[The value of selective screening for the diagnosis of hypercholesterolemia]

The aim of this study was to evaluate the decision of the Dutch Cholesterol Consensus Meeting in 1987 to screen selected groups of persons for the identification of individuals with hypercholesterolemia. During 8 weeks serum cholesterol levels were measured in all 305 newly referred patients to the outpatient clinic of general internal medicine of the University Hospital Nijmegen. Information on age, sex, smoking habits and the selection variables was obtained. Of the patients 4 were referred because of hypercholesterolemia, 11 were excluded because of disease which could influence cholesterol levels. The results of 270 patients were available for evaluation. For statistical analysis the t test and linear and logistic regression analysis were used. Of the patients 87 individuals (32%) would have been candidates for selective screening; in 20 (23%) serum cholesterol levels were increased (greater than or equal to 6.5 mmol/l). Selective screening would have identified 20 of 42 hypercholesterolemic individuals (48%). Age showed a significant correlation with serum cholesterol levels (p less than 0.001); after correction for age, selective screening did not identify more hypercholesterolemic patients (p greater than 0.4). The serum cholesterol level appeared to be higher in the group meeting the selection criteria (p = 0.04). Among these criteria only hypertension was significantly correlated with serum cholesterol levels (p = 0.02) but sensitivity and predictive value for hypercholesterolemia were low. To identify a larger proportion of hypercholesterolemic individuals selective screening does not appear useful. If selective screening is chosen however, it should focus on hypertensive patients, also because the identification of patients with several risk factors is important from a prevention point of view.