Application of Framingham risk estimates to ethnic minorities in United Kingdom and implications for primary prevention of heart disease in general practice: cross sectional population based study

Objective To compare the 10 year risk of coronary heart disease (CHD), stroke, and combined cardiovascular disease (CVD) estimated from the Framingham equations. Design Population based cross sectional survey. Setting Nine general practices in south London. Population 1386 men and women, age 40-59 years, with no history of CVD (475 white people, 447 south Asian people, and 464 people of African origin), and a subgroup of 1069 without known diabetes, left ventricular hypertrophy, peripheral vascular disease, renal impairment, or target organ damage. Main outcome measures 10 year risk estimates. Results People of African origin had the lowest 10 year risk estimate of CHD adjusted for age and sex (7.0%, 95% confidence interval 6.5 to 7.5.) compared with white people (8.8%, 8.2 to 9.5) and south Asians (9.2%, 8.6 to 9.9) and the highest estimated risk of stroke (1.7% (1.5 to 1.9),1.4% (1.3 to 1.6),1.6% (1.5 to 1.8), respectively). The estimate risk of combined CVD, however, was highest in south Asians (12.5%, 11.6 to 13.4) compared with white people (11.9%, 11.0 to 12.7) and people of African origin (10.5%, 9.7 to 11.2). In the subgroup of 1069, the probability that a risk of CHD greater than or equal to 15% would identify risk of combined CVD greater than or equal to 20% was 91% in white people and 81% in both south Asians and people of African origin. The use of thresholds for risk of CHD of 12% in south Asians and 10% in people of African origin would increase the probability of identifying those at risk to 100% and 97%, respectively. Conclusion Primary care doctors should use a lower threshold of CHD risk when treating mild uncomplicated hypertension in people of African or south Asian origin.