Katie Richards is a Research Associate working in King’s Improvement Science (KIS) at the Centre for Implementation Science, King’s College London. Her role involves conducting mixed-methods improvement and implementation research projects across King’s Health Partners. Before starting the Research Associate post, she completed her PhD at King’s College London, where she evaluated the national implementation and outcomes of an early intervention service for eating disorders in the UK.
Theories provide evidence-based and flexible tools to evaluate implementation processes. The Normalisation Process Theory (NPT) is a widely used implementation theory with demonstrated utility in supporting process evaluations. This study evaluated the role of NPT mechanisms in the national implementation, embedding, and integration of an early intervention services for eating disorders.
A mixed method evaluation was conducted. Twenty-one clinicians completed semi-structured interviews, and 211 clinicians completed longitudinal NPT questionnaires (NoMAD) administered before and after training and at a 3-month follow-up. For the qualitative data, the NPT was applied to inductively derived themes/subthemes to further evaluate underlying implementation mechanisms. The questionnaire data were analysed using multi-level growth models.
The inductive thematic analysis yielded six themes and 15 subthemes outlining barriers and facilitators to implementation at the wider system, service, implementation strategy, intervention, clinician, and patient levels. The early intervention service was largely normalising in teams with high levels of sense-making, engagement, collection action, and appraisal work taking place. These NPT mechanisms were more evident for some subthemes (e.g., compatibility and integration) than others (e.g., patient complexity and comorbidities). Insufficient capacity was the main factor inhibiting the normalisation in services. The quantitative data paralleled the qualitative findings. Specifically, NPT mechanisms were high at the outset, especially ‘buy-in’ and engagement. The training led to significant improvements in the NPT subscales, which continued to improve or remained approximately the same at the 3-month follow-up. The exception to this were the items related to sufficient training and resources, which initial improved post-training, but reduced at the 3-month follow-up.
The NPT characterised key mechanisms that were shaped by and interacted with features of the early intervention service, implementation strategy, and context to facilitate or hinder implementation. However, not all aspects of the implementation were directly captured by the theory.