Reducing variability and improving diabetes care in general practices in deprived and ethnic areas

Improvement in key performance indicators following intervention and in qualitative measures of patient and staff satisfaction

The secondary outcome measures of this study aims to evaluate the experience of staff and patients who participated in the intervention within the Coventry & Rugby CCGs, using qualitative methods.

We will conduct a qualitative study in parallel to the trial, the aim of which is to provide insight into participants’ experiences of taking part in the experimental intervention. As the trial has two arms (intervention and control groups), three qualitative methods (group evaluation meetings [one pre-intervention and another post-intervention], observation, and 1-1 semi-structured face-to-face interviews) will be used to understand both the patients and staff views regarding the intervention. The research will be carried out by a researcher trained and experienced in qualitative research methodologies, supported by second researcher as follows:

1. First, a one-off purposive evaluation group meeting involving both patients and staff will be organised prior to the intervention with 8-12 key participants. The researcher will share existing evidence of some of the benefits of implementing diabetes services in the primary care settings rather than in hospital clinics. The meeting will then seek the views and reasons for participation and what the participants intend to achieve. Participants will be engaged to identify and discuss the types of clinical sessions that can be removed from hospital clinics and transferred into general practices as part of the integrated Community Diabetes Service, and the potential benefits.

2. Observations - Observations will take place at the beginning, middle and towards the end of the of the intervention at 2-3 clinics/consultations to make comparisons of what has been said/agreed or done to what actually happened during the intervention duration, whilst making notes on the changes or differences. This approach is useful in collecting data in naturally occurring behaviours and contexts and will enable the researcher to question, understand and document how things work in their natural settings. It will also complement the study by enriching all the other qualitative data (evaluation meetings and the semi-structured interviews).

3. At the end of the intervention, another one-off evaluation meeting will be organised with all those who participated at the initial meeting, including two additional participants. This meeting will seek to understand the views/experiences of the participants and whether the intervention met their expectations – what was planned and achieved or planned but not achieved or subsequently changed. The additional two new participants will be able to follow the discussion and comment from a neutral view point.

4. All participants who participated in the post evaluation meeting, plus any other key stakeholder (maximum of 20 participants and/or reach achieve data saturation) will be invited to take part in a 1-1 face-to-face semi-structured interviews using topic guide questions to provide a level of consistency between interviews, whilst maintaining some flexibility to capture ideas raised by the interviewees. The interview topic guides will used mostly open-ended questions to focus the discussion on the phenomena of interest. The interviews will address some of the issues that were not fully explored at the evaluation meetings, what worked and/or did not work in the intervention, how could the intervention be improved and sustained to become a truly integrated Community Diabetes Service to benefit patients and the community at large and be transferable nationally. It is also give opportunity to some of the participants who may not have been fully opened to express their views in the group settings about our studied phenomenon.

Regarding 4 above, we will interview participants until we feel confident that we have reached data saturation. Based on our previous experiences we anticipate that the above stated maximum of 20 participants will be sufficient. Sampling in 1 – 4 will be purposive and will be aimed at reflecting the population of the intervention arm of the trial in relation to gender, ethnicity and social class both for the patients and staff. Pseudonyms will be used and any other identifying information will be anonymised to ensure participants’ confidentiality is maintained. Evaluation and interview transcripts and field notes will be uploaded into NVivo (QSR), a computerised programme for qualitative data, to assist with data management and analysis.

Our approach to data analysis will be thematic and will broadly follow the principles of grounded theory (Charmaz, 2006).

The meetings will be conducted at convenient locations of the participants but away from clinical areas in order to minimise any interruptions. The 1-1 interviews will be held at locations preferable to the participants, either in their office or home or NHS buildings within Coventry and Rugby CCGs. The duration of evaluation meetings will range from 60-90 minutes, interviews 45-60 minutes and the observations last between 2-4 hours each.

To ensure that participants are fully aware of their rights and responsibilities in the study, a detailed PIL/SIL will be giving to them - clarifying the objectives of the study, assuring them of their confidentiality and of the fact that participation in the study was entirely voluntary and of their right to withdraw at any stage without affecting their NHS care.