Accelerating Delivery of Psychological Therapies after Stroke: A Feasibility Stepped Wedge Cluster Randomised Controlled Trial

Abstract

Background: Psychological problems post-stroke are common and debilitating, yet insufficient evidence-based psychological support exists for stroke survivors, either in stroke or general mental health services. Many stroke survivors with significant needs remain unsupported. To address this problem, we need pathways to identify, treat and manage psychological difficulties after stroke. The Accelerating Delivery of Psychological Therapies after Stroke (ADOPTS) study aimed to explore the feasibility of collaboratively developing, implementing and evaluating intervention packages (IPs) to facilitate access to, and increase the provision of, psychological support post-stroke. Methods: Stakeholder groups were formed across four sites in north-west England, comprising stroke and psychological services, to collaboratively develop site-specific IPs incorporating a psychological care pathway, staff training, a staff manual for stroke-specific psychological support and supervision. A feasibility stepped-wedge cluster randomised trial recruited patients admitted with stroke during the usual care (pre-implementation of the IP) and intervention (post-implementation) periods. The feasibility of IP implementation and their potential usefulness were evaluated through assessing wellbeing and the support received, and through a process evaluation incorporating interviews with staff, patients and carers. Feasibility evaluation included the recruitment rate and attrition rate; exploratory analysis (mixed-effects linear or logistic regression models) was used to assess the ‘promise’ of the intervention in achieving psychological distress outcomes (mood (PHQ-9), anxiety (GAD-7)), assessed using validated measures at 6 weeks and 6 months. Results: IPs were collaboratively developed at each site but implementation took longer than the per-study-protocol duration of three months. Nineteen training sessions (152 attendees) were delivered for nursing, therapy, NHS Talking Therapies and voluntary staff. Nursing staff were underrepresented due to difficulties with releasing staff. Manuals were developed for each site, incorporating a mood screening and referral algorithm, but these were not finalised at one site. Stroke and NHS Talking Therapies champions were identified in each site to facilitate cross-service staff supervision. A total of 270 patients were recruited over 14 months (133 usual care, 137 intervention), with 227 and 198 at 6 weeks and 6 months, respectively. Stroke staff found the training, manual and pathway helpful, and reported greater confidence in managing and referring psychological issues. NHS Talking Therapies staff found the training useful for adapting their therapy. However, the intervention took longer to implement in all sites, requiring an additional time period to be added to the stepped-wedge design. Conclusions: It is feasible to collaboratively develop and implement IPs for post-stroke psychological support. However, an alternative to the stepped-wedge design used here would be more appropriate for a future study. This study was registered in ISRCTN—the UK’s Clinical Study Registry (trial registration: ISRCTN12868810, registration date: 4 February 2016).