Cognitive behavioural therapy stabilises glycaemic control in adolescents with type 1 diabetes-Outcomes from a randomised control trial.

Abstract

To compare the impact of cognitive behavioural therapy (CBT) with non-directive supportive counselling (NDC) on glycaemic control and psychological well-being in adolescents with type 1 diabetes mellitus (T1DM). Participants aged 11 to 16 years with T1DM (duration ≥1 year) from 4 UK-based paediatric diabetes centres were randomised to receive either 6 weekly sessions of 1-to-1 CBT (n = 43) or NDC (n = 42), with 2 further sessions at 6 and 12 months. Follow-up continued for 12 months postintervention. Outcome measures included glycated haemoglobin A1c (HbA1c) and psychological scores. The HbA1c levels were available in 33 patients in each group for analysis. Between group difference of the overall changes in HbA1c across the study period was statically significant (P = .018). Geometric mean (range) HbA1c in the NDC group deteriorated from 68 (46-113) to 78 (48-128) mmol/mol (ie, 8.4 [6.4-12.5]% to 9.3 [6.5-13.9]%; P = .001), but was maintained in the CBT group from 72 (46-129) to 73 (51-128) mmol/mol (P = .51) (ie, 8.7 [6.4-14]% to 8.9 [6.8-13.9]%). More patients who have undergone CBT showed an improved or maintained HbA1c levels at 24 months (62.5% vs 35.5%, P = .032). Patients offered CBT with depressive scores in the lowest tertile (least depressive symptoms) showed improvement in HbA1c over time from 70 (46-102) to 67 (57-87) mmol/mol (P = .041) (ie, 8.6 [6.4-11.5]% to 8.3 [7.4-10.1]%), but not in the NDC group. The CBT showed borderline improvements in Children's Health Locus of Control (internal) scores over time compared with NDC (P = .05). The self-efficacy score showed significant improvement in both CBT (P < .001) and NDC (P = .03) groups over time. CBT demonstrated better maintenance of glycaemic control compared with NDC.