Guided self-help for depression in autistic adults: the ADEPT feasibility RCT

Background: Co-occurring depression frequently occurs in autism. Evidence-based psychological interventions have been successfully adapted to treat co-occurring anxiety, but there is little evidence about the usefulness of adapted cognitive-behavioural therapy for depression. To the authors' knowledge, to date there have been no randomised trials investigating the usefulness of low-intensity cognitive-behavioural therapy for depression in autism. Objectives: The objectives of the study were to (1) develop a low-intensity psychological intervention for depression adapted for autism, (2) assess the feasibility and patient and therapist acceptability of the intervention, (3) estimate the rates of recruitment and retention for a full-scale randomised controlled trial and (4) identify an appropriate measure of depression to be used in a full-scale randomised controlled trial. Design: The study comprised a randomised controlled trial (n = 70) with a nested qualitative evaluation (n = 21). Seventy eligible and consenting participants were randomly allocated to guided self-help or to treatment as usual. Setting: Adult autism services in two NHS regions. Participants: Adults with a diagnosis of autism spectrum disorder with depression, that is, a Patient Health Questionnaire-9 items score of >= 10. People who had attended more than six sessions of cognitive-behavioural therapy in the previous 6 months were excluded. Interventions: The low-intensity intervention (guided self-help) comprised materials for nine individual sessions, based on behavioural activation adapted for autism, facilitated by therapist guides (coaches) who were graduate-level psychologists who attended training and regular supervision. Treatment as usual was standard NHS care for depression. Main outcome measures: Outcomes were measured 10, 16 and 24 weeks post randomisation using self-report and interview measures of depression, anxiety, obsessive-compulsive symptoms, social function and quality of life, and a health-care and service use questionnaire. As this was a feasibility study also designed to identify the most appropriate measure of depression, it was not possible to specify the primary outcome measure or outcome point a priori. Results: The aims of the study were met in full. The guided self-help intervention was feasible and well received by participants and coaches. The majority of allocated participants attended the intervention in full. The most practical outcome point was determined to be 16 weeks. There were differential rates of attrition across the treatment groups: 86% of the guided self-help group remained in the study at 24 weeks, compared with 54% of treatment as usual group. The qualitative study suggested that guided self-help had enhanced credibility with participants at the point of randomisation. Inter-rater reliability of the interview measure of depression was less than adequate, limiting the conclusions that can be drawn from the prespecified sensitivity to change analyses. Conclusions: The intervention was feasible and well received. Although this feasibility study was not a fully powered trial, it provided some evidence that the guided self-help intervention was effective in reducing depressive symptoms. A full-scale clinical effectiveness and cost-effectiveness trial of the intervention is warranted. Future work: Improvements to the intervention materials as a result of qualitative interviews. Stakeholder consultation to consider future trial design, consider strategies to improve retention in a treatment as usual arm and select a self-report measure of depression to serve as the primary outcome measure.