In support of the observed reduction in HbA 1c levels from insulin pump use (continuous subcutaneous insulin infusion) shown in this study, previous studies performed in similarly aged cohorts with comparable follow-up time found a decrease in HbA 1c level of 0.6% (6.6 mmol/mol) from 1998 to 2005 ( 24) and an even larger reduction of 1.5% (16.4 mmol/mol) from 2000 to 2011 ( 25). Therefore, it is unlikely that disparate access to care could explain the observed sustained differences in HbA 1c over time. Since 2002, the BDC Hispanic/Latino Childhood Diabetes Program has provided bilingual culturally sensitive diabetes care service for Spanish-speaking and bilingual families. The Center provides, as needed, free diabetes care supplies (Helping Hands by the Children’s Diabetes Foundation) as well as free transportation and parking. Although adjusting for insurance status and insulin pump treatment may not fully control for disparities in access to care, the BDC provides the same level of diabetes management to all children, regardless of the family’s ability to pay. These associations are likely multifactorial and inherently incorporate language barriers, parental education, socioeconomic status, and cultural inequalities. Consistent with other studies ( 1, 23), we found a high percentage of children diagnosed in DKA were nonwhite or Hispanic and had government-provided or no medical insurance. Children that presented in DKA were younger than those who presented with milder symptoms, and younger age predicted higher HbA 1c over time (0.5% per 10 years in age at diagnosis), similar to previous studies ( 21, 22). Our study confirmed previously known risk factors for DKA at diagnosis of childhood type 1 diabetes ( Table 1). At 60 days post type 1 diabetes diagnosis *

Type 1 rta