Estimating prevalence of type 1 and type 2 diabetes in a population of African Americans with diabetes mellitus

The pathogenesis, treatment, and outcomes of type 1 and type 2 diabetes dif fer. Current surveys derive population-based estimates of diabetes prevalen ce by type using limited clinical information and applying classification r ules developed in white populations. How well these rules perform when deri ving similar estimates in African American populations is unknown. For this study, data were collected on a group of African Americans with diabetes w ho enrolled at the Diabetes Unit of Grady Memorial Hospital in Atlanta, Geo rgia, from April 16, 1991, to November 1, 1996. The data were used to devel op some simple classification rules for African Americans based on a classi fication tree and a logistic regression model. Sensitivities and specificit ies, in which fasting C-peptide was used as the gold standard, were determi ned for these rules and for two current rules developed in mostly white, no n-Hispanic populations. Rules that yielded precise (minimum variance unbias ed) estimates of the prevalence of type 1 diabetes were preferred. The auth ors found that a rule based on the logistic regression model was best for e stimating type 1 prevalences ranging from 1% to 17%. They concluded that si mple classification rules can be used to estimate prevalence of diabetes by type in African American populations and that the optimal rule differs som ewhat from the current rules.