Effect of psychotherapy for depression via home telehealth on glycemic control in adults with type 2 diabetes: Subgroup analysis of a randomized clinical trial
09/2017
Journal Article
Authors:
Egede, L. E.;
Walker, R. J.;
Payne, E. H.;
Knapp, R. G.;
Acierno, R.;
Frueh, B. C.
Volume:
24
Pagination:
596-602
Issue:
9
Journal:
J Telemed Telecare
PMID:
28945160
URL:
https://www.ncbi.nlm.nih.gov/pubmed/28945160
Keywords:
Aged
Anxiety Disorders/therapy
Behavior Therapy/*methods
Blood Glucose/analysis
Depressive Disorder/*therapy
Diabetes Mellitus, Type 2/metabolism/*psychology
Female
Glycated Hemoglobin/analysis
Humans
Male
Middle Aged
Psychotherapy/methods/*organization & administration
Telemedicine/*methods
Telehealth
home telecare
telepsychology
Abstract:
Objective We evaluated the impact of telemedicine-delivered behaviour activation treatment (BAT) on glycemic control in a subgroup of older adults with diabetes who participated in a randomized controlled trial for depression. Research design and methods We randomized older adults with major depression to same-room or telemedicine BAT. Each group received eight weekly sessions. For the subgroup analysis, we identified individuals with type 2 diabetes and obtained hemoglobin A1c at baseline and 12 months' follow-up. We used mixed-effects models (MEM) for repeated measures analysis to compare the longitudinal mean A1c. We estimated model-derived mean A1c values and considered an adjusted model to account for baseline health status. Results We included 90 individuals with type 2 diabetes of the original 241 in the subgroup analysis (43 in telemedicine and 47 in same room). Treatment groups were not significantly different at baseline for demographics, depression, anxiety or A1c levels (telemedicine 6.9 vs. same room 7.3, p = 0.19). Baseline mean A1c for the telemedicine group remained at 6.9 (55 mmol/mol) at 12 months, whereas baseline mean A1c for the same-room group increased to 7.7 (61 mmol/mol). Longitudinal trajectories of model-derived mean A1c indicated a significant main effect of treatment group on mean A1c value at study end (difference = -0.82, 95% CI -1.41, -0.24). Adjusted analyses gave comparable results. Conclusions Telemedicine-delivered BAT was superior to same room in achieving lower mean A1c values in participants with type 2 diabetes, suggesting BAT-delivered via telemedicine is a viable treatment option for adults with diabetes.
