Theoretical framework
This study sought to address important gaps in evidence about which people with SMI experience poor diabetes outcomes and why, and how health-care services could be changed to improve physical and mental health outcomes.
For our theoretical framework, we conceptualised socioeconomic conditions as a fundamental contributor to health outcomes.66 As described in Chapter 1, Social inequalities, people with SMI are more likely to be socially disadvantaged,68,69,74,79 leading to a reduced ability to take advantage of resources that improve health, or prevent or treat illness, which ultimately contributes to the significant inequalities in health outcomes seen in this population. We acknowledge that, as a ‘downstream’ determinant of health,67 the health-care system cannot completely remediate the social causes of poor health. However, using a social inequalities lens, we aimed to understand how health-care delivery could better respond to the physical health needs of people with SMI and diabetes, and identify areas for action in which health care and its organisation appear to generate further inequalities in this vulnerable population.
Study design
We used a concurrent triangulation mixed-methods design,105 comprising:
- A quantitative longitudinal observational study. The quantitative study interrogated CPRD-linked primary care and hospital records, along with linked HES, ONS and IMD data of a large sample of adults with SMI, adults with SMI and diabetes and a sample of matched controls with diabetes but no SMI.
- A qualitative interview study. The qualitative study sought to identify the health-care needs and health-care delivery concerns of this population through thematic analysis of semistructured interviews exploring how diabetes is managed alongside SMI, and how diabetes care is experienced by people with diabetes and SMI, their family members and informal supporters, and health-care professionals. The study drew on critical realism as a guiding methodological framework,106,107 which acknowledges that there is an external reality to be observed (e.g. through interrogation of data such as CPRD), but that there is also a lived reality that can be accessed only through the perspectives of individual actors who make sense of, and interact with, the world in which they live.
The mixed-methods design was underpinned by a pragmatic paradigm, which acknowledges that each data type provides a different, but equally important, worldview, and that, merged together, they will enable us to develop a more complete understanding of health inequalities in this population than would be possible from using either method alone.105 In particular, this approach allowed for an exploration of how people with SMI and diabetes navigate the health-care system and the range of factors that might influence their health and access to health care.
The definitive design for the study was informed by the results of an initial expert consultation and scoping of the literature, which aimed to develop a better understanding of the potential factors that could influence diabetes outcomes for people with SMI (see Chapter 4). This was used to determine which variables were important to explore in both the quantitative and qualitative studies and consider how to include them.
In line with the concurrent triangulation model, the quantitative and qualitative studies were conducted in parallel, and findings were merged at the interpretation stages to develop a fuller understanding of the factors contributing to poor outcomes and the drivers of improved health outcomes in SMI and diabetes. As part of the integration process, we conducted two co-design workshops involving members of the patient and public involvement (PPI) panel and participants from the qualitative interview study, and a final research team workshop, to iteratively make sense of and interpret the study findings and translate findings into key messages and implications for research and practice (see Chapter 7). Following Medical Research Council guidance for the development of complex interventions,108 we also used the co-design workshops to assess the potential acceptability and feasibility of potential interventions and service improvements into routine health care.
Figure 1 illustrates the four distinct components of the study and how they inter-relate. Each component will be discussed in detail in the upcoming chapters of this report:
- scoping of the literature and expert consultation (see Chapter 4)
- CPRD patient record analysis (quantitative longitudinal observational study) (see Chapter 5)
- qualitative interview study with people with diabetes and SMI, their family members and health-care staff (see Chapter 6)
- study integration, including co-design workshops (see Chapter 7).
FIGURE 1 The four components of this study and the relationships between them. (Figure not reproduced here; see the original article.)
Study organisation and governance
The study research team, comprising all co-applicants, was responsible for day-to-day management of the project, and held bimonthly project management meetings. A Study Steering Committee (SSC) with an independent chairperson and service user, and clinical, academic and commissioning representation was established, meeting every 6 months. The study was sponsored by Bradford District Care NHS Foundation Trust, which was also represented on the SSC.
Patient and public involvement
Overview of patient and public involvement
This study has been supported by DIAbetes and Mental illness: improving Outcomes aND Services (DIAMONDS) VOICE, a PPI group that has provided ‘expert by experience’ input to the study, with the aim of ensuring that the research remains relevant to people with SMI and that the outcomes of the research have the potential to have a meaningful impact on care and services.
Established in 2015, DIAMONDS VOICE comprises service users living with SMI and diabetes, and family members who provide support. The group is facilitated by a PPI co-ordinator and has 12 members who have been involved throughout the study, and a further two who have joined more recently. Three of these are family members or carers of people with diabetes and SMI. The group has met quarterly to discuss study progress, to offer advice on challenges faced by the research team and to undertake activities to support and inform the study. In addition, individual members of the group have contributed beyond the regular meetings, attending the co-design workshops and helping to raise the profile of the study through their networks and at local and regional dissemination events.
John Radford is a member of DIAMONDS VOICE and also a co-investigator on this study. He has participated in study research team meetings and SSC meetings, as well as co-design workshops (see Chapter 7, Co-design workshops).
Patient and public involvement activities
DIAMONDS VOICE has been engaged in project-related activities, starting from the grant application stages and continuing throughout the study duration (described below). The group has led the write-up of this section of the report.
Pre project
During the planning stages, the group assisted with the prioritisation of research questions, expressing a need for a better understanding of how diabetes is prevented and managed in people with SMI. The group suggested that it was important to understand the lived experience of comorbid SMI and diabetes, which influenced the research team to integrate qualitative inquiry into the study.
DIAMONDS VOICE representatives were invited to join the study research team and the SSC (see Study organisation and governance for study governance details) to ensure that the service user perspective was incorporated into the management and governance of the study.
Year 1
Activities included reviewing patient- and public-facing documentation, providing advice on interview topic guides and publicity materials, and suggesting ways to ask sensitive questions relating to finances and health. One member acted as a practice interviewee, which enabled researchers to refine the topic guide and to assess participant burden. In addition, members publicised the project across their own networks and at NHS events, for example running a stall at the sponsoring trust’s research and development conference in May 2018. Membership of DIAMONDS VOICE also expanded as a result of these engagement activities.
Year 2
DIAMONDS VOICE was involved in advising on participant recruitment, interpreting findings and in engagement activities. The research team had experienced difficulties engaging family members in the study, and advice was sought from the group on how to overcome this challenge. Members increased promotion of the study and expanded the stakeholder network, using their personal networks to engage family members supporting people with SMI and diabetes. This resulted in the creation of an ongoing connection with organisations such as Roshni Ghar, a local charity group that was able to provide input into the research and raise the study’s profile in the Asian community.
DIAMONDS VOICE also participated in the study co-design workshops, which took place in May and July 2019 (see Chapter 7, Co-design workshops). In advance of the workshops, members were asked to review invitation documents sent to potential service user and family member participants to ensure that they were clear and engaging. Group members also offered to support service user attendees with no previous experience of contributing to research workshops. During the sessions, members offered valuable insights into the authenticity of the interpretation of the study findings.
Group members played an important role in dissemination and engagement activities. For example, John Radford was a co-author on the protocol paper published in the Journal of Medical Internet Research.103 In addition, members continued to play an active role in engaging varied audiences, for example at a regional National Institute for Health Research (NIHR) event on multimorbidity research.
Ongoing patient and public involvement
DIAMONDS VOICE will continue to be involved after this study ends, for example through advising on key events for dissemination, giving feedback on materials used in dissemination of research to the public to ensure that they are effective and appropriate, and advising about future research priorities.
Impact of patient and public involvement on the research
Patient and public involvement has had a clear impact on the design, management and dissemination of this study. Importantly, the DIAMONDS VOICE group influenced the choice of mixed methods through their recommendation to explore the lived experience of SMI and diabetes in addition to conducting analyses of longitudinal patient records.
The study has benefited from PPI representation at study research team and SSC levels of governance, which has ensured that the research has continued to address questions that are meaningful to people living with SMI and diabetes. For example, although sleep difficulties were not identified as an important variable in the scoping of the literature, DIAMONDS VOICE representatives highlighted the significance of the relationship between sleep and the day-to-day management of SMI and diabetes. As a result, and in recognition that CPRD recording of sleep difficulties would be incomplete, a decision was taken to explore this issue through the qualitative interviews.
Input from DIAMONDS VOICE has also had an impact at the operational level of the study. Group members reviewed all patient- and public-facing project documentation, indicating ways to improve clarity and readability. They also advised on ways to make promotional materials engaging and appealing and have been instrumental in promoting the study across their mental health networks and expanding the stakeholder network. Topic guides and recruitment strategies were refined in accordance with advice from the group.
Reflections on the experience of participating in patient and public involvement
Researcher and member feedback on PPI involvement in this and other studies is collected annually by the PPI co-ordinator (see Appendix 1, Reflections on the experience of participating in patient and public involvement).