Objective: To explore facilitators and barriers to using Experience-based Co-design (EBCD) and accelerated EBCD (AEBCD) in the development and implementation of interventions to increase activity opportunities for inpatient stroke survivors.

Design: Mixed-methods process evaluation underpinned by normalisation process theory (NPT).

Setting: Four post-acute rehabilitation stroke units in England.

Participants: Stroke survivors, family members, stroke unit staff, hospital managers, support staff and volunteers. Data informing our NPT analysis comprised: ethnographic observations, n= 366 hours; semi-structured interviews with 76 staff, 53 stroke survivors and 27 family members pre or post EBCD/AEBCD implementation; and observation of 43 co-design meetings involving 23 stroke survivors, 21 family carers and 54 staff

Results: Former patients and families valued participation in EBCD/AEBCD perceiving they were equal partners in co-design. Staff engaged with EBCD/AEBCD, reporting it as valuable improvement approach leading to increased activity opportunities. The structured EBCD/AEBCD approach was influential in terms of progressing coherence and cognitive participation and legitimated staff involvement in the process. Researcher facilitation of EBCD/AEBCD supported cognitive participation collective action and reflexive monitoring which was important in implementing and sustaining co-design activities. Observations and interviews post EBCD/AEBCD cycles confirmed creation and use of new social spaces and increased activity opportunities in all units. Changes occurred without increased staffing or organisational resource allocation. EBCD/AEBCD facilitated engagement with wider hospital resources and local communities increasing and enhancing activity opportunities. However, outside of structured group activity many individual staff-patient interactions remained task-focused with limited focus on enabling patient activity.

Conclusions: Using EBCD and AEBCD facilitated the development and implementation of environmental changes and revisions to work routines which supported increased activity opportunities in stroke units providing post-acute and rehabilitation care. Former stroke patients and carers contributed to improvements. Normalisation process theory’s generative mechanisms were instrumental in analysis and interpretation of facilitators and barriers at the individual, group and organisational levels.