CV event predict

CV event predict predicts cardiovascular (CV) events and all-cause mortality among incident dialysis patients using multivariable Cox proportional hazard regression models derived from a multicenter prospective cohort of 1,520 patients.


Key Features:

  • Cohort: Multicenter prospective cohort of 1,520 consecutive incident dialysis patients with median age 70 years and 32.4% women (492 individuals).
  • Predictive modeling: Implements simple and complex multivariable Cox proportional hazard regression models incorporating variables such as age and diabetes comorbidity.
  • Primary endpoint: Composite endpoint of first CV event or all-cause death with median follow-up 1,285 days; 506 CV events, 392 deaths (152 CV-related), and 692 patients reaching the primary endpoint.
  • Model evaluation: Performance assessed by area under the receiver operating characteristic curve (AUROC) with the simple model AUROC 0.737 and complex model AUROC 0.765, and improvements quantified using net reclassification improvement (NRI) and integrated discrimination improvement (IDI).
  • Clinical utility: Comparison of simple versus complex models highlights trade-offs between predictive accuracy and reliance on routinely available clinical variables for risk stratification.

Scientific Applications:

  • Risk stratification: Predicting individual risk of first CV event or all-cause mortality at dialysis initiation for nephrology clinical decision-making and prognosis.
  • Personalized management: Informing individualized patient management strategies by estimating event probabilities from baseline clinical variables.
  • Prognostic research: Benchmarking and comparing prognostic models using AUROC, NRI, and IDI in dialysis populations.

Methodology:

Multivariable Cox proportional hazard regression models (simple and complex) fitted to a multicenter prospective cohort of 1,520 incident dialysis patients, with model performance evaluated by AUROC, net reclassification improvement (NRI), and integrated discrimination improvement (IDI), and a primary composite endpoint of first CV event or all-cause death with median follow-up 1,285 days.

Topics

Details

Programming Languages:
R
Added:
11/14/2019
Last Updated:
12/9/2020

Operations

Publications

Inaguma D, Morii D, Kabata D, Yoshida H, Tanaka A, Koshi-Ito E, Takahashi K, Hayashi H, Koide S, Tsuboi N, Hasegawa M, Shintani A, Yuzawa Y. Prediction model for cardiovascular events or all-cause mortality in incident dialysis patients. PLOS ONE. 2019;14(8):e0221352. doi:10.1371/journal.pone.0221352. PMID:31437231. PMCID:PMC6705850.