Aim
Develop the intervention to be tested in WS3.
Develop intervention
The development details are published elsewhere.19,20 In brief, development comprised three phases using normalisation process theory (NPT) as the underpinning, flexible framework. This was only the third time that NPT has been used to develop a primary care intervention.20
Phase 1 comprised secondary analysis of existing qualitative data (3 data sets; 31 SSME recipients, 30 frontline delivery staff, 3 commissioners) and a systematic literature review20 of NPT applications for informing and assessing implementation processes in UK primary care settings (23 articles). The coded data were merged and analysed resulting in 46 attributes influencing SSME referral and uptake and the implementation processes involved. We produced a list of resources (pre-existing or in need of creation) needed to realise each attribute, for example, training for referrers and example referral pathways (for the full list, see Turner et al.19).
Phase 2 organised the 46 attributes into a viable intervention and ensured relevance to the current organisational context of primary care. A stakeholder workshop (n = 13) condensed the attributes to 11 priority attributes with explanatory notes and a refined compendium of resources needed to implement these attributes (see Turner et al.19 for a full list). A second set of stakeholders ranked the 11 priority attributes via a ‘diabetes update’ meeting (n = 26) and individual interviews (n = 16). The attributes ranked as most important were the need for:
- increased SSME awareness among primary care staff
- increased SSME awareness among patients and the public
- improved referral processes and booking systems
- tailored SSME for a range of audiences
- increased SSME accessibility for patients.
Phase 3 identified key intervention components based on the top five attributes through fortnightly meetings between the phase 1 and 2 analysis team and the intervention development team. This led to identification of four key intervention components (Box 2).
- Clear marketing strategy for SSME.
- User-friendly and effective referral pathway.
- New/amended professional roles.
- Toolkit of resources.
Build capacity and develop resources
Following completion of this mixed-methods work, the analysis and intervention development teams designed an intervention focused on the evidence-based components in Box 2. The development team then translated this into materials. Full details of the methodology used to develop the intervention are provided elsewhere.19 The intervention is described in the next section.
Changes from workstream 1 original plan
We intended to use two conceptual frameworks: NPT21,22 and the Consolidated Framework for Implementation Research (CFIR).23 However, early examination of the papers identified in the literature review indicated that the focus of these papers, and the level of detail they included, meant that coding according to the detailed domains of CFIR would add limited value. We therefore only used NPT due its flexibility for our approach to, and analysis of, data relating to the contextual issues impacting implementation and embedding. A description of NPT and the rationale for using it in WS1 is published elsewhere.20
We planned to review literature on implementing change in complex systems pertaining to SSME interventions in primary care (phase 1). To specifically inform intervention development, we narrowed the search to focus on the application of NPT in informing and assessing implementation processes in UK primary care settings.
We planned to involve people with T2DM in phase 3. However, phase 1 paid specific attention to this group’s views, while the views of those working in primary care (potential referrers to SSME) and service provision were underexplored. We were keen to explore how an intervention would fit with the realities of primary care, so we recruited accordingly. We had planned to facilitate stakeholder consultations using nominal group technique. Instead, we used a more flexible and semistructured brainstorming approach for our first stakeholder meeting and a ranking task conducted individually for our second set of stakeholder consultations for pragmatic reasons (e.g. the challenge of bringing all stakeholders together at the same time).