ObjectivesThe objective of this study was to test whether glycemic control varies between adolescent patients diagnosed with type 1 or type 2 diabetes who are depressed and those who are not, after controlling for confounding factors. We hypothesized that diabetic children who have depression or a high risk to develop depression will have worse glycemic control, as indicated by higher hemoglobin A1c (HbA1c) values. Study designThis was a retrospective case-control study. MethodsA chart review was conducted in the Section of Endocrinology at St. Christopher's Hospital for Children in Philadelphia. Multivariate linear regression was used to determine effects of individual variables. ResultsA total of 214 records were included out of 263 reviewed. Significant differences were observed in type 1 diabetics (n = 156) between depressed and non-depressed patients in the percentage of females in the group (P = .002), the duration of diabetes (P = .005), age at diagnosis (P = .01), hemoglobin A1c (P = .03), and the percentage of those with a HbA1c greater than 14% (P = .03). Depression was associated with significant increases in HbA1c values in type 1 diabetics (P < .001). An interaction effect (P = .055) was observed between sex and depression. Given the small sample of children with type 2 diabetes, we were unable to perform any meaningful statistical analysis in this subgroup of patients. ConclusionsWe have detected a significant association between depression and glycemic control in adolescent girls with type 1 diabetes. This association appears to be moderated by sex. Depressed patients with type 2 diabetes generally display higher HbA1c values than their non-depressed counterparts.