Category: Obesity / Overweight

PS2- #B61 - Mechanisms of Therapeutic Change in a Randomized Control Trial of CBT to Decrease Depressive Symptoms in Adolescents At-Risk for Type 2 Diabetes

Friday, Nov 17
9:45 AM – 10:45 AM
Location: Indigo Ballroom CDGH, Level 2, Indigo Level

Keywords: Obesity / Overweight | Adolescent Depression | CBT

Elevated depressive symptoms (DepSx) predict worsening insulin resistance and the onset of T2D in adolescents. We conducted a randomized, controlled trial to evaluate if a 6-session CBT depression prevention program decreased DepSx and improved insulin resistance in a diverse sample of 119 girls (12-17 y, 64% African American, 11% Hispanic) with overweight/obesity (BMI≥85th %ile) and a family history of T2D. In a sample subset with higher baseline DepSx (CES-D>20; N=78), those randomized to CBT decreased DepSx at post-treatment and lessened insulin resistance indicators at 1-year follow-up, compared to those randomized to a standard-of-care health education (HE) control group (Shomaker et al. Ann Behav Med 2016; Dep Anx In Press). CBT is presumed to decrease DepSx by restructuring thoughts and increasing activities; yet, to what extent these factors underlie the effect of CBT to decrease DepSx in adolescents at-risk for T2D is unknown. We therefore conducted a secondary data analysis to evaluate potential mediators of therapeutic change, including negative/positive automatic thoughts and frequency/pleasantness of physical and social activities. At baseline and post-treatment, DepSx were assessed with the CES-D, automatic thoughts with the Automatic Thoughts Questionnaire, and pleasant activities with the Pleasant Events Schedule. Hayes Process Macro for SPSS with multiple imputed data was used to evaluate direct and indirect effects of treatment (CBT vs. HE) and changes in thoughts and activities on change in post-treatment DepSx. Models controlled for race, degree of T2D relative, therapy facilitator, puberty, body fat, and body fat change. Assignment to CBT had a direct effect on decreasing DepSx (p=.03) and social activities (p=.04). Increased frequency of physical (p=.02) predicted decreases in DepSx. However, the indirect effects of treatment on DepSx via changes in pleasantness or frequency of activities were not significant. Regardless of condition, increases in positive automatic thoughts predicted decreases in DepSx (p < .001). In sum, greater decreases in DepSx were observed following participation in CBT (vs. HE) in adolescent girls at-risk for T2D with moderate DepSx. Changes in pleasant activities and automatic thoughts did not explain the effect of CBT on DepSx. More frequent measurement of mechanisms via ecological assessments may shed light on temporal factors driving therapeutic change. Future studies should test if interventions that increase physical and social activities, increase positive thinking, and lessen negative thinking are particularly beneficial for adolescent girls at-risk for T2D with elevated DepSx.

Maeve B. O'Donnell

Graduate Student
Colorado State University

Ryan L. Rahm-Knigge

Graduate Student
Colorado State University

Lauren B. Shomaker

Assistant Professor
Colorado State University

Nichole R. Kelly

Assistant Professor
University of Oregon

Omni Cassidy

Graduate Student
Uniformed Services University of the Health Sciences

Rachel M. Radin

Post-Doctoral Research Fellow
University of California San Francisco

Marian Tanofsky-Kraff

Professor
Uniformed Services University of the Health Sciences
Bethesda, Maryland

Jack A. Yanovski

Chief, Section on Growth and Obesity
National Institutes of Health