Use of Continuous Glucose Sensors by Adolescents With Inadequate Diabetic Control (CGM-Teens)

Management of type 1 diabetes mellitus (T1DM) in adolescents is very difficult and innovative approaches are needed to help them achieve better glycemic control and behavioral outcomes. Continuous glucose sensors (CGS) have been refined progressively and provide acceptably accurate, nearly continuous estimates of glucose levels and trends. This increased quality and quantity of glucose data could be an excellent adjunct to conventional self-monitoring of blood glucose, permitting more informed diabetes decision-making. CGS could yield medical, educational and psychological benefits for adolescents with T1DM, but those with extremely variable self-management habits and suboptimal glycemic control may not realize these benefits readily. We hypothesize that a targeted, family-focused behavioral intervention could optimize benefit from adding CGS to T1DM therapy for youths with glycosylated hemoglobin (HbA1c) > 7.5%. A multi-site sample of 150 adolescents with T1DM and HbA1C of 7.5% to 10.0% will be randomized to either Standard Care for T1DM (SC), or to augmentation of SC with 9 months' use of a CGS device (CGS) or use of a CGS device supplemented with a targeted behavior therapy intervention (CGS+BT). Multiple measures of glycemic control, glycemic variability and health care use will be obtained during the study and there will be periodic assessments of demographic factors, diabetes self-management, family relations and psychological adjustment. Three specific aims will be addressed: 1. Evaluate whether CGS+BT yields more improvement in glycemic outcomes than CGS or SC; 2. Evaluate whether CGS+BT yields more improvement in behavioral outcomes than CGS or SC; and 3. Identify behavioral variables that mediate and moderate glycemic benefit from use of the CGS device. The study will also compare the cost effectiveness of CGS and CGS+BT relative to SC and evaluate the predictive utility of various indices of glycemic variability in youths. We hypothesize that, compared with SC and CGS, CGS+BT will yield significantly better biomedical outcomes (HbA1C; severe hypoglycemia; glycemic variability; proportion of glucose readings in the normal range) and behavioral outcomes (treatment adherence; parent adolescent teamwork; diabetes-related family conflict; quality of life; fear of hypoglycemia; and treatment satisfaction). After the 9 month randomized trial, all youths will be allowed to use the CGS device during an additional 3-month continuation phase. Statistical analyses will be based on individual growth modeling techniques. The application capitalizes on the Principal Investigator's prior and ongoing funded research on family management of T1DM, including trials of family-focused behavioral interventions, intensive therapy regimens, and clinical evaluations of continuous glucose sensors. The proposed study will determine whether a targeted behavioral intervention improves CGS benefits among adolescents with previously inadequate glycemic control. These results could demonstrate that adolescents with previously suboptimal diabetic control could realize multiple benefits from CGS use if they are provided with a specialized behavioral intervention.