Safety of a co-designed cognitive behavioural therapy intervention for people with type 1 diabetes and eating disorders (STEADY): a feasibility randomised controlled trial

Background Safe management of people with Type 1 diabetes and EAting Disorders studY (STEADY) is a complex intervention for people with type 1 diabetes and mild-to-moderate disordered eating (T1DE) integrating cognitive behavioural therapy (CBT) with diabetes education. Aim was to test feasibility of STEADY in a randomised controlled trial. Methods Feasibility parallel-group, randomised (blocks of four) controlled open-label trial (RCT) of STEADY against usual care (Control) at King's College London, UK. Participants were referred by clinicians or self-referred via social media advertisements. Forty adults with T1DE (Hba1c < 15%, body mass index 15-35 kg/m(2), age >= 18 years) were randomised. STEADY was delivered in 12 sessions by a CBT-trained Diabetes Specialist Nurse through video-conferencing and mobile app. Main outcome at 6 months post-randomisation was feasibility. Baseline mental health data (Structured Clinical Interview for DSM-5, SCID-5RV), and secondary biomedical outcomes (HbA1c; glucose time in range; TIR) and person-reported outcome measures (PROM: Diabetes Eating Problems Survey-Revised, DEPS-R; Eating Disorder Examination Questionnaire Short, EDE-QS; Type 1 Diabetes Distress Scale, T1DDS; Generalised Anxiety Disorder Assessment, GAD-7; Patient Health Questionnaire, PHQ-9; Impact of Diabetes Profile, DIDP) were collected. Analyses were conducted as intention-to-treat. ClinicalTrials.gov NCT05140564. Findings Of the 98 screened, 40 participants with T1DE were randomised (recruitment rate: 40.81%; 95% CI: 31.60%, 50.72%): 38 women, 1 man, 1 trans man (37 White, 1 White/Asian, 1 Black; 39 +/- 11 years old, diabetes duration 22 +/- 15 years, HbA1c 9.1 +/- 2.6%). The drop-out rate was 3/20 = 15% (4.39%, 36.55%) in STEADY and 2/20 = 10% (1.57%, 31.32%) in Control. STEADY reported lower GAD-7 (5.75 +/- 2.89 vs 10.18 +/- 5.31, p = 0.0060) and higher DIDP (3.13 +/- 0.63 vs 2.46 +/- 0.87, p = 0.020) at follow-up compared with Control, indicating lower anxiety and higher diabetes-specific quality-of-life. Compared to baseline, STEADY improved in DEPS-R, EDE-QS, GAD-7, PHQ-9 and T1DDS. Interpretation The STEADY-feasibility RCT demonstrated proof-of-concept for feasibility and mental health improvements in T1DE without deteriorating glycaemic control. A full scale RCT of STEADY will test effectiveness and implementation. Copyright (c) 2025 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).