Prediction of adult cardiovascular multifactorial risk status from childhood risk factor levels. The Bogalusa Heart Study

Abstract

There is increasing interest in identifying children at risk for later development of cardiovascular disease. The authors studied 1,457 children who were first examined as part of the Bogalusa Heart Study in 1973 and again 15 years later as young adults. Age-, race-, and sex-specific quartiles were defined for each of three risk factor variables-ponderal index (weight/height3), systolic blood pressure, and cholesterol--for both the child and adult measures. Adults were classified as clustered if they were in the top quartile for each of the variables. Clustered adults had higher levels of several risk factor variables, in addition to the criteria variables, than did nonclustered individuals. Of children who placed in the top quartile on three factors, 21.8% were clustered as adults. Only 1.1% of those with no risk factor levels in the top quartile were clustered as adults (p < 0.0001). Logistic regression was used to predict adult cluster status from childhood variables levels. All three factors were significant predictors, with blood pressure being the most powerful. This well-fitting model is easily interpretable in terms of standard deviations and can be a useful model for identifying at-risk children.