Understanding implementation of a complex intervention in a stroke rehabilitation research trial: A qualitative evaluation using Normalisation Process Theory.

Background: The Implicit Learning in Stroke study was a pilot cluster randomised controlled trial, investigating the use of different motor learning strategies in acute stroke rehabilitation. Participating Stroke Units (n = 8) were from the South East/West regions of the UK, with the experimental intervention (implicit learning) being delivered by clinical teams. It required therapists to change how they gave instructions and feedback to patients during rehabilitation. This paper reports the processes underpinning implementation of the implicit learning intervention. The evaluation aimed to i) understand how therapists made sense of, engaged with and interpreted the effects of the intervention; ii) compare this to the experience reported by patients; iii) extrapolate learning of broader relevance to the design and conduct of research involving complex interventions in stroke rehabilitation. Methods: Qualitative evaluation, with data collected through focus groups with clinical staff (n = 20) and semi structured interviews with people with stroke (n = 19). Mixed inductive and theory driven analysis, underpinned by Normalisation Process Theory. Results: How therapists made sense of and experienced the intervention impacted how it was implemented. The intervention was delivered by individual therapists, and was influenced by their individual values, beliefs and concerns. However, how teams worked together to build a shared (team) understanding, also played a key role. Teams with a more "flexible" interpretation, reported the view that the intervention could have benefits in a wide range of scenarios. Those with a more fixed, "rule based" interpretation, found it harder to implement, and perceived the benefits to be more limited. Therapists' concerns that the intervention may impair therapeutic relationships and patient learning were not reflected in how patients experienced it. Conclusions: Changing practice, whether in a research study or in the "real world", is complex. Understanding the process of implementation is crucial to effective research delivery. Implementation frameworks facilitate understanding, and subsequently the systematic and iterative development of strategies for this to be addressed. How teams (rather than individuals) work together is central to how complex interventions are understood and implemented. It is possible that new complex interventions work best in contexts where there are 'flexible' cultures. Researchers should consider, and potentially measure this, before they can effectively implement and evaluate an intervention. Trial registration: Clinical Trials - NCT03792126. [ABSTRACT FROM AUTHOR]