Principal findings and achievements
This multimethod longitudinal study using ethnography, interviews, document analysis and workshops revealed wide diversity of experience and strategic decision-making in a sample of 12 general practices. This was in the context of a policy push for digitalisation and multiple contextual complexities. We documented infrastructural weaknesses and challenges with procurement of new technologies. We demonstrated that remote and digital care in general practice is not, in general, unsafe, but it does have potential downsides including increased complexity of care, widening of inequities (especially for multiply disadvantaged groups), fragmentation of care with reduced relational continuity, and technology-induced inefficiencies (such as double-handling of patients and processing of lengthy forms). Other downsides include unmet training needs, technostress and threats to team relations among staff, and pressure on an already-strained material and technical infrastructure. We also demonstrated that while short-term inefficiencies are almost inevitable when introducing novel digital pathways and processes, significant longer-term increases in efficiency can be achieved with attention to strategic leadership and the work of embedding and adaptation, especially for non-clinical tasks and processes. The decision to revert wholly or partially to traditional in-person services may be strategic and warranted, but it may reflect unmet support needs for practices. We have produced a new typology of digital maturity based on our empirical findings.
Contribution to existing knowledge
Our findings affirm and extend previous studies which demonstrated that remote and digital modalities in general practice may be acceptable and even preferred by some patients,12,29,106 but that organisational transformation to such modalities is rarely unproblematic.107 Challenges include
- potential increases in workload and reductions in efficiency14–16,29,106 at a time when capacity is severely constrained;28,45
- complexification of staff roles and interactions, with increase in role stress;16,108 distortion of demand for patient-initiated consultations (leading to a mismatch between provision and need);4,16
- a shift towards more transactional care and loss of opportunities for preventive and long-term condition care;20,109,110
- reduced continuity of care;28,110
- documented27 or perceived16,29,106 safety challenges; and
- widening of inequities.10,13,16,20,106
Our finding that some patients may be completely unable to use key digital access routes because of physical, cognitive or mental health conditions aligns with recently published work by other groups.111,112 Other groups have also shown, like us, that inequities in digital health are especially dramatic for most disadvantaged patients because axes of disadvantage are often mutually reinforcing.
We have extended the existing evidence base by producing depth and detail in accounts of how British general practices coped with the disruptive innovation of remote and digital services from 2020 to the end of 2023. We have distinguished warranted variation (based on sound strategic choices) from variation that may reflect the need for additional support or resources. We have shown that while the negative impacts of digitalisation are common, they are not inevitable; they may not be permanent; and they do not affect all tasks and processes equally. We have also surfaced important issues for national policy-makers, Integrated Care Boards (ICBs), health boards and practices to address. These issues relate to safety, equity, continuity, non-acute services, training (including a comprehensive list of staff competencies – see Appendix 1, Table 482), workforce support, technologies and infrastructure, and patient information and support (including professionally designed patient-facing materials and resources).
Theoretically, we have made novel and important contributions to the theorisation of safety (showing how to combine a ‘Safety I’ and ‘Safety II’ approach30), equity (introducing a new theory of digital candidacy,63 adapting Archer’s theory of fractured reflexivity to address digital healthcare access,66 and extending theories of intersectionality to inform the production of digital personas71), continuity (producing a new ontology of continuity to reflect the realities of contemporary hybrid health care72,73), workforce (producing a novel theory of technostress, workplace suffering and strained team relations in the primary care context91) and infrastructure [extending sociotechnical theories of infrastructure to incorporate major disruptive innovation in an austerity context in the public sector (paper in preparation)].
Strengths and limitations
A significant strength of this study was the use of longitudinal ethnography in individual general practices, with a dedicated researcher for each practice. Using a purposively diverse sample of general practices drawn from across England, Scotland and Wales, we built relationships with clinical and non-clinical contacts, developed mutual trust, and visited regularly to collect data. The 28-month data collection period allowed us to observe the longer-term impacts of disruptive innovations introduced in the acute phase of the pandemic and to capture how ‘clunky’ and ‘inefficient’ technology-supported innovations could become more efficient over time through mutual adaptation and customisation of both the technology and the work process. Data analysis included extensive discussion among researchers to examine similarities and contrasts between practices, and member checking with practices.
Another strength was the adaptive use of additional data sources as the study unfolded. The 39 ‘elite’ stakeholder interviews and 4 multi-stakeholder workshops were part of the original study design, but as summarised in Appendix 1, Table 2, we later added 5 supplementary data sets to pursue key areas identified in the workshops as requiring further in-depth exploration: safety incidents including never-events relating to remote and digital care; the perspectives of clinical trainees and trainers; two data sets to address multiply disadvantaged patients (those with complex health and social care needs including homelessness and drug/alcohol abuse, and elderly first-generation immigrants with low digital literacy and limited English); and an examination of press coverage.
Limitations included that the study was limited to Great Britain (we did not have a practice in Northern Ireland); we had no practices in the north of England (though we did have some in Scotland); and patient interviews were relatively sparse. Also, while we managed to study a range of disadvantage, we did not capture the full range of disabilities and circumstances that can affect access (we did not, for instance, have an example of a patient with hearing or visual impairment, nor one where the patient was clinically extremely vulnerable and hence seeking to avoid in-person contact if possible for infection control reasons). While the study generated rich qualitative data which was useful in its own right, a larger study with a quantitative component could support formal hypothesis-testing (e.g. around the organisational-level determinants of digital innovation). A health economics component to such a study could allow firmer statements about efficiency.
Reviewers highlighted two classic papers on access to health care (both from North America113,114) which we might have referenced; both discuss patient factors, organisational factors, resources and appropriateness of services to need. To these, we would add a classic UK text on the important links between poverty and access to health care (Julian Tudor Hart’s ‘Inverse Care Law’ paper, which is written from a political economy perspective115) and Michael Marmot’s work demonstrating that such structural influences remain pervasive and often overlooked.116
Take-home messages
The study generated a number of important take-home messages.
The patient perspective: accessing and using remote and digital services Patients vary in their ability to use remote and digital services. Many patients already possess digital devices (e.g. smartphones) and skills to use them, or have family members who can help them access care. These individuals often find remote and digital services convenient and accessible for most contacts, most of the time. Some patients’ abilities to use remote and digital services are limited by sight or hearing impairment, neurodiversity or other conditions that make social interaction challenging, conditions affecting cognition (e.g. learning difficulties, memory loss, though both these conditions affect people in different ways and in varying levels of severity), mental health conditions, or low health, digital or system literacy. Notwithstanding the special needs of these groups, everyone might experience some difficulties in connecting at some times (e.g. when particularly distressed or concerned about their health or that of e.g. a child, or when working in a role that does not allow for ‘call backs’ during the working day). People with fluctuating physical or mental conditions vary from day to day in their ability to use digital access routes – hence a rigid coding of patients’ digital capability will not solve access challenges (rather, staff need the capability and flexibility to respond adaptively in real time).
Some low-income groups possess only basic technologies and limited data. Many patients who appear ‘digitally connected’ may struggle to use the full functionality of a digitalised primary care service as they possess only conventional mobile phones (or entry-level smartphones) and a ‘low-cost’ data package. The environment from which they seek access may not be conducive for private conversations. Low-income individuals may also be time-poor (e.g. they may work long hours, have long commutes and have limited negotiating power to gain time off for appointments) and some may struggle with the cognitive load117 of digital access routes. For all these reasons, they may value ‘low-tech’ remote services such as text messaging and telephone calls. Such messaging needs to be used carefully, avoiding extraneous information and clearly marked as coming from the patient's general practice (to avoid being misinterpreted and ignored as ‘cold call’ marketing).
Patients with multiple markers of disadvantage find remote and digital services hard or impossible to access. Disadvantage includes (in addition to the conditions mentioned in 1a above) extreme poverty, inadequate housing (or homelessness), precarious jobs (or unemployed), adverse social circumstances (including survivors of adverse childhood experiences), displacement (e.g. refugees), drug or alcohol misuse, and weak social and family networks. These axes of disadvantage play out differently in different individuals. In many people with multiple disadvantage, accessing care is extremely problematic. The addition of remote and digital options for such people can complicate the process of accessing care and increase the likelihood of disengagement. The needs of such patients may be best met with a traditional, walk-up-and-book/wait, in-person service. For some groups, specialist outreach services are required.
Human intermediation may be helpful but is not a panacea. ‘Digital navigators’ are people who help patients use digital access routes – for example, by helping patients complete an online digital-by-default consultation request. These staff are typically receptionists who have undertaken extra training. Some patients with low digital literacy (e.g. elderly people) may benefit from such human intermediation. However, there are many challenges to an intermediation service, including training and supervision, trust (some patients and staff are troubled by confidentiality issues), affordability (some patients require a great deal of human input and staff time is a limited resource), and the cognitive and emotional demands on the patient (who is usually required to answer a long list of questions in a structured format). Engaging with an intermediary requires additional interactions which some people find difficult both practically (locating an appropriate service) and interpersonally (e.g. if neurodiverse or upset). Digital services, even with intermediation, are unsuitable for some patients.
The quality challenge: benefits and trade-offs of digitalisation
In addition to the access and equity issues described above, a number of key messages relate to quality of care (including efficiency, effectiveness, safety, continuity and patient-centredness).
Digitalisation may produce inefficiencies as well as efficiencies. Potential inefficiencies include generation and circulation of large amounts of text and other material, double or triple handling (e.g. in which a single problem generates an e-consultation, a phone call and an in-person visit), loss of relational continuity (so that multiple staff members need to get to know the patient and the problem), a need for new or different supervision of roles, and an increase in task complexity. Ongoing adjustment of tasks and processes may (in some but not all cases) iron out some of these problems and turn an ‘inefficient’ process into a highly efficient one. The inefficiencies of digitalisation may compound those linked to the growing diversification of staff roles in general practice.48,49
Remote and digital interactions are socially very different from in-person ones. When the clinician–patient interaction occurs by telephone, video or asynchronous electronic communication or text message, there is less of a sense of ‘presence’ (the intersubjective awareness of the other person being close and attentive). In physically distant and asynchronous interactions, the human elements of traditional general practice (e.g. continuity of the therapeutic relationship, compassion, support) are more difficult. This does not matter much in more transactional consultations (e.g. seeking a repeat prescription for an uncomplicated condition) but may be crucial in others (e.g. especially when the patient has serious or complex clinical or social needs).
Safety incidents are rare in remote general practice, but they do occur. Staff tend to err on the side of caution and use creativity and ‘workarounds’ to make care safer; hence serious safety incidents (deaths or harms relating to e.g. missed or delayed diagnoses, failure to refer urgently or inadequate follow-up) relating to remote modalities are extremely rare. When safety incidents do occur, they may be partially attributed to excessive workload, interruptions and other sources of stress, and inadequate training among support staff. Safety in the clinical consultation can be compromised by inappropriate modality (e.g. offering a telephone call when the patient should be seen in person), poor rapport building, inadequate information gathering, limited clinical assessment, inappropriate pathway (e.g. staff member selects the wrong algorithm) and inadequate attention to social circumstances. Special care should be paid when patients have complex conditions, cardiac or abdominal emergencies, vague or generalised symptoms, safeguarding issues, failure to respond to previous treatment or difficulty communicating.
Continuity is changing. Traditional continuity of the GP–patient relationship is less common and harder to achieve as practices advance their digital services. In some practices, commitment to traditional continuity is a reason for cautious progress and even strategic de-digitalisation. More commonly, practices orient around an expanded definition of continuity which includes continuity of the illness episode (e.g. a clinician following up the test results on patients they have seen), continuity of distributed work (ensuring that all staff can access, and contribute to, an unfolding digital record of what has been done and what is planned for the patient), and continuity of commitment to the community served by the practice (e.g. in a Deep End practice, engagement with the community’s complex and ongoing needs and placing these centrally in any plans to digitalise – or not). As practices redesign their clinical pathways to accommodate digital technologies, it is essential to ‘design in’ relational continuity for patients with repeat attendances for still-undiagnosed problem and those with complex physical, social and or mental health problems. This includes developing ways to identify these patients and triage rules which allocate them to a GP or other clinician who knows them.
Efficiency-driven protocols for non-acute services may compromise care. In long-term condition monitoring, for example, tasks such as structured information-gathering can safely occur by text message or e-template for many patients. However, some key components of a check-up (e.g. diabetes foot examination) cannot be done adequately without an in-person assessment, and some patients may not disclose problems through a digital modality. As with most digital health technology, high-quality care requires an evidence-informed blend of technology and human input and designing care pathways with this in mind is an essential part of implementing and embedding digital care in general practice.
The staff perspective: training and workforce issues
Learning to provide digital services is more than learning how the technologies work. The knowledge needed to deliver high-quality remote services is complex, collective and organisationally embedded. Providing person-centred and safe care requires individuals and teams to be able to quickly assess patient capacity to use digital technologies and to use initiative and adapt to particular circumstances. Staff need to know how their input fits in with what other team members are contributing, and how to interact with them, hence on-the-job training and team training is essential.
Training needs are multiple and currently often unmet. Many staff have been taught to use a technology in a classroom or via didactic, self-completion online study modules, but lack confidence to use it in real workflows in a busy general practice. Clinical trainees may have low confidence and skills in telephone triage and consultations; trainers are concerned about the level of some trainees’ competence in the virtual environment. Staff turnover is sometimes high, and many staff work part-time, which makes it harder to extend training to all practice staff and co-ordinate team-based training.
Competencies for remote and digital services vary by staff group. Key competencies for clinical students and trainees include basic technical, triage, communication and information governance skills. Those for more senior clinicians include making complex judgements, working within the limitations of technologies and overseeing the distributed work of a multidisciplinary team.
Competencies for support staff include basic technical skills, system knowledge, safety-critical clinical knowledge for safe triaging and how to support patients in navigating services. This raises questions about whether and how to professionalise the triage support role (perhaps looking to the Netherlands for an example) with a longer and more standardised training and qualification for receptionist staff.
Competencies for those with a strategic role include a system-level understanding of how remote and digital services could help deliver the organisation’s mission, awareness of underserved populations and other complex need groups who may need bespoke arrangements, a sound grasp of workforce and training issues, and knowledge and skills to evaluate and monitor a dynamically unfolding digital service.
Digitalisation increases pressure on an already-stressed primary care workforce. In order to provide adaptive, person-centred, co-ordinated care in a high-pressure, safety-critical environment, healthcare staff must be able to do their job confidently, experience their work as fulfilling, feel safe to speak up and have positive and emotionally warm relationships with one another. Staff who are overworked, under-trained, under-supported and required to go beyond their scope of practice or work in ways they feel are causing harm or distress to patients are likely to become burnt out, distressed and unproductive.
The innovation and routinisation process in general practice
Most practices need considerable support and guidance to introduce digital technology safely, efficiently and strategically. Digital technologies introduced in the complex world of general practice are never plug-and-play. They require much embedding, in which work processes are adjusted to accommodate the technology and maximise the value generated. Fundamental redesign of tasks and processes is often needed, and this can take some time to optimise. In the current context of GP shortages, workforce pressures and burnout, few practices have the time, resources or skills to rethink care delivery, including consulting with patients about this. As noted above, staff need to learn to use technologies both independently and as part of a team, and to do this in an ongoing way as technologies evolve. Dialogue between technology users and designers may lead to key adaptations which make the technology more fit for purpose and transform the value proposition. Patients and carers also need to acquire key knowledge and skills.
One size does not fit all. UK general practices have responded in different ways to the strategic question of whether and how to continue, adapt, extend and abandon remote and digital services. Mostly, but not always, their decisions reflect the priorities and constraints of the local context, especially in relation to the communities they serve. Practices serving multiply disadvantaged communities may choose to prioritise in-person services (‘strategically traditional’). Some practices have reached the ability to select and assimilate technologies as part of a wider strategy to meet the needs of their population (‘digitally strategic’). A few practices are digitally advanced and innovative (‘trailblazers’). Practices which recognise the potential for digital technology to support and improve care often the material, technical, financial and human preconditions to get started (‘digitally hesitant’). Some are still experimenting with digital innovations in a somewhat unsystematic way, and mostly as a reaction to policy incentives or must-dos (‘digitally reactive’). Each of these archetypes needs a different approach to support but with current workload pressures on practices, additional ringfenced resources for digital transformation are essential if rapid progress is to be made (see Implications for decision-makers and Appendix 1, Table 3).
Larger, better-resourced, horizon-scanning practices usually innovate more readily. Practices meeting the preconditions for innovativeness (e.g. large size, well-resourced, high existing digital knowledge and capability, good team relations, well-networked with other practices, on the lookout for new solutions) generally find it easier to identify, procure, assimilate, embed and adapt new technologies. Smaller, more isolated practices may have the advantage of strong traditions and deep commitment to their local communities. Whatever the size of the practice, high levels of trust from patients and strong team relations seem to help innovation and vice versa.
Technologies and infrastructure
The context for digital innovation in general practice is adverse, with weak and fragmented national and local infrastructure, workforce shortages, demand–capacity misalignments and constraints on procurement. There is much policy emphasis on technological novelty. However, in general, there is less policy attention paid to patients’ and practices’ experiences and needs – for example, relating to issues of equity (the need for multiple routes of access), to infrastructural weaknesses, and to the costly and time-consuming process of embedding and adaptation. Standardised procurement protocols and top-down imposition of particular technologies or targets may restrict practices’ ability to address local needs flexibly and adaptively.
The range of technologies is wide and potentially confusing. A wide range of software applications is available, and many more products are in development, to support practices with access, triage and remote consulting. The usability and functionalities of these products, and their scope for integration into existing systems, vary. Identifying, procuring and integrating these products is challenging for practices. While some localities have good horizon-scanning and assessment mechanisms in place, allowing practices (individually or collectively) to identify and obtain new technologies that fit well with their vision and existing set-up, others struggle to navigate the market and identify and integrate appropriate products.
Practices need to build infrastructure as well as purchase new technologies. Practice infrastructure is critical and needs ongoing strategic investment and support. Some practices (‘digital trailblazer’ and ‘digitally strategic’ in our taxonomy) have sound existing material, human and technological infrastructure and are able to assimilate new technologies to augment and extend their existing platforms. This is challenging (and often unrecognised) work that involves incremental, step-by-step engagement with digitalisation, leveraging of people and resources (both often in short supply), and a learning-driven approach across the practice. The push to extend digitalisation to all practices will need significant support to build this digital infrastructure and to hold it all together.
The picture is dynamic. Procuring, adapting and embedding digital technologies will be an ongoing challenge as products evolve and new technologies become available. This is in part due to a diversity of suppliers and products and limited incentives for companies to design, develop or adapt platforms that bear practices’ routines and workflows in mind. There is no ‘one-size-fits-all’ technological solution. Rather, platforms that allow for ongoing adaptation (with practice staff working with the supplier) have far greater potential than those that that are inflexible. As artificial intelligence (AI)-enabled products become available, identifying and integrating such products into the busy and complex routines of general practice are likely to be challenging for all practices and (without additional support) impossible for some.
Reflections
See also Lessons learnt for future research.
Our original study design included a codesign component in 3 of our 12 practices, in which an independent design company attempted to work with patients and staff to try to optimise key access pathways using a popular design tool (the ‘double diamond’ method, which consists of four phases: discover – define – develop – deliver118). This approach did not produce workable solutions in any of the three practices. On reflection, this approach was ill-matched to the highly complex and dynamic challenges faced by participating practices. The approach assumed that a particular ‘design problem’ could be readily surfaced by an external design agency talking to staff and patients; that a solution to this problem could emerge with facilitated, creative thinking in a collaborative workshop format; and that this solution would be implementable, helpful and relatively enduring.
But as the empirical work described elsewhere58 has demonstrated, practices faced multiple and sometimes ‘wicked’ problems unfolding in dynamic and unpredictable ways across multiple interacting subsystems (e.g. technical, clinical, administrative, policy) in a context of severe resource constraints. As the RBD2 study unfolded, we moved from the ‘double diamond’ model to a more flexible and agile approach to design, in which a separate small design company was commissioned to respond rapidly to practices’ design needs as these emerged. We found that the major need in this space was for high-quality, accurate and appealing information and guidance for both practice staff and patients to explain new digital technologies, pathways and ways of working.
Also noteworthy is that we recruited one additional practice to avoid research fatigue as various PhD students joined the study team. This proved straightforward since all materials and processes were in place and it just required an amendment to our ethics approvals. The study design was emergent, with the intention to pursue key themes as they were identified. As it turned out, we collected five supplementary data sets (see Appendix 1, Table 2) to pursue additional research questions which emerged during our study. Again, this proved very straightforward because we had ‘designed in’ the possibility of doing this (though we did not identify in advance which data sets would be needed or what methods we would use). Additional funding was secured to examine processes of digital care navigation for people with social care needs, and also for several linked PhD projects as reported below.
We reflect on the challenges of involving patients and the lay public in the separate section below.
Regarding the implementation, embedding and post hoc adjustment of digital technology, we have closely observed the gradual adjustment of digital services and the impact of these adjustments on quality, safety, access, equity and staff and patient well-being. Alongside the considerable benefits and opportunities of digital technology in general practice, our observations highlight examples of inefficient and stressful implementation and the ways in which poorly designed clinical pathways embedded in practices that are intolerant of rule bending and flexible interpretation of practice policies by staff can create clinical risk and occasional safety incidents. These insights are important for the future design of clinical pathways and practice processes for safe and effective care as well as highlighting the need for some discretion in reception and clinical staff to act flexibly if their instincts alert them to clinical risk.
Engagement with partners and stakeholders
In the design, writing, delivery and dissemination of the RBD2 study, we used three main mechanisms to ensure a broad perspective from diverse stakeholders, including patients.
Firstly, we formed an External Advisory Group (EAG) consisting of a lay chair, senior policy-makers from NHS England and the Scottish Government, digital health providers, academics from outside our own institutions, and patients with lived experience. The EAG provided input into writing the grant proposal and regular links into their networks to share our emerging findings. Importantly, the EAG served as a constant ‘critical friend’, reviewing quarterly progress reports and (for each report) joining a 90-minute webinar discussion between the EAG and research team. Through these regular written reports and discussion meetings, we were able to articulate our progress, discuss evolving research findings and sense-check new research questions and ideas. The EAG was important in helping maintain focus in a rapidly evolving digital and policy environment.
Secondly, The Nuffield Trust (an independent think tank), our research collaborator, was integral in developing policy-related outputs from the start – that is identifying messages to target decision-makers at national, regional and local level. Their extensive networks of cross-sector health policy leads and decision-makers across the four nations were leveraged to bring key people together for the various dissemination events (described below). The Nuffield Trust have also created policy briefs and blogs of key study findings and distributed them through the same networks.
Thirdly, the four large, multi-stakeholder workshops, each on a different priority theme (access and equity, quality and safety, workforce and training, and technologies and infrastructure), required us to create and maintain a map of stakeholders across government, industry, third sector, NHS and patient groups. The design of the workshops, which included virtual breakout groups, allowed extensive networking and interaction across sectors. Post-workshop follow-up activity strengthened these links further. The relationships and alignments developed during and after the workshops are described elsewhere in this report.
Individual training and capacity-strengthening activities
The RBD2 study has built individual capacity in multiple areas within and beyond academia.
The study was large and lasted for almost 3 years (including writing up), allowing us to accommodate three PhD students who conducted focused studies on topics related to the core RBD2 study – relational continuity in the digital age (Emma Ladds), intersectionality of race/age/deprivation in explaining access challenges in ethnic and racial minorities (Laiba Husain), and staff working conditions, well-being, and relational work (Francesca Dakin). These PhDs were made possible by additional funding leveraged on the strength of the RBD2 study, from (respectively) Wellcome Trust, THIS Institute and NIHR School of Primary Care Research (SPCR). All three PhD students have already achieved several first-author publications and presented their work at international conferences; at the time of writing, two have now passed their PhDs and secured postdoctoral positions at leading universities.
We also accommodated four clinical pre-doctoral GP researchers under the NIHR In-Practice Fellow Scheme (Rebecca Payne and Ellen MacIver), NIHR Pre-doctoral Fellowship (Asli Kalin) and a John Monash scholarship Fellowship (Isabel Hanson). Rebecca Payne led two major substudies, leading to two first-author publications in leading journals. She successfully applied for a PhD (DPhil) place to take forward her work in digital innovation and was awarded a prestigious Clarendon Scholarship from the University of Oxford. She has also been awarded a Churchill scholarship to visit digital primary care services in remote Alaska and Australia. Ellen MacIver, who joined a year later, contributed substantially to various papers and earned authorship. Asli Kalin was a researcher-in-residence on RBD2 and has built her CV with numerous coauthored papers, with a view to applying for PhD following a career break. Isabel Hanson led a substudy on trainees’ learning needs using supplementary data set 1 and gained coauthorship on two papers, including one paper led by trainees and written for trainees, and published in a trainee journal.119 Isabel Hanson undertook a study of GP trainees’ confidence in remote consultations, collecting supplementary data set 1 for her MSc dissertation, earning authorship on several papers and gaining a DPhil place at the University of Oxford.
Early-career academics also developed significantly. Post-doctoral researchers included Sarah Rybczynska-Bunt and Joseph Wherton, both of whom have published first-author papers in leading journals; other postdocs (including Sarah O'Rourke and Stuart Faulkner) have coauthored outputs and are working on further publications. During the course of the study, Gemma Hughes secured a tenured Associate Professorship at the University of Leicester and Sara E Shaw was promoted from Associate to Full Professor at the University of Oxford.
More informally, several clinical academics (Aileen Clarke, Sara E Shaw, Emma Ladds, Richard Byng, Asli Kalin, Rebecca Rosen, Rebecca Payne) obtained exposure to interdisciplinary training opportunities with social scientists; and, conversely, social scientists (Francesca Dakin, Laiba Husain, Joseph Wherton, Sara E Shaw, Sarah O'Rourke, Sarah Rybczynska-Bunt) benefited from working closely with clinical academics (Trisha Greenhalgh and Lucy Moore are dual qualified in social sciences and medicine/nursing).
Two attached medical students (Kai Song and Molly Hey) conducted a media analysis of how remote and digital general practice was covered in newspapers and gained authorship on a paper published in a leading journal. A further medical student (Tabitha Pring) supported data collection and analysis on the quality workstream, contributing to the quality paper. In-Practice Fellows gained experience supervising these students on an academic project. A lived-experience researcher (Nadia Swann) contributed to the safety paper and wrote the patient advice section for that, helping build her CV for a PhD application (which was successful).
People with lived experience of remote general practice gained research experience and were welcomed as partners in the collection and analysis of data and writing up of findings. One individual with lived experience of a safety near-miss (Nadia Swann) undertook data collection and analysis for the safety paper,30 and wrote the patient advice section of that paper.
Institutional capacity strengthening
This study was a partnership between a large team based at the Interdisciplinary Research In Health Sciences (IRIHS) research group at the University of Oxford and smaller teams based at the University of Plymouth and the policy-focused think tank, Nuffield Trust. Through this collaboration, both academic departments gained insights into policy analysis and policy-focused impact work from working with the Nuffield Trust and vice versa.
The study was led from the IRIHS research unit at the Nuffield Department of Primary Care Health Sciences, University of Oxford. IRIHS seeks to use creative interdisciplinary methods to bring both clinical and social science insights to bear on applied health research. The RBD2 grant provided significant resource for the IRIHS team over almost 3 years (including a no-cost extension). In addition to supporting individual career development, this grant enabled IRIHS to consolidate and expand a diverse and talented team which included clinicians (Trisha Greenhalgh, Aileen Clarke, Sara E Shaw, Isabel Hanson, Rebecca Payne, Ellen MacIver, Asli Kalin, Lucy Moore, Emma Ladds), anthropologist (Francesca Dakin), psychologist (Laiba Husain), human–computer interaction scholar (Joseph Wherton), sociologists (Sara E Shaw, as well as Trisha Greenhalgh and Lucy Moore who are dual qualified) and people with lived experience (Nadia Swann).
The RBD2 study supported a small team led by Richard Byng within the University of Plymouth’s Community and Primary Care Research Group. The collaboration provided opportunities to share expertise in addressing inequalities for marginalised populations and has extended the group’s experience in social sciences and digital health. It has provided valuable experience for two postdoctoral researchers and developed new partnerships.