Early and later onset type 2 diabetes in uninsured patients: clinical and behavioral differences.

OBJECTIVE: The national burden of type 2 diabetes mellitus (T2DM) is increasing rapidly. This study investigated a) clinical differences between early onset and later onset T2DM; and b) if specific risk factors were associated with age at diagnosis or clinical outcomes among uninsured adults in a large urban setting. METHODS: We compared 417 adults diagnosed under age 30 with 968 adults diagnosed ages 50-58 on clinical and social measures using standard parametric tests. RESULTS: Early onset patients had higher hemoglobin A1c, were more likely to smoke and to be depressed, and had more emergency department visits. Insulin monotherapy was more common in early onset patients (32% vs. 11%). Complications were already present in 11% of early onset patients and 29% of later onset patients within one year of diagnosis. CONCLUSION: Early onset patients had more acute beta-cell failure and coped less well with their diabetes. It is crucial to expand specialized diabetes resources for young, medically indigent patients.

Ethnic differences predicting ketonuria in patients with Type 2 diabetes.

UNLABELLED: Ketonuria occurs more frequently than ketoacidosis does in patients with Type 2 diabetes and may portend serious future events. We studied the correlates of ketonuria among African American and Hispanic/Latino patients with Type 2 diabetes who did not have ketoacidosis. METHODS: We performed a retrospective cross-sectional study of patients referred for diabetes education from the emergency department or the urgent care clinic of a large, public, urban hospital from 2001 until 2003. Eligible patients had newly diagnosed Type 2 diabetes or diabetes diagnosed within 15 years but without treatment for the previous 2 months. Using logistic regression analysis, we analyzed the predictors of ketonuria and assessed whether they varied by ethnic group. RESULTS: Of 228 adult African American and Hispanic/Latino patients, 112 (49%) had ketonuria. Independent predictors of ketonuria were younger age (OR: 2.3), higher hemoglobin A(1c) (OR: 3.0), and a history of substance abuse (OR: 1.9) (for all: P<.05). Among Hispanic/Latino patients, ketonuria was also independently associated with low HDL-cholesterol (adjusted OR: 6.6; 95% CI: 2-20), but this relationship was not seen in African American patients. CONCLUSIONS: Ketonuria is common among newly diagnosed or untreated patients with Type 2 diabetes in a public hospital serving a predominantly minority population. Some correlates of ketonuria differ among ethnic groups, suggesting potential ethnic differences in ketogenesis. Prospective identification and study of patients with ketonuria may have important implications for appropriate therapy, especially if it is shown that ketonuria is a predictor of symptomatic ketoacidosis.