. / Blanche, Paul; Kattan, Michael W; Gerds, Thomas A.

In: Biostatistics, Vol. 20, No. 2, 2019, p. 347-357.

Research output: Contribution to journal › Journal article › Research › peer-review

Harvard

Blanche, P, Kattan, MW & Gerds, TA 2019, 'The c-index is not proper for the evaluation of t-year predicted risks', Biostatistics, vol. 20, no. 2, pp. 347-357. https://doi.org/10.1093/biostatistics/kxy006

APA

Blanche, P., Kattan, M. W., & Gerds, T. A. (2019). The c-index is not proper for the evaluation of t-year predicted risks. Biostatistics, 20(2), 347-357. https://doi.org/10.1093/biostatistics/kxy006

Vancouver

Blanche P, Kattan MW, Gerds TA. The c-index is not proper for the evaluation of t-year predicted risks. Biostatistics. 2019;20(2):347-357. https://doi.org/10.1093/biostatistics/kxy006

Author

Blanche, Paul ; Kattan, Michael W ; Gerds, Thomas A. / The c-index is not proper for the evaluation of t-year predicted risks. In: Biostatistics. 2019 ; Vol. 20, No. 2. pp. 347-357.

Bibtex

@article{25be87f84e224a02b317abbc07eff7e9,
title = "The c-index is not proper for the evaluation of t-year predicted risks",
abstract = "We show that the widely used concordance index for time to event outcome is not proper when interest is in predicting a $t$-year risk of an event, for example 10-year mortality. In the situation with a fixed prediction horizon, the concordance index can be higher for a misspecified model than for a correctly specified model. Impropriety happens because the concordance index assesses the order of the event times and not the order of the event status at the prediction horizon. The time-dependent area under the receiver operating characteristic curve does not have this problem and is proper in this context.",
author = "Paul Blanche and Kattan, {Michael W} and Gerds, {Thomas A}",
year = "2019",
doi = "10.1093/biostatistics/kxy006",
language = "English",
volume = "20",
pages = "347--357",
journal = "Biostatistics",
issn = "1465-4644",
publisher = "Oxford University Press",
number = "2",

}

RIS

TY - JOUR

T1 - The c-index is not proper for the evaluation of t-year predicted risks

AU - Blanche, Paul

AU - Kattan, Michael W

AU - Gerds, Thomas A

PY - 2019

Y1 - 2019

N2 - We show that the widely used concordance index for time to event outcome is not proper when interest is in predicting a $t$-year risk of an event, for example 10-year mortality. In the situation with a fixed prediction horizon, the concordance index can be higher for a misspecified model than for a correctly specified model. Impropriety happens because the concordance index assesses the order of the event times and not the order of the event status at the prediction horizon. The time-dependent area under the receiver operating characteristic curve does not have this problem and is proper in this context.

AB - We show that the widely used concordance index for time to event outcome is not proper when interest is in predicting a $t$-year risk of an event, for example 10-year mortality. In the situation with a fixed prediction horizon, the concordance index can be higher for a misspecified model than for a correctly specified model. Impropriety happens because the concordance index assesses the order of the event times and not the order of the event status at the prediction horizon. The time-dependent area under the receiver operating characteristic curve does not have this problem and is proper in this context.

U2 - 10.1093/biostatistics/kxy006

DO - 10.1093/biostatistics/kxy006

M3 - Journal article

C2 - 29462286

VL - 20

SP - 347

EP - 357

JO - Biostatistics

JF - Biostatistics

SN - 1465-4644

IS - 2

ER -

ID: 198707021