Skip to main content

07-07-2011 | Diabetes | Article

New model for 5-year CVD risk in Type 2 diabetes developed


Free abstract

Medwire news: A new risk model has been developed for estimating the 5-year risk for cardiovascular disease (CVD) in Type 2 diabetes patients, report Swedish researchers.

The model was developed from a large observational study of derivative data available for 24,000 diabetes patients, aged 30 to 74 years, who were registered in the Swedish National Diabetes Register.

As reported in the journal Diabetes Research and Clinical Practice, Bjorn Zethelius (Uppsala University) and team assessed the association between risk factors and the risk for CVD (defined as myocardial infarction, unstable angina, percutaneous coronary intervention, or stroke) over 5 years among these individuals.

Cox regression analysis was used to identify 12 independent CVD risk factors.

These included the following six continuous covariates, with hazard ratios (HRs) associated with a 1-standard deviation increase in each: onset age (HR=1.59); diabetes duration (HR=55); ratio of total to high-density lipoprotein cholesterol (HR=1.20); glycated hemoglobin (HR=1.12); systolic blood pressure (HR=1.09); and body mass index (HR=1.07).

The dichotomous variables male gender (HR=1.41), smoker (HR=1.35), microalbuminuria (HR=1.27), macroalbuminuria (HR=1.53), atrial fibrillation (HR=1.50), and previous CVD (HR=1.98) were also independent predictors in this model.

All 12 of these variables were then incorporated into an equation to give the 5-year CVD risk model.

In a separate validation sample of 5000 patients, the equation showed adequate calibration, including an excellent predicted-to-observed rate ratio of 0.97, and discrimination, at a C statistic of 0.72, for use in clinical practice, the authors say.

Commenting on the strengths of their model, they write: "Multivariable risk assessment avoids overlooking high-risk CVD patients and avoids needlessly alarming persons with only a single risk factor."

They add that estimation of 5-year CVD risk should be more robust and accurate, as well as being more comprehensible to the patient, than the usual 10-year risk.

MedWire ( is an independent clinical news service provided by Springer Healthcare Limited. © Springer Healthcare Ltd; 2011

By Sally Robertson