Based on data from the African American population, it appears that the Framingham estimates for the risk of coronary heart disease are reasonably predictive of cardiovascular outcomes in this group. However, a variety of factors exist in this population that may lead to a greater degree of misclassification than might be expected. Therefore, caution should be exercised when implementing risk algorithms, especially when effects in observational studies are incongruent with the experience of controlled clinical trials. Estimation of risk for cardiovascular events is a dynamic field, and current approaches will likely be modified as newer information is obtained.