Establishing surveillance for diabetes in American Indian youth.

OBJECTIVE: To determine prevalence estimates in order to monitor diabetes, particularly type 2 diabetes, in American Indian youth. RESEARCH DESIGN AND METHODS: To explore the feasibility of developing a case definition using information from primary care records, all youth aged <20 years with an outpatient visit or hospitalization for diabetes were identified from the Billings Area Indian Health Service database in Montana and Wyoming from 1997 to 1999, and the medical records were reviewed. Classification for probable type 1 diabetes was based on age < or =5 years, weight per age < or =15th percentile at diagnosis, or positive results of islet cell antibody test. Classification for probable type 2 diabetes was based on weight per age > or =85th percentile or presence of acanthosis nigricans at diagnosis, elevated C-peptide or insulin, family history for type 2 diabetes, or use of oral hypoglycemic agents with or without insulin or absence of current treatment 1 year after diagnosis.
RESULTS: A total of 52 case subjects with diabetes were identified, 3 of whom had diabetes secondary to other conditions. Of the remaining 49 case subjects, 25 (51%) were categorized as having probable type 2 diabetes, 14 (29%) as having probable type 1 diabetes, and 10 (20%) could not be categorized because of missing or negative information. Prevalence estimates for diabetes of all types, type 1 diabetes, and type 2 diabetes were 2.3, 0.6, and 1.1, respectively, per 1,000 youth aged <20 years.
CONCLUSIONS: Our definitions may be useful for surveillance in primary care settings until further studies develop feasible case definitions for monitoring trends in diabetes among youth.