Embedding a user-centred approach in the development of complex behaviour change intervention to improve outcomes for young adults living with type 1 diabetes: The D1 Now Study

Phase 1: Formation of a Young Adult Panel (YAP)

Throughout the development work, a core work stream was the formation of the YAP. Full details of the process of the formation of the YAP are published elsewhere²⁷. The panel consisted of 8 young people aged between 18 and 25 years living with T1D who volunteered as co-researchers with the study team. The structured consultation process used to form the YAP commenced in February 2014 and the first official meeting of the YAP was held in June 2014. In line with INVOLVE guidelines²⁸ training was provided in research methods, processes and terminology, in committee skills and in qualitative methods. YAP members have contributed to all aspects of intervention development including reviewing study materials, abstracts, and manuscripts, and conference presentations.

Phase 2: Reviewing existing literature

In order to establish and synthesise research in this area, a systematic review to explore the evidence regarding the effectiveness of interventions aimed at improving clinical, behavioural or psychosocial outcomes for young adults with T1D was conducted⁹. This phase is aligned to the first stage of the MRC, the development phase which seeks to identify the existing evidence base. All interventions aimed at improving clinical, behavioural and psycho-social outcomes for young adults with T1D were included. A narrative synthesis (rather than a meta-analysis) was undertaken due to the large degree of heterogeneity between studies. Full methods and findings have been previously published⁹. The key objectives were to identify components of interventions and to measure the effectiveness of these interventions on outcomes for young adults. Findings were presented to the YAP for assistance with interpreting results prior to publication.

Phase 3: Qualitative study to understand young adults’ diabetes self-management behaviour

The next phase of intervention development consisted of a qualitative study with key stakeholders to 1) understand the factors which influence diabetes self-management and 2) explore how services and support could be improved. Ethical approval was obtained separately at each hospital recruitment site (NUI Galway Ref: 13/NOV/15, Galway University Hospitals Ref: C.A.1018, NHS National Research Ethics Service Ref: 14/LO/2254 and St. Vincent’s Healthcare Ref: O’Shea/10 Mar 2014). Interviews and focus group scripts were co-produced with YAP members. Semi-structured interviews were conducted with service providers (including doctors, diabetes specialist nurses and dieticians; n = 15; for interview guide see Supplementary File 1) and parents of young adults with T1D (n = 10; for interview guide see Supplementary File 2). Three focus groups were conducted with young adults with T1D (Galway focus group n=5; Belfast focus group n= 9; Dublin focus group n= 4; for interview guide see Supplementary File 3). All potential participants were invited to participate via post, apart from service providers who were contacted via email. Potential participants responded to author MCOH to schedule an interview or participation in a focus group. Thematic analysis was used to analyse the data, which was then further categorised using the framework of the COM-B model to identify the factors that drive type 1 diabetes self-management behaviour among young adults (manuscript in preparation; led by author LH).

As described above the COM-B model aims to better understand a target behaviour in context. Therefore, the COM-B model was used to aid identification of appropriate target behaviours for this complex intervention. For example, Capability is the individual’s ability to perform a behaviour (i.e., both physical and psychological). Opportunity describes the specific context that may facilitate the behaviour, while Motivation describes the processes that fuel and influence the behaviour²³.

Phase 4: Information sharing symposium and expert panel consensus meeting

In June 2016, a 3-day event was organised in Galway, Ireland, entitled: ‘Strength in Numbers: Teaming up to improve the health of young adults with type 1 diabetes’²⁹. The overall aim of the Strength in Numbers event was to exchange knowledge and experiences among stakeholders to gain input into the D1 Now intervention development process. As part of this event, a range of key international and local stakeholders, including young adults with T1D, health professionals, policy makers and researchers, presented on a range of topics relevant to develop effective interventions to improve outcomes for young adults. These included: Supporting self-management; Digital technology for supporting self-management among young adults; Engaging young adults in research and service design; and Innovations in service delivery for young adults with type 1 diabetes. Following this, 18 experts including young adults with type 1 diabetes, researchers, service providers and policy influencers, took part in an Expert Panel meeting. The breakdown of Expert Panel meeting participants illustrates the diversity of roles and perspectives, and is displayed in Table 1.

Representative groups were formed and each group was assigned a facilitator from the D1 Now research team. One of the authors (MB), an expert in developing complex behaviour change interventions and gathering stakeholder consensus, facilitated this meeting. The findings of the systematic review⁹ and qualitative study were summarised, and the take home messages from the previous day’s conference were briefly revisited. As a result of the previous phases of the D1 Now intervention development process, three areas emerged as potentially important targets for improving outcomes among young adults with type 1 diabetes. Prior to the meeting, preparatory materials were sent to the Expert Panel members, including a breakdown of the D1 Now study, and descriptions of these three focus areas. The focus areas provided the basis for the Expert Panel meeting activities. The three areas that emerged as potentially important targets for outcomes from Phase 1 and 3 were outlined to experts (see Results section).

Each team was asked to examine and debate two of the three focus areas over two rounds of discussion within each team and with the larger group. Teams were asked to focus on only two of the three focus areas for pragmatic reasons to facilitate in-depth group discussion of the core areas within the allotted short timeframe. Using a structured behavioural analysis recommended as part of intervention development²³ each group identified specific strategies which could be used to address their assigned focus area. This involved considering the framework of the COM-B model to identify the factors that drive type 1 diabetes self-management behaviour among young adults and how the suggested strategies would impact on young adults’ ‘Capability, Opportunity and Motivation’ to perform this target behaviour. The strategies identified by each group to address each of the three focus areas were summarised (Supplementary File 4), which were assessed using ratings of high, medium or low, according to the criteria of; impact on young adult self-management; how possible or feasible the strategy was; and the potential for each strategy to have subsequent positive effects. Each group was asked to generate 3 strategies per session, but time constraints meant this was not achieved in every case.

Expert Panel teams were also tasked with brainstorming plans and barriers for implementing their most promising strategy before presenting to the larger group. These were then grouped to identify relevant themes. Figure 1 in the Results section depicts important themes that emerged during discussion on the plans and barriers to implementation of the promising strategies.

Figure 1. Themes emerging from expert panel discussions.

Phase 1: Formation of a Young Adult Panel (YAP)

The development of the YAP resulted in meaningful involvement between young adults, researchers and service providers. The YAP made significant contributions to all aspects of the study development, in particular developing the qualitative interview topic guides, the participant invitation letters, consent forms and information sheets, and disseminating study findings’ by submitting scientific abstracts to national conferences, being invited speakers at two national conferences, being interviewed for local newspapers and radio, and by entering national science competitions. The YAP members were part of the organisation committee for our international symposium in June 2016 (see Phase 4) and two YAP members were elected to sit on the study’s steering committee. The D1 Now YAP were named as official Collaborators on a successful Health Research Board (HRB) Definitive Intervention and Feasibility Award to further progress the D1 Now study (Ref: DIFA_2017-034).

Phase 2: Reviewing existing literature

Eighteen studies were included in the review of interventions aimed at improving clinical, behavioural or psychosocial outcomes for young adults with Type 1 diabetes and categorised as follows: Health Services Delivery (n = 4), Group Education and Peer Support (n = 6), Digital Platforms (n = 4) and Diabetes Devices (n = 4). Study designs included 1 randomised controlled trial, 3 retrospective, 7 feasibility/ acceptability studies and 8 studies with a pre/ post design. Continuity, support, education and tailoring of interventions to young adults were the most common themes across studies. In particular continuity and support have emerged as crucial through other key phases of the current intervention development process. Glycaemic control was the most frequently measured outcome but only 5 of 12 studies that measured it showed a significant improvement. Due to the wide variance of quality and reporting among the studies, the effectiveness of individual interventions outcomes could not be established and warrants further investigation.

Phase 3: Qualitative study to understand young adults’ diabetes self-management behaviour.

The qualitative engagement study with key stakeholders elicited rich information regarding how self-management of T1D is influenced by interactions between capability, opportunity and motivation. The findings of this study demonstrate that social and physical opportunity drive self-management through access to supportive diabetes service providers which facilitates young adults accessing the health care they want, in a way that is most appropriate and useful to them (i.e., age-appropriate consultation style and frequent follow-up through different modes of contact as well as traditional clinics). Findings suggest that diabetes teams can also optimise external physical and social factors such as access to resources such as diabetes devices and peer networks to enhance self-management. Importantly, an individual’s perception of their capability to self-manage impacts their motivation to perform self-management behaviour in what appears to be a bi-directional relationship. Further, the role of the clinic and the diabetes team were foundational to self-management. According to these findings, all components of the COM-B model play a role in determining self-management behaviour. Physical and social opportunity factors, such as the influence of service providers, emerged as dominant drivers of self-management.

Phase 4: Information sharing symposium and expert panel consensus meeting

As described in the Methods section, during the Expert Panel meeting, each team was asked to examine and debate two of the three focus areas below over two rounds of discussion within each team and with the larger group. Using a structured behavioural analysis recommended as part of intervention development²³ each group identified specific strategies which could be used to address their assigned focus area.

These focus areas to improve self-management were:

The full list of strategies identified by each group to address each of the three focus areas can be seen in Supplementary File 4. The strategies deemed to be most promising for addressing each focus area (printed in bold) according to the Expert Panel are listed in Table 2.

Results of Expert Panel team discussions on plans and barriers for implementing promising strategies are depicted in Figure 1. Seven key areas were identified: 1) strategies to facilitate continuity of care, which was seen as important for relationship building within the therapeutic relationship and highlights the desire to work with a healthcare professional who is familiar with each individual and their history; 2) the needs of young adults should be frequently reviewed, emphasising the nature of young adulthood as a time of transition which requires the healthcare team takes a dynamic approach to adapt to changing demands; 3) joint engagement of both young adults and service providers in diabetes management. Joint engagement in diabetes management requires collaborative discussions whereby both young adults and healthcare professionals outline current care needs and future goals and action plans on a regular basis; 4) a choice of when to transition from paediatric services and to which clinic in the adult service. Young adult involvement in the transition process ensures that there is limited stress experienced and enables the young adult to adapt to the expectations associated with attending an adult service in a way that is appropriate for them; 5) use strategies from relevant arenas (i.e., social psychology and marketing) to engage young adults. Services should leverage more contemporary means of engaging with this younger cohort. Traditional methods of initiating engagement (e.g., appointment letters) may not be the most appropriate for this target group; 6) multiple modes of contact and engagement should be available. A variety of modes of communication are now needed to ensure that the care needed is delivered as efficiently as possible. This could potentially include text messages and other online services to complement more traditional modes of communication; and 7) clinics designed to be accessed by young adults flexibly, as and when they are needed. This key area of accessibility is very important for this cohort. Flexibility is integral to this group due to the constant changing needs (e.g., moving out of home for the first time, work/study schedules, travelling abroad alone) and healthcare professionals need to have the capacity and methods to respond and deliver care as needed.

There was significant overlap in the details of the implementation plans described across the groups. According to the Expert Panel, the most promising strategies for addressing the focus areas will involve action on the part of the diabetes service organisation, service providers and young adults, often in partnership. The philosophy and running of the clinic will need to change, for example staff must be open to integrating new approaches such as online systems, communicate in new ways and adopt a young adult-centred ethos. Flexibility is the theme which describes panel members’ consensus around when, where and how often a strategy should be implemented. According to the Expert Panel, strategies such as meeting with a key worker or youth worker should occur at the beginning of an intervention and intermittently as required thereafter in a personalised manner and when the young adult requires the support. Platforms or locations for connecting with key workers, such as online or in person, in the clinic or another suitable location, should also be utilised. Bringing everyone to the table to implement strategies effectively was also a strong theme across groups and focus areas. Successfully implementing strategies will involve the participation of all stakeholders including young adults and service providers and depending on the strategy and the preferences of young adults, it may also involve friends and family.

The barriers to implementing each strategy included funding, acceptance and engagement, training of service providers and young adults, logistical issues such as planning, time, staffing levels and access to technology within the HSE, identification or recruitment of appropriate key workers and accountability of key workers. Solutions for overcoming barriers included engaging with Diabetes Ireland (Irish national charity) and with young adults themselves, for example learning from other service-user groups seeking similar changes in other services, referring to existing best practice guidelines, and pilot testing strategies and collecting feedback. In addition, it may be beneficial to seek funding from other sources to support this work or aspects of it, such as Social Entrepreneurs Ireland [national group aimed at solving big issue societal problems). Issues related to engagement among young adults and service providers could be addressed using marketing and social psychology approaches, by building flexibility within strategies and addressing technology gaps (e.g., leveraging existing IT solutions and services to deliver care in a more nuanced way).

D1 Now: Current proposed intervention components

Subsequent to the above development process, three components of the D1 Now intervention have been identified: 1) a key worker to introduce the young adult to the diabetes service and flexibly address the needs of young adults within the clinic; 2) an online Young Adult Service Portal to facilitate relationship building between staff and young adults; and 3) an agenda setting tool to facilitate joint decision making and goal-setting within clinic appointments. These three components are currently being refined within the intervention optimisation phase. The operationalisation of these components for the final intervention content will be developed based on iterative cycles of pilot testing and feedback from young adults and healthcare professionals. The final version of the D1 Now intervention will then be subject to feasibility and piloting phases (funded by a HRB Definitive Intervention Feasibility Award).

Reflections on the user-centred approach within D1 Now intervention development

Engaging with key stakeholders is recommended as best practice for effective intervention development^10,13 and researchers are increasingly encouraged to adopt a PPI framework within health research. The D1 Now study team has demonstrated that it is both feasible and desirable to include a PPI panel of young adults with T1D in health research^27,29.

Formation of the YAP was time-intensive and required commitment from the research team to drive recruitment and identified training with the YAP members. Acting as a member of the YAP required regular meetings, up-skilling as many members became familiar with new terms and reading different types of material (participant information sheets, grant applications and manuscripts being prepared for submission) and engagement with an iterative research process. This involved managing expectations of both the YAP and the research team in terms of what is feasible, and implementable, and how quickly change can happen given research designs and processes. Expectation management was undertaken by a neutral ‘knowledge broker’ Jigsaw [a youth mental health service with experience in young adult panels]^30–32. The process of understanding and integrating the ethos of PPI occurred over the course of the study. The trust between the research team and the YAP grew through multiple interactions, culminating in the large scale event described in Phase 4 (Information sharing symposium and expert panel consensus meeting). Currently, a sub-study aiming to explore the impact of PPI on the intervention development and overall study is underway.

Implications for research and practice

It has become clear that D1 Now needs to reimagine T1D young adult care from its very foundation with the guidance of the YAP^9,33. Therefore, an exploratory approach underpinned by best practice and theory was chosen to address the aims of this study. This user-centred approach acts as a template for other research teams to work collaboratively with people living with a chronic condition to develop meaningful strategies for impacting health services.

The current study highlights the importance of consistent engagement with key stakeholders throughout intervention development. Theory (e.g., the BCW) is an important tool used throughout this process but all theory and interventions require grounding within a health services context if they are to contribute to successful service re-design. Our PPI approach is particularly important to identify potential barriers to implementation within the T1D population.

Use of the BCW, and identifying aspects of capability, opportunity and motivation that should be targeted will facilitate explicit linking of these ‘drivers’ of behaviour to intervention functions which will allow us to explicitly outline the proposed mechanisms of action of the D1 Now study intervention. Investigating these hypothesised mechanisms of action, and whether included BCTs are appropriate and acceptable for our target population, will be further examined within future testing and refining phases and subsequently with a feasibility phase and randomised pilot.

Strengths and limitations

The systematic intervention development process addressed an important criticism of previous behaviour change interventions targeting complex self-management behaviours. As identified by O’Hara and colleagues⁹, previous interventions are poorly described and their effectiveness cannot be accurately evaluated. However, a phased intervention development process is hugely resource-intensive³⁴. Various outputs, individual and team decisions and decision logic can be difficult to document and present in a coherent reusable fashion. The use of the BCW framework along with checklists such as the Template for Intervention Description and Replication (TIDieR) checklist³⁵ to report the intervention content will assist in future replication and prevent recurring research waste as demonstrated through previous reviews of interventions unable to yield tangible effective intervention mechanisms to be implemented in future research.

A further strength of this study, was the multidisciplinary nature of the stakeholders and the PPI approach adopted throughout the D1 Now study. This facilitated discussion between allied health professionals, patients and researchers which resulted in rich and nuanced output to explore within the next phase of the D1 Now feasibility and pilot phases.