Assessing 10-year coronary heart disease risk in people with Type 2 diabetes mellitus: Framingham versus United Kingdom Prospective Diabetes Study

Abstract

Aims: Previous studies have suggested that the Framingham coronary heart disease risk prediction equation underestimates risk among people with Type 2 diabetes. We compared the 10-year absolute risks of coronary heart disease (CHD) using a Framingham equation and a United Kingdom Prospective Diabetes Study (UKPDS) equation in adults with Type 2 diabetes. Methods: Participants were from a cross-sectional survey of a randomly selected population. There were 461 people with newly (n = 132) or previously diagnosed (n = 329) diabetes aged 35 to 74 years with no past history of cardiovascular disease or nephropathy. We examined predicted 10-year CHD risk by age, gender, and newly or previously diagnosed diabetes. Results: Overall the mean 10-year CHD risks predicted by the two equations were similar. Among men, the UKPDS and Framingham scores were almost identical below 60 years of age but at older ages, the UKPDS score was 4% - 11% higher than Framingham. For women, the Framingham score was higher than the UKPDS score between ages 40 and 65 years, but the UKPDS score was about 4% - 5% higher for women aged 70 years and over. The UKPDS equation tended to give higher risk estimates in people with a predicted 10-year Framingham CHD risk above 15%. Conclusion: Framingham CHD risk scores tended to be lower than UKPDS scores primarily in people above standard thresholds for drug treatment, so the clinical impact of underestimating risk is likely to be limited. Moreover, the UKPDS equation predicted lower risks than Framingham for women and newly diagnosed diabetes at otherwise low to moderate CHD risk, which could result in la- ter initiation of therapy in these groups if the UKPDS score was used instead of the Framingham score.