Coronary heart disease risk assessment in diabetes mellitus: comparison of UKPDS risk engine with Framingham risk assessment function and its clinical implications
Article first published online: 22 JAN 2004
Volume 21, Issue 3, pages 238–245, March 2004
How to Cite
Song, S. H. and Brown, P. M. (2004), Coronary heart disease risk assessment in diabetes mellitus: comparison of UKPDS risk engine with Framingham risk assessment function and its clinical implications. Diabetic Medicine, 21: 238–245. doi: 10.1111/j.1464-5491.2004.01116.x
- Issue published online: 22 JAN 2004
- Article first published online: 22 JAN 2004
- Accepted 17 June 2003
- coronary heart disease;
- diabetes mellitus;
- primary prevention;
- risk assessment
Aims To assess differences between absolute coronary heart disease (CHD) risks calculated by Joint British Societies (JBS) risk calculator and UKPDS risk engine and its impact on CHD primary prevention management in diabetes mellitus (DM).
Methods Seven hundred Type 2 DM patients without arterial complications were identified from nine general practices in the Scarborough area. Their absolute 10-year CHD risks were calculated. The differences in the proportion of patients identified for aspirin and statin under JBS and National Institute for Clinical Excellence (NICE) guidelines by these two methods were determined. The proportion of additional patients identified for statin in the Scarborough population as a consequence of CHD risk threshold reduction from 30 to 15% (as recommended by NICE) was also determined.
Results UKPDS risk engine calculated significantly higher mean 10-year CHD risk (UKPDS vs. JBS, 21.5 vs. 18.3%, P < 0.0001). Both methods identified approximately 65% of patients to be eligible for aspirin and statin if NICE recommendations were followed. At a risk threshold of 30%, the UKPDS risk engine identified more patients for statin. Reducing the CHD risk threshold from 30 to 15% for statin initiation will identify an additional 0.5% of the total population for this treatment.
Conclusions Both methods are comparable in identifying at-risk patients under NICE recommendations. A high proportion has risk levels that merits primary CHD prevention. Lowering the risk threshold for statin treatment has a small numerical impact on the whole population.