Background: The prevalence of youth-onset type 2 diabetes (T2D) has been rising. There is a paucity of data on the risk factors for hyperosmolar hyperglycemic state (HHS) compared to diabetic ketoacidosis (DKA) in this population.

Methods: We used the national Kids’ Inpatient Database to identify pediatric admissions for DKA and HHS among those with T2D in years 2006, 2009, 2012, and 2016. Admissions were identified using ICD codes. We used descriptive statistics to summarize baseline characteristics and used Chi-squared test and logistic regression to evaluate factors associated with admissions for HHS compared to DKA in unadjusted and adjusted models.

Results: We found 6,8admissions for hyperglycemic emergencies in youth with T2D, of which 9.8% were due to HHS and 90.2% were for DKA. These admissions occurred mostly in youth 16-20 years old (70%) , females (52%) , who were non-White (Black 29%, Hispanic 17%) with public insurance (46%) and from the lowest income quartile (39.4%) . Most hospitalizations occurred in urban areas (82%) , southern US (47%) in private (58%) , large hospitals (61%) . In adjusted models, there was increased odds for HHS compared to DKA in males (OR 1.84, 95%CI 1.47-2.30) and those of Black race compared to those of White race (1.69, 95%CI 1.25-2.28) . There was no significant difference between those admitted for HHS compared to DKA with regards to age, other racial-ethnic groups, insurance payer, income, hospital region, size or ownership. In patients admitted with HHS compared to DKA, severity of illness (54% vs. 12.9% with major illness; p<0.001) and mortality were significantly higher (1.8% vs. 0.2%, p<0.001) .

Conclusion: While DKA represents most admissions for hyperglycemic emergencies among youth-onset T2D, those admitted for HHS had higher severity of illness and mortality. Male gender and Black race were associated with HHS admission compared to DKA. Additional studies are needed to understand the drivers of these risk factors.


E.M.Everett: None. L.E.Wisk: None. L.Chao: None.


National Institute of Diabetes and Digestive and Kidney Diseases (K01DK116932, L40DK129996)

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at