OA Epidemiology

The Competing Risk Effect of Non-Cardiovascular Mortality on Total Cardiovascular Risk Prediction: Non Cardiovascular Mortality Should Be Considered in the Systematic Coronary Risk Evaluation (SCORE) Model

Proceedings of the 2013 annual meeting of the Netherlands Epidemiology Society

Volume 1 Issue S1 Abstract 44

 

B.G. Demissei, University Medical Centre, Groningen, the Netherlands
D. Postmus, University Medical Centre, Groningen, the Netherlands
R.T. Gansevoort, University Medical Centre, Groningen, the Netherlands
P. van der Harst, University Medical Centre, Groningen, the Netherlands
H.L. Hillege, University Medical Centre, Groningen, the Netherlands

Background
The current version of the Systematic Coronary Risk Evaluation (SCORE) model doesn’t take the competing risk of non-cardiovascular deaths into account. The aim of this study was to assess the impact of non-cardiovascular mortality on the predictive ability and potential yield of traditional cardiovascular risk factors regarding 10-year cardiovascular mortality in the general population.

Method
5871 Participants, from the Prevention of Renal and Vascular End stage Disease (PREVEND) study, aged 40 years and older who were free of atherosclerotic cardiovascular disease (CVD) at baseline were included. A standard Cox-model and Fine & Gray model (in which non-CVD deaths were treated as competing events) were fit.

Result
During a median follow-up of 10.5 years, 130 deaths from CVD and 375 deaths from non- CVD causes were reported. Compared to standard Cox-model, the Fine & Gray model showed better calibration, particularly at higher levels of actual risk. The two models showed comparable discrimination ability; however 7.3% of individuals classified into high risk group by the standard Cox-model were reclassified into low risk group by the Fine & Gray model.

Conclusion
A competing risks model taking non-cardiovascular mortality into account yields more accurate estimates of 10-year risk of CVD mortality with more appropriate classification of individuals into the different cardiovascular risk groups. This will reduce unnecessary overtreatment and associated medical and economical consequences resulting from misclassification of individuals into the high risk group. We strongly propagate that the competing risk of non-cardiovascular mortality should be accounted for in the SCORE risk prediction model.

Published: 06 Jun, 2013

 
Licensee OA Publishing London 2013. Creative Commons Attribution License (CC-BY)