Screening 1140 fifth graders for hypercholesterolemia: family history inadequate to predict results.

Cholesterol screening was performed on 1140 fifth-grade students in Scottsdale, AZ, as part of a school-affiliated, health-education program. The goals were to determine whether family history of heart disease or high cholesterol can predict which children have high cholesterol levels and to examine the feasibility of screening large numbers of elementary school students. Among the children studied, the mean cholesterol level was 168.3 mg/dL (4.35 mmol/L), and 13 percent had cholesterol levels above 200 mg/dL (5.20 mmol/L). Fifty-four percent had a family member with high cholesterol or a heart attack before age 60 years, but 36 percent of the students with cholesterol levels greater than 200 mg/dL (5.20 mmol/L) had a negative family history. Family history was neither sensitive nor specific as a predictor of elevated cholesterol levels (sensitivity 0.64, specificity 0.47, and positive predictive value 0.16 for predicting cholesterol levels greater than 200 mg/dL [5.20 mmol/L]). Large numbers of children were screened safely and efficiently with good student and parental cooperation. Results of this study do not support the current recommendations to screen children for hypercholesterolemia based upon their family histories.