Modified realist evaluation of a complex, multi-centred, multi-intervention programme.

Well North was a complex, multi-intervention health improvement programme spanning 10 sites across the North of England. The aim was to address inequalities by improving the health of the poorest fastest, increasing resilience and reducing levels of worklessness. The intention of the programme was for all sites to have freedom and flexibility to conduct different interventions reflecting local priorities. Evaluation ran concurrently with the programme, and an iterative approach was required to ensure constant feedback, allowing the programme to be adapted and improved as necessary. Realist methodology was chosen for evaluation, as it provides insight into what works, for whom and in what circumstances. Due to the complex nature of the programme and diverse approaches, it was necessary to adapt the methodology to meet the needs of the evaluation. The Evaluation Team utilized a range of qualitative and quantitative techniques within the context of a Rapid Cycle Evaluation framework. For each project, Contexts, Mechanisms and Outcomes (CMOs) were identified at three stages and were incorporated into the CMO configuration, leading to the development of a middle range theory. Validation and testing of theory took place at every stage. Realist methodology was the most appropriate existing method. However, it still necessitated modification.