TY - JOUR
T1 - Calibrated predictions for multivariate competing risks models
AU - Gorfine, Malka
AU - Hsu, Li
AU - Zucker, David M.
AU - Parmigiani, Giovanni
N1 - Funding Information:
Acknowledgments Malka Gorfine’s work was supported by Israel Science Foundation (ISF) Grant 2012898. Li Hsu’s work was supported by NIH Grants P01 CA53996 and R01AG14358. Giovanni Parmi-giani’s work was supported by NIH/NCI 5P30 CA006516-46 and Komen KG081303.
PY - 2014/4
Y1 - 2014/4
N2 - Prediction models for time-to-event data play a prominent role in assessing the individual risk of a disease, such as cancer. Accurate disease prediction models provide an efficient tool for identifying individuals at high risk, and provide the groundwork for estimating the population burden and cost of disease and for developing patient care guidelines. We focus on risk prediction of a disease in which family history is an important risk factor that reflects inherited genetic susceptibility, shared environment, and common behavior patterns. In this work family history is accommodated using frailty models, with the main novel feature being allowing for competing risks, such as other diseases or mortality. We show through a simulation study that naively treating competing risks as independent right censoring events results in non-calibrated predictions, with the expected number of events overestimated. Discrimination performance is not affected by ignoring competing risks. Our proposed prediction methodologies correctly account for competing events, are very well calibrated, and easy to implement.
AB - Prediction models for time-to-event data play a prominent role in assessing the individual risk of a disease, such as cancer. Accurate disease prediction models provide an efficient tool for identifying individuals at high risk, and provide the groundwork for estimating the population burden and cost of disease and for developing patient care guidelines. We focus on risk prediction of a disease in which family history is an important risk factor that reflects inherited genetic susceptibility, shared environment, and common behavior patterns. In this work family history is accommodated using frailty models, with the main novel feature being allowing for competing risks, such as other diseases or mortality. We show through a simulation study that naively treating competing risks as independent right censoring events results in non-calibrated predictions, with the expected number of events overestimated. Discrimination performance is not affected by ignoring competing risks. Our proposed prediction methodologies correctly account for competing events, are very well calibrated, and easy to implement.
KW - Calibration
KW - Competing risks
KW - Frailty model
KW - Multivariate survival model
KW - ROC analysis
KW - Risk prediction
UR - http://www.scopus.com/inward/record.url?scp=84897065350&partnerID=8YFLogxK
U2 - 10.1007/s10985-013-9260-x
DO - 10.1007/s10985-013-9260-x
M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???
C2 - 23737081
AN - SCOPUS:84897065350
SN - 1380-7870
VL - 20
SP - 234
EP - 251
JO - Lifetime Data Analysis
JF - Lifetime Data Analysis
IS - 2
ER -