A Hybrid Effectiveness-implementation Trial of an Intervention to Reduce Diabetes-specific Emotional Distress in Teenagers

Diabetes distress among teens is common and associated with poor glycemic control and poor psychosocial outcomes. Diabetes-specific emotional distress (abbreviated DD in this proposal) is emotional distress culminating from living with diabetes and the burden of relentless daily self-management. Teens are at high risk for DD as diabetes self-management behaviors decline, and glycemic control worsens. Only 17% of teens achieve recommended glycemic targets for time-in-range (TIR) of 70-180 mg/dL and hemoglobin A1c <7.0%.7,8 DD is highly prevalent in teenagers (21%-52%) and is associated with suboptimal glycemic control, and an increased risk of future complications, even when controlling for demographic variables such as race and socioeconomic status. DD is a stronger predictor of A1c than depression. In a study of DD trajectories, teens with higher A1cs at baseline were more likely to experience chronically high levels of DD over time. Alarmingly, prospects do not improve as teens transition to adulthood without treatment. The American Diabetes Association's (ADA), the International Association for Pediatric and Adolescent Diabetes (ISPAD), and NIDDK highlight DD as a vital intervention target for youth.

Supporting Teen Problem Solving (STePS) is a highly scalable, group-based DD intervention for teens that builds diabetes resilience. In the sole efficacy trial to date, our team found that STePS demonstrated: 1) continued improvements in DD over time, with effect sizes increasing up to 3 years post-intervention, a statistically significant and clinically meaningful outcome, and 2) prevented worsening A1c, with 3-year post-intervention results of stable A1c. The latter is noteworthy because longitudinal evidence shows A1cs worsen in the teen years, not often improving until early adulthood (>25 years of age). While STePS has shown the rare capacity to move the needle on teens' glycemic control, a seemingly intractable outcome, it has not been tested in real-world contexts within existing pediatric diabetes programs.

Given the urgent need to translate research-based interventions into real-world settings, and the unacceptable lag-time between efficacy trials and routine uptake of clinical interventions, the investigators are currently conducting a type 1 hybrid effectiveness-implementation trial, funded by the ADA and by BreakthroughT1D (formerly known as Juvenile Diabetes Research Foundation), in 6 geographically distinct pediatric diabetes centers. Each center will enroll and randomly assign 60 economically, ethnically, and racially diverse teens aged 14-18 (N=360) to receive STePS either in-person (n=20) or via telehealth (n=20), or receive an educational telehealth control (n=20).

Design Considerations. The investigators will assess both effectiveness (primary outcomes) and implementation outcomes (secondary outcomes) via a Type 1 Hybrid Effectiveness Implementation design. The investigators chose a clinical effectiveness approach to evaluate STePS delivered in real-world contexts, using already-embedded behavioral health providers. The investigators prioritize external validity through recruiting a representative population, using an active comparison condition, and integrating the intervention into diabetes clinics. The investigators also chose an implementation approach to compare in-person versus telehealth strategies for improving uptake, fidelity, and sustainability of STePS. As noted by Curran et al. the time lag between research discovery and routine uptake is inflated by the step-wise progression of research from efficacy to effectiveness, and then only if effectiveness is established, to implementation. Blending effectiveness and implementation research is more desirable than pursing these lines of research independently and can "promote more rapid translational gains, more effective implementation strategies and more useful information for decision makers.

Effectiveness Outcomes: Improving Distress, A1c and Time-In-Range: Assessment of pre-post outcomes includes evaluating clinically meaningful changes in distress, A1c, and TIR. With respect to both A1c and TIR metrics, 0.3% is a clinically meaningful shift in the distribution of these measures of glycemic control as it generally represents about a 50% relative increase in the proportion of patients improving by 0.5% or more. However, 0.4% is generally accepted as evidence of a clinically meaningful change by the FDA. Therefore, if participant's A1c's improve by 0.4% at the end of 6 months, the investigators will conclude that STePS showed a clinically meaningfully improved metabolic control. With respect to diabetes-specific emotional distress, the minimal clinically important difference (MCID) of any psychosocial measure represents a threshold value of change in the score that represents a meaningful change in comparison to a statistically significant change. For this study, the investigators selected a stringent definition of MCID where the MCID is beyond the 95% confidence interval of the expected random variation in the distress score. For the PAID measures, that would be 8.3. Therefore, if participant's scores on the distress measures are reduced by >8.3 at the end of 6 months, then the study team will conclude that STePS has clinically meaningfully reduced distress. The investigators will also examine intervention delivery (in-person vs. telehealth) as a moderator of both STePS' effectiveness and implementation potential, increasing its potential for scalability.