Effect of computerized coronary heart disease risk assessment on the use of lipid-lowering therapy in general practice patients

Abstract

UK guidelines recommend the use of cholesterol-lowering drugs for the primary prevention of coronary heart disease (CHD) when the 10 year risk of CHD is ≥ 30%. Absolute CHD risks are conventionally calculated using predictive equations derived from the Framingham Heart Study. We have previously adapted the laboratory information software to use the Framingham equation to calculate patients’ CHD risks, and have now evaluated the subsequent use of lipid-lowering therapy by those general practitioners who elected to use the system. 1132 risk calculation requests were received during 1998 from 14 general practices (range 23 to 186 requests). 999 patient records (88%) were available for review, of which 93 patients were excluded since they had recorded vascular disease (and were candidates for secondary prevention). Of the remaining 906 patients, 500 were male and 406 female, age 54.9 (11.2) years [mean (SD)]. 197 (22%) smoked, 180 (20%) had diabetes mellitus, 441 (49%) had a family history of CHD, 476 (53%) had hypertension and 223 (25%) were hyperlipidaemic. Median 10 year CHD risk was 13.2% (range 1–58%) and 81 patients (8.9%) had 10 year CHD risks ≥ 30%. Statins had been prescribed to 97 patients prior to CHD risk assessment, of whom 62 had calculated risks <30% and 35 risks ≥ 30%. Following CHD risk assessment, statins were prescribed to a further 3 patients with CHD risks ≥ 30%, and discontinued in 4 who had risks < 30%. Of the 43 patients with calculated CHD risks ≥ 30% who were not given a statin, 31 (72%) had a serum cholesterol below 6.5 mmol/L, the traditional threshold for considering a patient to be hypercholesterolaemic. CHD risk assessment has only a marginal impact on the use of lipid-lowering therapy even in a group of motivated general practitioners.