Advantages and limitations of virtual online consultations in a NHS acute trust: the VOCAL mixed-methods study

Article history

The research reported in this issue of the journal was funded by the HS&DR programme or one of its preceding programmes as project number 13/59/26. The contractual start date was in March 2015. The final report began editorial review in July 2017 and was accepted for publication in November 2017. The authors have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The HS&DR editors and production house have tried to ensure the accuracy of the authors’ report and would like to thank the reviewers for their constructive comments on the final report document. However, they do not accept liability for damages or losses arising from material published in this report.

Declared competing interests of authors

none

Copyright statement

© Queen’s Printer and Controller of HMSO 2018. This work was produced by Shaw et al. under the terms of a commissioning contract issued by the Secretary of State for Health and Social Care. This issue may be freely reproduced for the purposes of private research and study and extracts (or indeed, the full report) may be included in professional journals provided that suitable acknowledgement is made and the reproduction is not associated with any form of advertising. Applications for commercial reproduction should be addressed to: NIHR Journals Library, National Institute for Health Research, Evaluation, Trials and Studies Coordinating Centre, Alpha House, University of Southampton Science Park, Southampton SO16 7NS, UK.

2018 Queen’s Printer and Controller of HMSO

Introduction to the research

Health services face rising costs as a result of increasing disease prevalence, high ‘did not attend’ (DNA) rates and poor patient engagement, resulting in poor health outcomes and greater use of emergency care. 1,2 Most outpatient models fail to reliably provide responsive care when patients need intervention. Reducing hospital follow-up appointments is a priority in the NHS. Unless new ways of delivering care are found, chronic disease management will be unaffordable and undeliverable. Current policy places considerable faith in digital technologies and their potential to deliver more efficient, effective, patient-centric care in the community. 3–6 The UK’s National Information Board (see the glossary in Appendix 1 for a description of national organisations and initiatives) has argued that, in order to respond effectively to these demographic and epidemiological trends, we need a different kind of health service, in which the traditional outpatient consultation, for example, will become increasingly obsolete. 5 Technology-supported consulting is viewed as at least a partial solution to the current challenges of delivering health care.

Digital technology plays a significant (although varied) role in local plans across the NHS to reconfigure hospital services and transform the delivery of health services. 7 Attending regular clinics can be expensive, physically challenging and inconvenient for patients. Virtual consultations (using Skype or similar media) have the potential to fundamentally change the way in which patients interact with clinicians.

Following proof of concept work with NHS Choices, we were funded by the Health Foundation8,9 to explore the scope and feasibility of video outpatient consultations in the diabetes clinic in Newham. The Diabetes Appointments via Webcam in Newham (DAWN) project documented 62% acceptance rates across all ages, a reduction in DNA (from 40% to 23% at the end of year 1) and small efficiency savings of £63K. 8,9 Our subsequent project on Diabetes Review, Education And Management via Skype (DREAMS) was also funded by the Health Foundation (2012–14), and explored Skype-supported video consultations in ‘hard-to-reach’ patients with diabetes mellitus. 10 As well as improved clinical outcomes, DREAMS showed improved engagement and better self-management among regular users. 11

The online environment is known to produce subtle alterations in the dynamics of human interaction, with a potential risk that clinical clues will be missed or the clinician–patient dynamic will be altered adversely. 12 As a new service model, it also brings operational and cultural challenges, including training and supporting staff, as well as patients, in using digital technologies. For this reason, we sought to undertake further in-depth research.

Aims and objectives

As set out in our protocol published previously,13 the VOCAL (Virtual Online Consultations – Advantages and Limitations) study aimed to define good practice and inform its implementation in relation to clinician–patient consultations via Skype and similar media, addressing three key objectives:

1. at the macro level, to build relationships with key stakeholders nationally and identify from their perspective how to overcome policy and legal barriers to the introduction of remote consultations as a regular service option

2. at the meso level, to illuminate and explore the sociotechnical microsystem that supports the remote consultation, thereby identifying how organisations can best support the introduction and sustainability of this service model in areas where it proves to be acceptable and effective

3. at the micro level, to study the clinician–patient interaction in a maximum variety sample of 30–45 remote outpatient consultations in two clinical areas; in particular, to highlight examples of good communicative practice, to identify and characterise examples of suboptimal communicative practice and to propose approaches for minimising the latter.

The study addressed the following research questions:

1. At the macro level: what is the national-level context for the introduction of virtual consultations in NHS organisations and what measures might incentivise and make these easier?

2. At the meso level: how is a successful virtual consultation achieved in an organisation in which the processes and systems are mostly oriented to more traditional consultations?

3. At the micro level: what defines ‘quality’ in a virtual consultation and what are the barriers to achieving this?

In the following sections, we outline how our research contributes to the NIHR brief and describe the current evidence base relating to virtual consultations. The VOCAL study was designed as a multilevel study capturing macro-level barriers and incentives to, and facilitators of, supporting virtual consultations (objective 1), meso-level administrative and clinical processes that need to change to embed online consultations (objective 2) and micro-level details of the interactions in consultations (objective 3). As this included a significant research component focused on the national-level context, we set out the detail of the wider landscape in which virtual consultations are evolving in Chapter 3.

The National Institute for Health Research Health Services and Delivery Research commissioning brief

The VOCAL study was funded in response to a call from the NIHR Health Services and Delivery Research (HSDR) programme for ‘assessing alternatives to face-to-face contact with patients’ (HSDR call number 13/59). The call emphasised the rapid increase in new forms of interaction between patients and health professionals, including consultations by Skype, e-mail and webcam, and the potential that these new forms of contact hold for improving the quality and experience of patient care and securing cost-savings. The two themes were:

1. primary research evaluating the cost-effectiveness of non-face-to-face contact with health professionals

2. qualitative research on the impact of new forms on the clinician–patient dynamic, and appropriateness for hard-to-reach groups.

Our study responds directly to the second theme, which offered a unique opportunity to build on our previous work in setting up Skype consultations for the diabetes clinic at Newham. 8–11 Guided by the brief, we sought to examine in depth any changes in the dynamics of the clinician–patient interaction and communication in virtual consultations in three clinics spread across two clinical areas (the Adult/Young Adult Diabetes and antenatal Diabetes clinics in the Diabetes service, and the Hepatobiliary And Pancreatic Cancer Surgery clinic in Cancer Surgery, enabling comparison across different conditions, clinics and teams), the appropriateness and satisfaction of staff and different patient groups using virtual consultations (including young adults, older people and those from ethnic minority groups) and the ways in which wider organisational- and national-level environments shape virtual consultation services.

In the following sections, we review the current evidence generally on alternatives to face-to-face consultations, and specifically on virtual consultations, and provide an overview of the strengths and weaknesses of different methodological approaches.

The development of alternatives to face-to-face consultations

New technologies that support alternatives to face-to-face consulting are often seen by decision-makers as potentially providing more flexible and convenient ways for patients to interact with health professionals, while at the same time improving financial efficiency and the clinical effectiveness of services. 2,14,15 However, the use of alternatives to face-to-face contact with patients is a relatively new area of activity in health care, with limited evidence currently available to inform decisions about how best to develop and use a range of technologies. As well as video, these technologies include telephone, text messaging, e-mail consultations, online portals, telemedicine and telehealth. We summarise the evidence on each of these below.

Evidence relating to the use of virtual consultations

The evidence base on virtual consultations (using Skype or similar technology) has been steadily accumulating. 88–98 A number of published studies focus broadly on video or remote consulting (e.g. exploring issues of usability and acceptability). 45,99–113 Many focus specifically on the clinical use of Skype,91,114–123 either on its own or in combination with other technologies [e.g. WhatsApp (WhatsApp Inc., Mountain View, CA, USA) messaging124 or FaceTime125,126]. A handful of papers examine the perceived ethical, legal or technical issues relating to virtual consultations. 127,128 Studies generally report positive benefits, particularly in terms of patient satisfaction and increased accessibility. However, most are brief descriptions of small, pilot-stage projects (some with as few as five patients), or randomised controlled trials (RCTs) offering one or more virtual consultations compared with traditional face-to-face contact, often with limited follow-up.

Below, we extend the literature review that we conducted at the start of the VOCAL study,13 by reviewing the higher-quality primary studies from a 2015 review88 of 27 published studies of the use of Skype, which reported largely positive benefits. We focus on studies from the review that are most relevant to the VOCAL study, along with some additional studies published since 2015. In doing so, we have adapted and extended the literature review that was previously published in our study protocol in BMJ Open. 13

A number of studies have focused on the use of virtual consultations for the treatment of chronic diseases. A study of family based behavioural support for adolescents with poorly controlled type 1 diabetes focused on the ‘working alliance’ (the strength of the relationships between patients, caregivers and health-care professionals). 129 Findings showed that 10 sessions via Skype were as effective at preserving the working alliance as 10 face-to-face sessions. 114 Adherence to treatment and glycaemic control were also similar. 91 However, losses to follow-up were high: of the 47 (out of 92) participants randomised to Skype sessions, follow-up data were available for only 32.

In our work in Newham, we introduced virtual consultations for Diabetes in 2011, with 480 remote consultations documented in 104 patients between 2011 and 2014. 10,11 Findings showed that virtual consultations were popular with both patients (especially young adults) and staff. In patients who chose to use the remote service, it appeared to be associated with increased engagement (overall DNA rates were 13% in patients accepting the Skype option and 28% in those who chose not to use this option, although denominator populations for these figures were self-selecting and hence not strictly comparable), improved glycaemic control (the average glycated haemoglobin level pre and post introduction of remote consulting was 70 mmol/l and 65 mmol/l, respectively, for those who used the service) and fewer accident and emergency (A&E) attendances than those not using the remote service (raw data on this were statistically significant, although numbers were small). These figures were encouraging; however, patients were not randomised and there were multiple potential confounders, and 45 patients who initially signed up to the remote service subsequently withdrew from it.

A recently published RCT130 showed similarly positive results for patients with type 2 diabetes who were not responding to usual care. A total of 102 participants (n = 165) were randomised to either monthly video conferences with a nurse via tablet computer over 32 weeks or usual clinic-based care. Allocation to the trial group required participants to regularly self-monitor blood glucose, blood pressure and weight, and to upload results so that these were available to both the nurse and patient during video consultations. Authors reported a significant improvement in glycaemic control in the virtual consultation arm compared with those receiving clinic-based care. However, these differences were no longer significant at 6 months, and authors were unable to tease out whether it was the video conference per se, the effect of immediate response to increased measurements, increased personal contact or a combination that led to the initial improvement.

In a study of the management of depression in older housebound adults, participants were randomised to receive in-person problem-solving therapy, Skype-delivered problem-solving therapy or a weekly telephone call with no therapeutic content. 92 Both the in-person and Skype-delivered therapy were effective at reducing depression scores and disability outcomes. However, at the 36-week follow-up, the participants in the Skype arm experienced significantly better outcomes than those receiving in-person therapy. The authors ventured that the more focused nature of the Skype sessions may be responsible for sustained benefits.

A number of studies have examined the use of virtual consultations for clinical follow-up, either as an alternative to or an addition to face-to-face consultations. A 2014 study reported on the use of Skype for orthopaedic follow-up. 94 The Skype service was offered to 78 patients, following total joint arthroplasty. Participants were invited to consult with their surgeon via Skype in addition to scheduled follow-up appointments at 1, 3, 4, 6 and 9 weeks. Just under half of the participants (n = 34) underwent at least one Skype consultation. The remainder (n = 44) did not have appropriate electronic devices or internet connection to use the Skype service. There was no significant difference in clinical outcomes for the users and non-users of this service (however, the study was probably underpowered). However, those participants followed up using the Skype service had fewer unscheduled in-clinic visits. Those using the Skype service rated their postoperative satisfaction as higher than those who were not using it. A key finding in a follow-on paper with 228 participants was that time spent on the consultation, and patient-borne costs, were lower in the Skype group. 96 A linked economic evaluation showed that service costs were also significantly lower in the Skype group. 95 Although no issues were missed in patients in the trial, a subsequent commentary queried whether or not remote assessment might be less safe. 97

Skype was used to deliver follow-up training for ‘pursed lips breathing’ (a technique used to manage breathlessness in chronic obstructive pulmonary disease). An initial study with 16 participants reported that those who received the follow-up sessions had better breathlessness management than those with basic training alone. 115 A subsequent small-scale trial confirmed these findings. 116

A small trial (n = 55) comparing remote ‘video visits’ during follow-up after prostate cancer surgery with usual care found video visits to be ‘equivalent in efficiency’ to conventional outpatient visits, as measured by the amount of time spent face to face, patient wait time and total time devoted to care. 99 There were no significant differences in patient perception of visit confidentiality, efficiency, education quality or overall satisfaction. Video visits incurred lower patient-borne costs and were associated with similar levels of urologist satisfaction to conventional outpatient visits. Other studies have shown similar levels of patient satisfaction, particularly in terms of time saved (as a result of not needing to travel to clinics) and costs saved. 101,104,107

In a randomised trial of Skype sessions versus standard home care in supporting families with premature infants, the nine families randomised to Skype sessions reported very positive experiences. All found the technology easy to use, noting that virtual consultations were better than telephone calls. 131 Tellingly, the authors commented that ‘The families readily embraced the use of ICT [information, communication and technology], whereas motivating some of the nurses to accept and use ICT was a major challenge’,131 raising questions about the ways in which virtual media might be embedded in clinical work.

Virtual clinics via Skype have also been used for counselling and mental health consultations. Skype proved to be an effective medium for supporting independence and self-confidence among those aged 12–18 years with spina bifida. 98 In a 15-minute consultation once per week, the nurse supported patients with continence care and self-care. Participants reported that they felt more confident talking about personal issues via Skype than face to face. They also valued the privacy that consultations via Skype allowed (e.g. enabling young people to speak about their care from a private space at home rather than in a face-to-face consultation with a parent or carer), and increased the accessibility of the service (particularly for those patients with complex physical needs).

Issues surrounding privacy, security and reimbursement for virtual consultation services are the topic of significant public and professional debate. Despite this, and often being mentioned in passing in discussion sections of studies,108 such issues have rarely been systematically explored. 127 Technical difficulties are also typically mentioned in passing, but are rarely explored in any depth. Studies beyond the medical literature have shown that Skype is often ‘laggy’ (e.g. audio and video data can become unsynchronised). However, in one study that looked at the effect of collaborative song-writing as therapy, some participants reported that this lag could be helpful, making them select words with care and focusing them more on the interaction. 117 There are times when Skype compresses the video, so that facial expressions are hard to interpret. 117 It may be that the quality of hardware or bandwidth is critical to some (although perhaps not all) kinds of clinical consultation. 103,128

In summary, the research literature on virtual consultations remains sparse. The contribution of virtual media to consultations in health care has been studied mainly by using experimental methods (especially RCTs), but with no adequately powered randomised trials and only a handful of controlled before-and-after studies conducted to date. These studies have generally focused on evaluating the outcomes of the technology. To date, there have been no rigorous and theoretically grounded qualitative or mixed-methods studies of the kind undertaken in the VOCAL study.

Overall, the literature suggests that there is great potential for the use of virtual media tools, such as Skype, for virtual communication between patients and clinicians. Although the studies reviewed are broadly positive, the small sample sizes, select nature of samples and high losses to follow-up call into question any unqualified conclusion that the technology is ‘effective’, and the lack of negative studies raises questions about potential publication bias. For the VOCAL study, current evidence raises questions about the relevance of video-based consultations for different clinical conditions and patients, the accessibility and acceptability of Skype and other virtual media to both patients and (clinical and non-clinical) staff, the ways in which practical and technical issues (e.g. the availability of smart technology, delays in data transfer) shape virtual consultations and how ethical, legal, regulatory and payment issues shape the adoption of virtual media. In addition, none of the studies reviewed examined the detail of interaction and what gets either added in or left out of consultations when they take place virtually, raising the question for the VOCAL study about what a good-quality interaction means in the context of a virtual consultation.

Although the RCT is widely viewed as a gold-standard design for testing the efficacy of an intervention experimentally, the trials that have been undertaken on remote consultations have provided few or no data on the organisational complexities of implementing a radically new technology-based service, and they cannot address the question of how video consultation services emerge and become embedded in a real-world setting (i.e. outside the specific confines of a randomised trial and with a view to long-term sustainability). To fill this gap in the literature, studies are needed to explore the emergence of video consultation services naturalistically (i.e. by capturing real-world quantitative and qualitative data on the emerging services and documenting the challenges faced).

Study design

In contrast to much of the published research literature described in the previous chapter, which has tended to compare ‘technology on’ with ‘technology off’ in an experimental or quasi-experimental design, the design of our study was oriented to teasing out the (often subtle) social and material interactions occurring between patient, staff member and technology(ies). Although experimental studies have their place, technology-focused approaches are crude and deterministic. 81,132–136 In-depth qualitative studies can reveal how individual identity, experience, expectations and material skill might shape and alter these technology-mediated interactions and make them more (or less) efficient and effective. 137

We conducted a multilevel mixed-methods study of virtual consultations with macro-, meso- and micro-level components. At the macro level, we combined interviews with 12 national policy-makers and other key stakeholders with analysis of national-level policy documents in order to explore barriers and incentives to, and facilitators of, supporting virtual consultations. At the meso level, we mapped the administrative and clinical processes that needed to change to embed online consultations [e.g. changes to clinical pathways, changes to staff roles, rethinking the use of traditional outpatient space, updating information governance (IG)]. At the micro level, we studied interactional dynamics by generating a multimodal data set (audio transcript, video and computer screen capture) of 30 virtual consultations in Diabetes and Cancer Surgery, alongside a matched data set of 17 (audio-recorded) face-to-face consultations.

Changes to the study protocol

In our original study protocol, we proposed collecting 30–45 virtual consultations between patients and clinicians in Diabetes and Cancer Surgery (10–15 in Cancer Surgery and 20–30 in Diabetes) and then, using fine-tuned linguistic techniques, examining the detail of the interaction. As we began to analyse these micro-level data and examine the interactional dynamics, we quickly realised that it would be helpful to compare these with usual care (i.e. traditional face-to-face consultations) and examine if/where any interactional differences occur. We therefore adapted our study design and protocol to incorporate a comparable sample of face-to-face consultations in Diabetes and Cancer Surgery, audio-recorded to enable analysis of the type of talk that takes place. We used the Roter interaction analysis system (RIAS) to compare different categories of talk in face-to-face and virtual consultations. As we found few significant differences in communication practices (see Chapters 5 and 6), we limited our proposals to minimise suboptimal communicative practices (objective 3) in the guidance for patients and providers on using Skype and similar virtual media for remote consultations (see Practitioner Resources on the NIHR Journals Library website: www.journalslibrary.nihr.ac.uk/programmes/hsdr/135926/#/).

Theoretical approach

Our study drew on strong structuration theory (SST),138,139 which sees society (through rules, values and norms) as profoundly shaping human behaviour, and human behaviour (through peoples’ interpretations and choices, as well as technology-mediated actions), in turn, changing society.

As we have set out in previous work,67,139,140 structuration theory links the macro of the social environment (social structures) with the micro of human action (agency) and considers how the relationship between structure and agency changes over time. 141 The structure–agency link is mediated through ‘scripts’ (patterns of behaviour and interaction in social settings, including the adoption and adaptation of particular technologies), which gradually change over time. 137 Scripts link to organisational routines and, hence, to the potential for innovations to become embedded and routinised in everyday practice. 142

Strong structuration theory proposes that external social structures (social norms, rules and so on) are mediated largely through position-practices (defined as a social position and associated identity and practices), together with the network of social relations that recognise and support it (‘position-practice relations’, of which the clinician–patient relationship is a good example). It also sees human agency as crucial to engaging with technologies. 143 Within the VOCAL study, SST therefore offered the potential to theorise human characteristics, such as identity and social role (e.g. what it means to be a ‘clinician’, ‘carer’ or ‘patient’), interpersonal relationships (e.g. the clinician–patient relationship), situational knowledge (e.g. patient expectations of a consultation) and the capabilities needed to operate technology.

Four components of SST guided our thinking and analysis of virtual consultations in the VOCAL study: external structures, internal structures, actions and outcomes (Figure 1).

FIGURE 1.

Strong structuration theory, adapted to encompass a technology dimension. Reprinted from Social Science & Medicine, Volume 70, Issue 9, Greenhalgh T and Stones R, Theorising big IT programmes in healthcare: Strong structuration theory meets actor-network theory, 1285–1294, Copyright 2010, with permission from Elsevier. 139

Reprinted from Social Science & Medicine, Volume 70, Issue 9, Greenhalgh T and Stones R, Theorising big IT programmes in healthcare: Strong structuration theory meets actor-network theory, 1285–1294, Copyright 2010, with permission from Elsevier

The first component of SST, ‘external structures’, refers to the position-practice relations that characterise the meso (organisational) and macro (national-level policy and decision-making) levels, and that change over time (e.g. the ways in which the medical profession is, arguably, less trusted in the face of media reporting about errors).

The second component of SST, ‘internal structures’, refers to the representations of society that we all have in our heads. They include general dispositions (e.g. discourses and world views, moral principles, attitudes, technical and other embodied skills and personal values – what Bourdieu called ‘habitus’144), as well as specific knowledge of an aspect of the world and how one is expected to act within it (e.g. a nurse’s understanding of what is involved in supporting patients with their self-management of diabetes).

The third component of SST is ‘activation/active agency’. We used ethnography to study specific examples of interactions – what SST calls ‘conjunctures’ (the medical consultation is a good example) – to capture how people play out their position-practice relations, behaving in a way that they believe is appropriate and responding in a moment-by-moment way to the other party(ies). To study the agency (i.e. human intention) behind these actions, SST incorporates theories from phenomenology (the study of people’s shifting fields and horizons of action arising from the focused activity at hand145), ethnomethodology (the study of how one person responds, moment by moment, to the talk and action of another146) and symbolic interactionism (the study of the subjective meaning and interpretation of human behaviour147).

The health-care setting is heavily institutionalised, and behaviour is often ritualised (i.e. we know, and play out, the roles expected of us as doctors, patients and so on). Behaviour in the consultation is strongly influenced by such things as regulations and other governance measures, norms, beliefs, professional and lay codes of practice and deeply held traditions (all of which are embodied and reproduced by human agents, including clinicians, administrators and patients), rather than exclusively by business concerns, such as efficiency and profit. A person’s knowledge of these institutional structures (in SST terms, the ‘strategic terrain’) may be more or less accurate and more or less adequate. A good example of this might be the older patient who retains the perception that it would be rude to offer suggestions to the doctor, whereas, in reality, the doctor is keen to promote shared decision-making.

The fourth component of SST is outcomes (see Figure 1). The outcome of human action in the consultation may be intended or unintended, and will feed back on external and internal structures – either preserving them or changing them as they are enacted. A good example of this in our study is whether or not a virtual consultation that is experienced positively will increase the likelihood that the patient will adhere to treatment and attend the next consultation (in person or virtually).

The clinical consultation is a social encounter shaped by social and institutional forces. For instance, clinicians resist technologies that (in their opinion) interfere with good clinical practice and the exercise of professional judgement,67 consulting patients will be more or less sick and have socioculturally shaped expectations of being cared for and comforted and their circumstances and/or illness may affect their ability to use the technology (e.g. those consulting virtually may be reliant on a relative or carer at home to access and use the technology). SST enabled us to focus on how bodily, emotional and cognitive functions interact with an individual’s disposition, symbolic interpretation and (imperfect) knowledge to affect how the consultation unfolds, and to consider the ways in which wider organisational, institutional and regulatory environments shape interactions in and around virtual consultations.

Action research

Our interests lay in studying the ways in which virtual consultations did (and did not) become embedded in the work of Diabetes and Cancer Surgery clinics. We were therefore keen to work with local managers and commissioners to understand the organisational change required to embed (and potentially spread) the virtual consultation option. Our study was therefore informed by the principles of action research. 148,149

Action research has been described as ‘a mutual learning process within which people work together to discover what the issues are, why they exist, and how they might be addressed’. 150 The idea is that practitioners and researchers work together to identify and seek to address issues as they arise in the context of research, and, in this case, in the development of a virtual consultation service. This meant that we keenly responded to requests for input (e.g. regarding issues with loading and updating Skype on clinic computers or plans to spread virtual consultations beyond Diabetes and Cancer Surgery clinics), and judiciously fed back emerging findings to local- and national-level decision-makers whose work was concerned with developing or spreading virtual consultations (e.g. relating to national payment systems).

Below, we summarise the main focus of our action research-related activities across the different levels of the study. Given that a virtual consultation service had already evolved in the Adult/Young Adult Diabetes clinic prior to the start of the study and the focus of Barts Health NHS Trust was on considering the potential for roll-out elsewhere within the organisation, much of our focus was at the organisational level, which appeared to be integral to the ongoing and measured development of virtual consultation services. Although all of our activities were broadly guided by the action research cycle (plan, act, observe, reflect148) and aimed to feed into service development, our contact with local and national stakeholders provided opportunities for increased insight and our extended fieldwork enabled insights into the complexities of relevant policies, organisational and decision-making processes that would not have been available to us through standard interviews or observations alone.

Project management and governance

The study was delivered by a core working group (TG, SV, JW, JM and SS), supported by a 6-monthly independent steering group and a patient advisory group (PAG; see Patient and public involvement). The steering group had a lay chairperson and cross-sector stakeholder representation, including patients, NHS stakeholders and national-level decision-makers (see Appendix 2).

The study received ethics approval from City Road and Hampstead NHS Research Ethics Committee on 9 December 2014 (reference number 14/LO/1883).

In line with changes to the VOCAL study protocol (see Changes to the study protocol), the following substantial amendments were sought and approved:

1. substantial amendment 1: audio-recording face-to-face consultations for comparison with Skype consultations; approved on 23 February 2016

2. substantial amendment 2: sharing selected video recordings (with patient consent) with technology developers to inform and improve the design of remote consulting technology; approved on 1 December 2016.

Patient and public involvement

Patients and their carers have been key to the VOCAL project and, in fact, the original impetus (in 2011) to use virtual consultations was initiated by service users (many from deprived backgrounds and/or minority ethnic groups) within the Diabetes clinic in Newham. At that time, we sought patient feedback as an integral part of both service development and ongoing evaluation (e.g. from focus groups, in-depth telephone surveys and online questionnaires, to refine the intervention in the Diabetes clinic10).

We set up a dedicated PAG in 2015. The main purpose of the PAG was to continue to incorporate patient feedback within our work and help to capture patients’ experiences of both the research process and the proposed virtual consultation services. The group was facilitated by Anna Collard, who has a background in community anthropology. The intention, set out in its terms of reference, was that the PAG would provide advice and feedback to the VOCAL research team and steering group. The intention was that the VOCAL research team and steering group would inform the PAG about research findings as they went along. The focus of the group was on the interactions between the clinician and the patient in Skype consultations, and not on any one particular condition (i.e. either diabetes or cancer). Patients from the PAG were also asked to review key documents, such as patient information leaflets.

Setting and context

We have been working for several years with the front-line clinical team in the Diabetes clinic to develop virtual consulting as part of business as usual in Barts Health NHS Trust, the UK’s largest acute trust (formed in 2012 when three trusts in different boroughs merged).

We studied three clinics on different sites: Diabetes (Adult/Young Adult), based at Newham hospital and with a community outreach clinic in a local general practitioner (GP) surgery (Shrewsbury Road), the Antenatal Diabetes service based at Mile End Hospital and the Hepatobiliary and Pancreatic Cancer Surgery service at the Royal London Hospital. These sites are located across adjacent London boroughs (Newham and Tower Hamlets), both of which are characterised by a high level of socioeconomic deprivation, ethnic and linguistic diversity, and a high burden of disease. 151,152 Like many acute trusts, Barts Health NHS Trust is under pressure to deliver services more cost-effectively, while responding to rising need and demand. 153,154

At the time of our study, the national health-care landscape was characterised by significant financial pressure, with NHS organisations struggling in the face of constrained budgets and growing demand, and technology being seen as a logical route towards achieving cost-savings and increasing quality. This national picture was mirrored in Barts Health NHS Trust, where remote consultation services in the Diabetes clinic and Cancer Surgery evolved in the context of considerable financial, organisational and staff pressures. 153,155 Barts Health NHS Trust serves a population of around 2.5 million in east London, with around 2000 clinics across eight different locations, over 15,000 staff and an annual turnover of £1.4B. In 2015, Barts Health NHS Trust was rated as ‘inadequate’ by the Care Quality Commission (the independent regulator of health and social care in England; see Appendix 1), with significant concerns reported in safety, effectiveness and responsiveness, and with the leadership of the trust. 156 The trust was also put into ‘special measures’ (a set of measures applied to NHS bodies with a view to resetting expectations of financial discipline and performance) in September 2016, following substantial and mounting financial deficits. 157 It is against this background that Barts Health NHS Trust set out a significant programme of improvement, both within the trust and with the relevant health and social care agencies (commissioners and providers), called Transforming Services Together: Strategy and Investment Case (TST). 158 The focus of TST is on radically changing the way in which services are designed and delivered, and this has fed into subsequent sustainability and transformation plans (see Appendix 1 for an overview). This includes the redesign of outpatient pathways, enabling quicker access to specialist advice, both virtually and face to face. As part of this work, and in line with the national-level impetus for technology-enabled care, Barts Health NHS Trust established an outpatient project strategy group (see Action research) focused on the potential roll-out of virtual consultations beyond those Diabetes and Cancer Surgery clinics included within the VOCAL study.

Sampling

Data collection

A summary of data collection and analysis across the different levels (macro, meso and micro) in the VOCAL study is presented in Table 1.

Data analysis

Data summary and synthesis of micro-, meso- and macro-level data

Table 3 summarises the final macro-, meso- and micro-level data sets for the VOCAL study.

The synthesis of the data sets summarised in Table 3 was guided by the following questions, informed by the SST framework (see Theoretical approach):

• How does the dynamic relationship between the macro (external social structures), meso (organisational routines and logics) and micro (individual understandings, dispositions and front-line actions) contexts explain how a real-time video consultation unfolds in the contrasting clinical settings of routine diabetes care and preoperative and postoperative cancer care?

• How do the outcomes of remote video consultations feed back in the short term to change (positively or negatively) the position-practice relations of the patient and the clinician, and in the longer term, the ability of the organisation and the health-care system to accommodate and sustain this service model?

In Chapters 3–5, we present the macro-, meso- and micro-level findings addressing these questions.

National policy on the digitisation of health and social care

The context for technological innovation in the UK public sector is currently extremely challenging. Analysis of our macro-level data set highlighted a significant policy push to develop the UK’s digital economy181–185 and digital government,186–189 as well as digital health. 4,5,15,182,190–196 However, constraints imposed by ongoing financial austerity have meant that there has been little slack to support either technological innovation or the piloting, organisational learning and extensive groundwork that is often needed to routinise new technologies or practices within the NHS and the wider public sector.

In health care, spending plans for the NHS set out an average real-terms increase per year of 0.75% above inflation between 2017–18 and 2020–21. 197 However, with costs and demand for health care rising faster than funding (as a result of treatments becoming more expensive, and people living longer and hence needing additional care),198 there are already significant overspending problems for the vast majority of NHS hospitals. 199 As set out in Figure 2, this means that there is a significant projected gap (£12B by 2022–23) between existing spending plans and the funding required to keep up with the rate of increase in NHS costs and demand, and little slack in the system to enable technological innovation.

FIGURE 2.

NHS spending scenarios. 199 Reproduced with permission from Appleby J and Gainsbury S. NHS Funding Choices and the 2017 General Election. London: Nuffield Trust; 2017.

Reproduced with permission from Appleby J and Gainsbury S. NHS Funding Choices and the 2017 General Election. London: Nuffield Trust; 2017.

At the time of writing, shortly after the general election in June 2017, and in the wake of the UK’s referendum decision in June 2016 to leave the European Union, there is significant uncertainty about the future direction of public spending. Spending plans indicate a decreasing share of gross domestic product (GDP) being devoted to the NHS (from around 7.3% in 2016–17 to 6.9% by 2022–23199), representing the largest sustained fall in NHS spending as a share of GDP in any period since 1951. 200 Compared with other countries, the UK is a relatively low spender on health care, and has already slipped into the bottom half of the Organisation for Economic Co-operation and Development’s health spending league198 and may well slip further. 201 Based on manifesto pledges ahead of the election, it seems likely that the projected decline in health spending as a proportion of GDP is set to continue. 199

Our analysis suggests that the low growth in NHS spending combined with sustained increases in demand are taking their toll on providers. 202 NHS trusts face significant challenges in meeting financial and performance targets (as evidenced, for instance, by the rising number of trusts missing the 4-hour target in A&E and the downgrading of the 18-week referral to treatment standard203). NHS finance directors are sceptical about the chances of the NHS delivering the required productivity gains over the next 5 years. 200

Against a background of worsening austerity, both the Five Year Forward View2 and the General Practice Forward View204 encouraged local health and social care economies to develop novel ways of working to deliver efficient, person-centred, cross-sector care. This was also a major focus of the Wachter Review,4 which centred specifically on the implementation of information systems across the NHS and social care, with particular attention given to the use of electronic health records and the ways in which digital health tools might support an (implicitly more efficient) ‘paperless’ NHS. Of the 10 recommendations made by Wachter, several addressed the need to train a national cohort of CCIOs in both primary and secondary care, who would have a dual role in supporting the digitisation of records and services and supporting the development of local improvement plans linked to new information systems. The Wachter review called for ‘new national funding to help Trusts go digital and achieve maximum benefit from digitisation’ (contains public sector information licensed under the Open Government Licence v3.0). 4 This led to the appointment of 12 NHS hospital trusts as ‘digital exemplars’,205 each receiving up to £10M to enable the use of digital technology to drive improvements in the care of patients (e.g. using real-time video links between ambulances and emergency departments to support better care during journeys to hospital).

The overall strategic terrain is thus characterised, on the one hand, by intense and worsening financial pressure, but on the other hand, by a ring-fenced funding stream for developing new structures and cross-sector linkages supported by information and communication technologies. In this context, technology appears to be seen as a logical route towards achieving cost-savings (e.g. by reducing staff time, supporting caring for people at home) and increasing quality (e.g. through monitoring care), and industry is depicted as an important partner in achieving this. Our data suggest that the stance taken by government and industry assumes that the digitisation of health and care services is the de facto option that needs to be engaged with across the public sector. As one GP involved in national-level debate about digital health put it, ‘The assumption is that if you don’t [engage with technology], you jolly well should.’ Although there was widespread acknowledgement of the need to train staff, few documents or informants mentioned the limited access, skills or motivation of the target patient population, 21% of whom (and a disproportionate number of poor, sick and elderly people) cannot use the web and 14% of whom are estimated to have no access to the Internet. 182,187,206

Despite the drive to digitise the NHS, we could find no formal national policy documents and limited guidance that explicitly related to virtual consultations. Guidance appeared to be restricted to a two-page guide on the use of remote consultations in the NHS produced by the Health and Social Care Information Centre (now NHS Digital; see Appendix 1), and focusing largely on issues of IG.

As indicated by the following extract from an interview in May 2016 with a senior decision-maker in NHS England, the push for virtual consultations was linked to the overarching theme of ‘telehealth’ and has come through a range of announcements and initiatives:

Would you say that there’s a coherent picture of what [telehealth] might look like?

No, there isn’t, not that I’ve seen. And I’m quite actively involved in this space. There might be some small documents floating around.

The policy of digitising health appears to be folded into other programmes – for instance, the ‘New Models of Care’ programme stemming from the Five Year Forward View2 that commenced in January 2015 and included the NHS ‘vanguards’, 50 local innovative services focused on delivering the ‘New Models of Care’ programme and supported by a dedicated tranche of innovation funding. 207 Although it is arguably commendable that policy has been programme focused and not technology focused, it has meant that virtual consultations (and/or the infrastructure, staff training and organisational development underpinning them) have not been actively commissioned or funded centrally. Rather, when technology-based new service models have emerged, they have arisen through local initiatives.

As the senior decision-maker quoted above went on to say: ‘There are pockets of success, and there are certain vanguards exploring it, there’s bits and bobs. But there’s not actively a digital fund for telehealth’. At the time of our fieldwork, this looks set to change, with a new NHS Digital Academy under development following the recommendations of the Wachter review. 4 The aim of the Academy is to develop a workforce of trained ‘clinician-informaticists’ in NHS trusts, give them appropriate resources and authority, and improve both capacity and capability in the health and care system. The Wachter review4 called for approximately £42M (1% of the £4.2B to be spent on digitising the NHS) to support workforce development and deployment. However, no confirmation has yet been given on the amount to be invested in the Digital Academy or the wider NHS.

Transformation, telehealth and innovation

Since the early 2000s, there has been a strong policy emphasis on the need for the ‘transformation’ of health and social care provision in response to the problems faced by the rising incidence of chronic illness and with growing pressures on services. 2,191,208–210 Telehealth (i.e. remote transfer of data and/or communication between a patient at home and a health-care provider) has been promoted as a key part of this ‘transformation’ agenda. There is strong policy support for ‘disruptive innovation’ (i.e. services transformed rapidly and radically through technology), on the assumption that change that is sufficiently radical to be experienced as uncomfortable in the short term is more likely to achieve the long-term goals of improved efficiency and effectiveness than more incremental approaches.

As described in National policy on the digitisation of health and social care, there is significant scepticism that transforming services and outcomes can be achieved in the context of current/planned NHS spending. Documents and interviews in our macro-level data set indicate that the NHS will struggle to meet the requirement, set by the Five Year Forward View,2 to save £22B by 2020. In an effort to address this, the government’s sustainability and transformation fund includes additional funding (£1.8B in 2016–17, accessible subject to meeting specific financial controls) for NHS trusts, to help reduce reported deficits. However, if some or all of those funds are used to plug the deficit year on year, there will be little in the way of funding (or the organisational ‘slack’ that tends to go with it) for the service change and new models of care that are thought to be required to modernise and reshape NHS services.

In the documents we reviewed, telehealth generally tended to be associated with the goals of personalisation, integration and collaboration, and is typically presented in deterministic terms (in other words, that simply having access to particular technologies would lead to desired outcomes such as improved health status, more personalised care and reduced costs):

[We have an] unexploited opportunity to combine different technologies and changed ways of working in order to transform care delivery. For example, equipping house-bound elderly patients who suffer from congestive heart failure with new biosensor technology that can be remotely monitored can enable community nursing teams to improve outcomes and reduce hospitalisations.

Five Year Forward View (p. 34). 2

Contains public sector information licensed under the Open Government Licence v3.0

As one GP with a role in national-level discussions about telehealth told us:

It seems to be that they will take less time and be cheaper than face to face. And the very, very subsidiary thing that it might be more convenient for patients . . . hasn’t been presented as a major factor.

These findings resonate with a previous study by our team of earlier policy documents on telehealth. 81

A prominent focus of recent policy has been to smooth, standardise and speed up the innovation process in order for NHS patients to more quickly access medicines, medical technologies, diagnostics and digital products, and make life sciences a more attractive prospect for the UK economy. Our analysis highlighted naïve assumptions underpinning recent policy about the innovation process, which is often situated as a linear pathway involving identification of ‘gaps in the technology pipeline’ that might be addressed simply by articulating to innovators ‘the technology requirements that would best support [NHS] needs’ (Accelerated Access Review, p. 18; contains public sector information licensed under the Open Government Licence v3.0),15 and so smooth the innovation process from ‘discovery to adoption and spread into clinical practice’ (Personalised Health and Care 2020, p. 43; contains public sector information licensed under the Open Government Licence v3.0). 5 This approach emphasises supply-side innovation, with little attention given to how demand might be mobilised, and the kinds of recognition and rewards that might stimulate adoption. 211–213 Although meaningful dialogue with patients is emphasised in documents recommending accelerated access,15 this tends to focus on developing a shared understanding of what matters to users (e.g. through horizon scanning and evaluation). Shared understanding is important, but is not sufficient; technological advances need to be underpinned by a user-centred approach to design and delivery. 214

In documents, interviews and field notes in our macro-level data set, telehealth was widely considered to have a role in delivering changes ‘at scale’ and ‘at pace’. Our analysis indicated limited recognition that virtual consultations and similar technology-supported service innovations may not be adopted at all by front-line staff (or may later be abandoned by them). Furthermore, the potential time lag between the adoption of technology and any realisation of productivity gains was rarely acknowledged either by interviewees or in policy documents.

It is widely acknowledged that it is very difficult to innovate in the NHS and that ‘transformative’ and ‘disruptive’ innovation in a cash-limited, public sector setting is significantly harder than incremental change (see, for example, a recent book, by Fitzgerald and McDermott, summarising key empirical studies and theoretical perspectives215). As Fitzgerald and McDermott215 point out, a crucial precondition for organisational innovation is the ability to capture emerging knowledge and feed it into organisational learning. Our macro-level findings suggested important sectoral differences in attitudes towards such knowledge capture. Industry interviewees expressed confidence in the standard commercial ‘fail early, fail often’ approach of iterating software design to optimise the use of a technology in a particular setting. In contrast, clinicians and policy-makers’ responses reflected what might be called the ‘does it already work?’ culture of UK public-sector health care (a prevailing expectation for RCT evidence generated elsewhere, but assumed to be transferable to the current setting). Senior national-level decision-makers from different departments in NHS England clearly acknowledged that ‘the evidence is still very limited when it comes to digital offering’ and that there is a need ‘for us to be producing some evidence that things are progressing well to keep the people at the political end happy’.

Interviewees from regulators, industry and professional organisations similarly talked about the limited evidence available and the need to know more about, for instance, ‘what makes a good-quality remote consultation?’. However, the emphasis in policy was very much on generating an evidence base focused on ‘expand[ing] NHS operational research, RCT capability and other methods to promote more rigorous ways of answering high impact questions in health services redesign’ (Five Year Forward View, p. 34; contains public sector information licensed under the Open Government Licence v3.0). 2 This approach to generating knowledge has characterised a number of government-funded initiatives designed to improve the evidence base and adoption of telehealth. The most widely cited is the Whole System Demonstrator (WSD) programme, established by the Department of Health to ‘provide a clear evidence base to support important investment decisions and show how the technology supports people to live independently, take control and be responsible for their own health and care’ (p. 1) (contains public sector information licensed under the Open Government Licence v3.0). 216 The WSD study involved a large RCT of telehealth and telecare, involving 6191 patients and 238 GP practices across three sites. The resulting evidence was viewed as ‘disappointing’, with further evidence required on cost-effectiveness, as well as mechanisms for reimbursement, and clinical buy-in, to support adoption of health care. 193

Other approaches (e.g. use of ‘test-bed’ sites – such as ‘NHS Innovation Test Beds’217 or ‘Healthy New Towns’;218 see Appendix 1 for an overview – bringing together new technologies, new staffing models and payment for outcomes) have been promoted. However, our findings suggest that public sector knowledge capture continues to be dominated by trial design, is a poor fit with the ongoing adaptation of digital technologies and is at odds with the commercial approach. As one interviewee from industry put it:

it’s a block to innovation . . . because everyone wants someone to prove they have done it before in exactly the same sort of customer as they are. Well, how does anyone ever start then! You know, someone at some point has to give someone a break. And that’s the thing that people find really hard: to get a break into the NHS.

Challenges for industry

In the documents we reviewed, the technology industry is depicted as the innovator and producer of transformative technologies15 that can change the lives of NHS patients; policy is concerned with speeding up access to such innovations and boosting the UK economy. In relation to virtual consultation technologies, it is assumed that if developed and implemented at scale by a thriving and innovation-driven technology industry, these will generate change and thereby (at least partly) solve the uncomfortable problem of how to address increasing pressures on the NHS (particularly with regard to the growing ‘burden’ of chronically ill and dependent citizens), while also saving money and creating ‘a virtuous circle of economic growth for the UK’ (p. 44) (contains public sector information licensed under the Open Government Licence v3.0). 5

Interviews and documents painted a picture of industry as a diverse group of companies, with some (typically larger ones) pursuing a strategy of offering a narrow range of products at scale and seeking to stifle competition across those ranges, and others [typically small and medium-sized enterprises (SMEs)] being more focused on identifying clinical or care issues in partnership with the NHS or social care, as well as users, and working to develop technologies that can help to address these issues. As one senior decision-maker said, ‘what we are seeing is a lot of telehealth organisations – small SMEs but also large-scale providers – working with large organisations, corporates, also some NHS organisations, to offer telehealth services’.

While a thriving technology industry that generates innovations is a key component of the policy vision for a more efficient health-care sector, the relationship between the technology industry and the public sector is complex and sometimes awkward, and makes health and social care a challenging market to enter and survive in. For example, procurement processes are typically long and complex; the market is known to have limited and shrinking resources; and block contracts often favour larger providers, making health care a challenging market to enter and survive in. There was clear recognition of these challenges within interviews and policy documents, for instance:

[Industry] largely recognise that a state-funded health care system such as the NHS must take issues of affordability seriously but find the approach to innovation, the slow track to reimbursement, and low prices without generating wider diffusion and volumes, all make the UK a challenging market.

Accelerated Access Review: Final Report Review of Innovative Medicines and Medical Technologies (p. 8). 15

Contains public sector information licensed under the Open Government Licence v3.0

Part of why progress has not been as fast as it should have been is that the NHS has oscillated between two opposite approaches to information technology adoption – neither of which now makes sense. At times we have tried highly centralised national procurements and implementations. When they have failed due to lack of local engagement and lack of sensitivity to local circumstances, we have veered to the opposite extreme of ‘letting a thousand flowers bloom’.

Five Year Forward View (p. 31). 2

Contains public sector information licensed under the Open Government Licence v3.0

The extension of NHS contracts to a wider range of providers, including SMEs, marks a pronounced shift from the National Programme for IT (2007–11), in which the strategy was ‘ruthless standardisation’ and potentially highly lucrative contracts were restricted to a very small number of preferred providers. 219 Our findings suggest that the balance is still not right, and that a focus on centralised versus decentralised approaches to technology adoption does not address the whole picture. As one senior representative from industry put it when we interviewed them in April 2016:

They’re all negotiating locally . . . So, I agree that the National Programme [for IT] was wrong. But where we are now is wrong. We have gone far too far the other way. Because you do still have to think about the whole-system cost, don’t you, and the whole-system benefit. And there are some things where it just makes sense, it just makes sense.

Some larger companies (e.g. Microsoft) appear to have an evolving process of technology development and supply that (at least on the surface) acknowledges and works with health-care organisations and individuals, as well as patients and carers, in order to develop more nuanced and marketable products (though as noted above, this is not necessarily oriented to the kinds of technologies that might facilitate virtual consultations). This is, however, a resource-intensive process with no guarantee of meeting shareholder or executive expectations at the end of it. The same companies appear to be less concerned with technologies that might facilitate virtual consultations (e.g. Skype, FaceTime) and more with the ‘wellness and wearables markets’ (senior executive, industry). This is perhaps as a result of their awareness that adoption will be slow for complex institutional reasons and, hence, virtual consultation technologies offer limited potential for profit. Companies are also aware of the potential for reputational risk associated with profiting from virtual consultations in the NHS, whereas the wellness and wearables market is (or could be) effectively ‘direct to consumer’.

Our data suggest that both larger and smaller firms sometimes (although not always) struggle to engage with health-care organisations and systems. Take the following extract from a senior executive working in one multinational technology supplier:

I would like to have more of an influence, I think . . . Its very difficult for, not just for [our company], for the large suppliers, to engage with the NHS. This is one of my, this is the problem . . . because there’s a kind of a disconnect between industry and the NHS, people are working in the NHS on the test beds, on the vanguards, on these pilots, but they’re working within the boundaries of what they know and actually they have no idea what is possible or what is already mainstream. So they think that they’re coming up with like this ground-breaking stuff and it’s just like [cringes] . . . if only we could talk, if only we could find a mechanism to increase awareness about what can we do.

Our analysis suggests that the NHS does not have the ‘slack’ to receive industry in a way that could build the necessary engagement to support development of virtual consultations. This resonates with the wider literature that characterises the NHS as inherently complex and sees no simple solution for technological innovation. 215 Interviewees (both within and outside industry) suggested that the Department of Health and NHS England might do more to ease the process of working with NHS providers and commissioners. Interviews with NHS decision-makers indicated that significant work already occurs to engage with suppliers and to ‘realistically manage expectations’ about NHS procurement processes and time frames. However, this did not appear to have made a significant difference to suppliers who spoke to us about slow time frames, ‘decision-making by committee’, ‘so much duplication’, ‘consultation about everything’ and the ‘need for everyone [i.e. all NHS providers/commissioners] to go out and evaluate every single thing’. The emphasis on one-off procurement contracts for particular technologies ran contrary to the desire on the part of industry stakeholders that we spoke with to develop mature NHS–industry partnerships, in which the latter is committed to supporting an evolving service via an evolving package of technology and support.

Interviewees (from industry and the NHS) spoke positively about some national-level initiatives (e.g. NHS Innovation Test Beds), which they felt may prove helpful in facilitating partnerships and working patterns, and so enable a less technology-centric way of working. However, most felt that it is currently unclear whether or not such initiatives can facilitate the capacity-building and articulation work (i.e. all of the tasks involved in assembling, scheduling, monitoring and co-ordinating all of the steps necessary to complete a task220) needed to enable a more iterative approach.

Both SMEs and larger companies often appeared to adopt a ‘plug-and-play’ model of technology. Such a model might appeal to health-care organisations seeking ‘easy solutions’ to the challenge of rising costs and expanding need, but as our meso-level data illustrate (see Chapter 4), a business model that rests on ‘off-the-shelf’ technologies that can be bought, installed and made to work, while ignoring the ongoing work required to embed, routinise and sustain them, may achieve short-term sales at the expense of longer-term problems.

The above findings suggest that providing technological solutions to the NHS will not be easy either for small, ‘niche’ software companies (SMEs) or for the larger providers with which the NHS has traditionally contracted. Our contact with representatives from the third-sector organisations also suggested that, although there is significant interest in engaging with discussion about technological innovation in health care, there is currently little capacity to do so.

Rolling out remote consultations

As noted at the start of Chapter 3, interviewees from both policy and industry acknowledged a general policy push for virtual consultations to spread across the NHS. However, we documented numerous concerns, particularly from participants drawn from the NHS and professional organisations, that delivering changes ‘at scale’ would be far from straightforward, as roll-out in any locality would be influenced by (among other things) differences in organisational culture, infrastructure, the nature and causes of professional resistance, IG challenges and the logistics of payment.

A key issue repeatedly raised by interviewees, but rarely evident in published documents, was how reimbursement for virtual consultations would be implemented. Our wider data set showed that, in localities that have introduced virtual consultations (including, but not limited to, Barts Health NHS Trust), payment has been negotiated with local commissioners on a site-by-site basis. However, establishing virtual consultations as business-as-usual across the entire NHS would require a more systematic plan for reimbursement that does not yet exist (and which some interviewees indicated would be problematic to implement). As one senior decision-maker in NHS England told us:

We have a drug tariff that does prescriptions very well, but we don’t have anything for digital. And I personally believe that telehealth falls under that space because the service that could be commissioned or could be reimbursed for but we don’t have the route to do that yet.

Our analysis highlighted several strands of work being undertaken by NHS Improvement (previously Monitor) and NHS England (see Appendix 1) to review payment and pricing, and indicated significant awareness within these organisations that there are specific challenges relating to reimbursement of virtual consultations. The development of an innovation and technology tariff has gone some way to addressing this, by removing the need for multiple local price negotiations (instead guaranteeing automatic reimbursement when an approved innovation is used), but is currently limited to six medical devices or apps (and does not currently include the use of virtual consultation technologies). Our analysis suggests that the current need for individual provider organisations to negotiate payment with local CCGs will continue to present a significant barrier to the national roll-out of virtual consultations, and might also compromise the long-term sustainability of existing virtual consultation services.

Some of our interviewees suggested that, in the longer term, the proposed shift away from activity-based funding and towards the introduction of capitated payments might facilitate roll-out. Capitated payments have been put forward as a long-term vision for health reform, for instance:

We are interested in learning from commissioners and providers that are implementing alternative payment approaches to enhance system-wide incentives . . . for example . . . to focus . . . on integrated care. Alternative payment approaches might include pathway, capitation or outcomes-based payments.

Reproduced with permission from National Tariff Payment System 2014/15 (p. 138)221

At their most radical, PACS [primary and acute care systems] would take accountability for the whole health needs of a registered list of patients, under a delegated capitated budget – similar to the Accountable Care Organisations that are emerging in Spain, the United States, Singapore, and a number of other countries.

Five Year Forward View (p. 21). 2

Contains public sector information licensed under the Open Government Licence v3.0

In this model, a lump-sum payment to a provider (or group of providers) could cover the majority of care for a group of patients across different settings and, in theory, enable funds to be spent however is thought best locally (in negotiation within/across providers and CCGs). It seems likely that moves to develop a more effective payment strategy within the NHS will combine different mechanisms, and be complemented by other levers for influencing quality and efficiency. 222 Even if capitated payments are pursued at scale, our analysis suggests that redesign of the payment system will take time and any impact would probably not be felt for several years.

Our meso-level data indicated that IG was a significant concern when setting up virtual consultations (see Chapter 4). This concern was not mirrored in our meso-level data set, with interviews or documents situating IG as something that can be sensibly addressed and dealt with (ensuring appropriate privacy and security of data), rather than as a potential block to further roll-out. As one GP involved in national-level discussions about telehealth told us: ‘I think it has to be looked at but I don’t really see it as a major obstacle . . . the main issue is the IT connection which is a security issue, not an IG issue’.

Digital participation

Our macro-level data suggested a widespread assumption that digital literacy among the general population and NHS clinicians and staff would develop in pace with technological advances. Interviews indicated that, for instance, NHS England and NHS Digital not only wanted to engage with this agenda (e.g. by supporting people from disadvantaged backgrounds, and in particular ethnic minorities, to improve their digital literacy182,183), but are also taking action to do so. As one senior decision-maker in NHS Digital told us:

Quite an interesting part of the move to NHS Digital is not just what we do with the public . . . but being a bit of a kind of moral compass and challenge . . . so as NHS Digital develops its products and services – things like patient online, online bookings, those kind of things – challenging the organisation a bit to remember digital inclusion and the challenge of leaving people behind.

There remained an assumption across many (but not all) interviews and documents that people need to engage with the digitisation of health and social care, and there is little or no option for doing otherwise in the future. As one interviewee neatly put it, ‘we are perhaps not getting as far as we could with this’.

Initiatives such as the government’s ‘digital strategy’ (Cabinet Office 2012)223 aim for digital services to be the default option for public sector services, including health and social care. Our analysis indicates that the focus of much work to date has been on redesigning digital services to make them accessible and straightforward to use, as well as offering ‘assisted digital support’ for those who have never accessed the internet (estimated at 10% of the UK population224). In line with the increasing consumerisation of technology and raised expectations about access to digital services, the approach to developing digital health services appears to be shifting. The above interviewee described less of an emphasis on digital skills and support and more on making digital health an attractive option for potential users:

What we learned through the programme is that there is a digital skills problem still but that’s probably not the biggest barrier because people are just becoming more adept in their everyday lives, technology’s becoming easier to use . . . What we found was that, actually, when people had a motivation and a perceived benefit they would make the minimal effort to develop the skills. What was more a problem [was] that we were not making a compelling digital offer that people wanted to do . . . that people are not viewing digital information offers, digital products, digital means of engaging with the health-care system with sufficient enthusiasm for them to make the effort . . .

Despite this shift, our analysis of policy documents and interviews with policy-makers, regulators and industry representatives indicated that the approach to developing telehealth tends to assume that digital is the de facto option that needs to be ‘engaged with’ (and which forms the starting point for a project or programme). This contrasted with those from patient organisations who advocated the language of ‘digital inclusion’ and emphasised digital health as one option among many.

The framing of digital literacy and digital connectivity as ‘deficits’ in disadvantaged populations that can be made good through training and support contrasts with approaches to health inequalities taken by some third-sector and patient organisations. The latter typically frame the challenge in social terms (e.g. poverty or loneliness) rather than technological ones. A technological framing tends to hone in on the provision of particular technologies, whereas a social (or sociotechnical) framing is more likely to produce a more broad-based solution that includes an assessment of the user experience, patient and carer needs and the use of co-design methodologies.

Overview of virtual consultations in Diabetes and Cancer Surgery services

Trust policy and service-level agreements

The implementation and sustained use of virtual consultations in participating services at Barts Health NHS Trust required the development of technical support and IG protocols. This was attributable to the impact of the use of (in this case) Skype software on the trust’s technical infrastructure and the legal, regulatory and local policy issues relating to the protection of patient data, privacy and consent. A major challenge has been to address the lack of consensus on these issues (both locally and nationally).

In relation to technical support, and in line with local policy in most NHS trusts, Barts Health NHS Trust policy states that new software downloads must be performed by ICT staff and requested through the IT helpdesk as standard procedure. During project setup, however, and despite the VOCAL study including both the CCIO and the estates and facilities manager as co-signatories, we found that ICT staff simply did not respond to such requests when they related to the installation of Skype (which was not routinely provided on trust computers). This was attributable to the absence of trust policy and service-level agreements in relation to supporting Skype on NHS computers (and, thus, a mismatch between what had been offered in the research bid and what was de facto supported on the ground). The original piloting of Skype in the diabetes service had been achieved through ‘informal’ requests to install Skype on a handful of computers in a single clinic.

Following considerable negotiation from the research team and clinicians in participating services, a ‘bespoke’ agreement was established, in which requests to install Skype on additional computers were to be made directly to a designated ICT manager, who would then arrange for the downloads to be completed. The implementation of this arrangement took 2 months. The need for what might be called a workaround at such an early stage in the project was perhaps telling, and almost certainly reflects the extreme pressure under which the ICT department was operating at the time. Our interviews with local ICT staff and managers strongly suggest that, although they were comfortable with virtual consultations being undertaken via Skype as a small ‘demonstration’ project by a clinician with whom they were on first-name terms, they feared that extending the option to download Skype (or similar media) on any trust computer would ‘open the floodgates’ to requests for technical support at a time when they were only just managing to meet existing demand.

The action research component of the VOCAL study sought to establish how technical support for virtual consultations could be provided across Barts Health NHS Trust in the long term. Notwithstanding the short-term workaround described above, a mainstream trust policy on virtual consultations using Skype needed to be established in collaboration with the IG and ICT departments. Working with service managers and the Barts Health IG Manager, the research team produced an internal policy document, covering issues relevant to IG in relation to virtual consultations and SOPs on technical support. The draft document, which was based on the setting up and running of the Skype service within the diabetes clinic and took 2 months to develop, was then reviewed by representatives across key departments, took 9 months to gain formal approval and become trust policy, and has since been continually adapted. Establishing formal agreements with the ICT department, and the views of their staff on the impact of virtual media, such as Skype, on the service infrastructure proved to be key to both setting up the virtual consultations in our participating services and subsequently addressing wider roll-out across the trust. The approved document was subsequently made available to all departments via the trust intranet.

At the outset of this study, the need for a formal, written agreement with the ICT department and a formal Barts Health NHS Trust policy on virtual consultations more generally was not anticipated to be a rate-limiting step, either by the research team or by participating clinical teams. In retrospect, however, this document can be seen as what Latour would call an ‘immutable mobile’225 – that is, an artefact describing ‘the way things are’ that is accepted by all key actors, and which serves to stabilise what would otherwise be a highly unstable arrangement of people and technologies. 226 Immutable mobiles include maps, diagrams, graphs and lists of agreed procedures that move within a network and its nodal points of passage, but which retain the same essential properties in different contexts, thereby allowing relations to be performed in the same way in a variety of different settings and locations.

In relation to IG, local implementation was held back by both local and national concerns about the security of virtual consultations in general, and Skype consultations in particular. This concern relates to the security of data streams (the hypothetical risk of someone eavesdropping on calls or accessing messages sent through Skype) and account security (access to information saved on computers). As part of the action research element of this study, the research team worked with participating clinical services and IG teams at Barts Health NHS Trust, with the support of an IG specialist within the local CCG, to identify models of best practice and establish IG policy on the use of Skype within the trust.

As part of the collaborative work with the IG team, we undertook a review of Skype security, which found that the encryption meets the minimum NHS network requirements. Many other virtual media applications (e.g. FaceTime) also meet such criteria. However, it is important to highlight that alternative voice-over-internet protocol (VoIP) systems (a group of technologies for the delivery of voice communications over the internet), some of which are currently being considered by Barts Health NHS Trust and other NHS trusts as alternatives to Skype, would need to undergo similar reviews on their data security.

In relation to account security, our ethnographic fieldwork informed the development of a guiding document for clinic staff to minimise risk to patient privacy and confidentiality. That guidance has been included in the internal SOP document. In addition, we have produced a summary document for wider dissemination to other NHS trusts. This has been reviewed and endorsed by the UK IGA, which used it as the starting point for developing national policy guidelines on the use of Skype and FaceTime across the NHS.

New clinical roles and practices: triage, technical support and direct access

It was evident from our interviews and ethnographic observations that clinicians involved in virtual consultations took on new roles and practices in relation to patient triage (judging a patient’s suitability for virtual consultation), patient setup (ensuring that the technology worked, and supporting patients with its use) and medical documentation (changing how electronic and paper documents and other artefacts were used in consultations).

In relation to patient triage, there were no formal criteria in any specialty on how to assess patient suitability for virtual consultations. It was clear that such decisions were made by the clinician on a case-by-case basis, informed by a range of factors that varied across the different clinical contexts. Within the Adult/Young Adult Diabetes clinic, such decisions were discussed and mutually agreed between the clinician and patient, depending on the patient’s support needs and wishes. The reality was that most patients (although not necessarily most appointments) were considered eligible for the virtual option.

Within the Antenatal Diabetes Clinic, offering virtual consultations was thought to be considerably more risky, and there were two patients (mother and fetus) to consider. Although moderate perturbations in blood glucose levels would likely cause little harm to the mother, serious complications (e.g. congenital abnormalities, intrauterine death) were possible in the fetus. Furthermore, resistance to insulin tends to increase as pregnancy progresses, necessitating up-titration of the insulin dose, sometimes on a daily basis. Owing to the complexity and risk involved in this clinical context, the option for a virtual consultation was offered only to patients considered to have relatively stable blood sugar readings and the capacity to self-manage effectively. Furthermore, because the clinical picture can change on an almost daily basis, close familiarity with the patient’s history and background appeared to be crucial, especially since the virtual encounter was undertaken in the absence of a full medical record (since the patient held that record). This is illustrated by the following quotation from the consultant:

I always find when I have a patient who I haven’t really got under the skin of. So let’s say it’s a lady who has gestational diabetes, diagnosed at 28 weeks and I’ve met her at 32 weeks for literally less than 5 minutes, and I might have thought well sugars are all fine: ‘We’ll see you for a scan in 4 weeks, why don’t you Skype at 2 weeks?’ I’ve had a few of those because I think they’re all fine. That Skype appointment to me is terrifying because I then get my orange bit of paper. There’s not much information on it because at the time I wasn’t very worried and I’ve lost that opportunity to go through it again. I think as clinicians, if you can’t remember the patient, then you’re duty bound to go back and check in your mind that you have got all the information you need. As soon as you don’t have the notes, if you can’t remember the patient, you might have this orange piece of paper saying all seems fine, why not Skype in 2 weeks. But actually you’re just a bit worried. Whereas someone like [patient X, known to consultant], who I’m Skyping three times a week, I mean, I totally get her.

In the Hepatobiliary and Pancreatic Cancer Surgery clinic, virtual consultations were offered to patients for postoperative follow-up if the consultant was confident of the treatment outcome, based on the details of the operation performed and the postoperative CT scan result.

Another aspect of patient triage was confirming that the patient had access to Skype (on a computer or a mobile device) and that they or a family member knew how to use it, since these were prerequisites for inclusion in the study and the trust does not offer technical support for patient-owned technologies. Patients seeking to use the virtual consultation option must establish contact with the clinician’s Skype account before a consultation can take place.

As part of the action research for this study, the research team collaborated with clinicians, the IG department and local CCG IG managers to produce a patient leaflet containing information about using Skype and written details of the clinic Skype username/ID (identification). Although patients found this material useful, additional support was often provided by the clinician as part of the setup phase. For instance, prior to a first virtual consultation, the clinician typically contacted the patient (via e-mail or telephone) to confirm their account details, check that they had registered and confirm that a contact invitation had been received. Clinicians sometimes also conducted ‘test calls’ with patients before their consultation, so that the patient could familiarise themselves with the technology or check that the video and audio worked.

Patients and clinicians often shared documents (e.g. paper medical record) and similar artefacts (such as lists of blood sugar readings) as part of their consultation. The ease (or not) with which such material exchanges happened was of course influenced by whether the consultation was face to face or virtual; these data are discussed in more detail below in the section on the micro-level findings (see Chapter 5).

In relation to the organisational processes associated with material artefacts (e.g. creating or editing them), the introduction of Skype has had little impact within the Adult/Young Adult Diabetes and Hepatobiliary and Pancreatic Cancer Surgery clinics, as all key documents are stored and accessed electronically.

In the Antenatal Diabetes Clinic, however, a paper patient-held maternity record (on which all members of the multidisciplinary team entered data) was used. This was readily shared across the desk in the face-to-face setting, but impossible to share easily in the virtual setting – a restriction that had a considerable impact on how the clinical consultation unfolded, and strongly influenced the triage decision on who was ‘suitable’ for a virtual consultation. The consultant’s perception of risk was also influenced by the fact that the antenatal outpatient clinic was set up to generate additional medical information (collected by additional ancillary staff) to support their clinical assessment. For example, patients’ weight, blood pressure and urinary ketone levels are taken on arrival at the clinic, and these data are made available to the clinician along with the hospital and patient-held maternity record. In comparison, the virtual consultation with an antenatal patient was invariably data sparse.

In the Adult/Young Adult Diabetes clinic, the use of virtual consultations facilitated patients’ direct access to the clinician, because patients could send a short message service (SMS) message via Skype and/or contact a clinician informally for an ad hoc consultation (this process usually involved the consultant indicating on Skype that they were online and ‘not busy’, the patient messaging them to ask if they were free and then together arranging a time shortly after to connect). Although it had been known at the outset that this option was technically possible, neither the clinicians nor the research team had anticipated how this function would be repurposed by patients to bypass the appointment system and establish a direct communication channel with their doctor or nurse. This ad hoc part of the service was used by some patients to contact the clinician between their scheduled appointments to ask questions or raise concerns. This ‘open access’ aspect of the service was welcomed by the Diabetes team because it aligned with the clinic’s goals to enhance patient self-management and engagement (as noted in Virtual consultations in the Adult/Young Adult Diabetes service, the virtual consultation service had originally been established to solve the problem of a high DNA rate in patients who were typically from socioeconomically deprived and/or minority ethnic groups). Dealing with such messages and the ad hoc requests for an unscheduled Skype call appeared to be practically manageable, because the virtual consultations ran alongside the diabetes outpatient clinic. In contrast, clinicians’ access to Skype within the Antenatal Diabetes and Hepatobiliary and Pancreatic Cancer Surgery settings was restricted to specific times (and a narrow time window), which limited their capacity to respond in the same way.

The repurposing of Skype technology to contact clinicians outside scheduled appointments raised challenges in relation to managing patients’ expectations (e.g. about when and for what purposes a consultant can be contacted). We noted occasional examples of patients using the Skype messaging service to contact the consultant outside usual hours for non-medical and non-urgent matters, with an apparent expectation of ‘24/7’ service). This also has implications for clinician workload, reimbursement (see Chapter 3, Rolling out remote consultations) and medicolegal issues.

Clinic management and administration

Embedding virtual consultations in clinic routines was a complex and difficult process that took many months and which was strongly influenced by physical space and the material properties of technologies and artefacts. Our ethnographic data highlight how work practices and the often subtle, but important, social, professional and material contexts in which such work takes place have a bearing on the capacity to embed the technology in routine practice.

This process of embedding virtual consultations is presented in Figures 3–8, which map out the human and technical interactions and interdependencies on which the execution of both face-to-face and Skype consultations depended in the Adult/Young Adult Diabetes clinic (see Figures 3 and 4), the Antenatal Diabetes Clinic (see Figures 5 and 6), and the Hepatobiliary and Pancreatic Cancer Surgery clinic (see Figures 7 and 8). Drawing on Feldman’s notion of ‘organisational routine’142,167 – defined as ‘a repetitive, recognizable pattern of interdependent actions, involving multiple actors’159 – we explored how face-to-face and virtual consultations were enacted, purposefully capturing the complex (and very different) actions involved in trying to make face-to-face and virtual consultations happen in different clinical and material contexts. In contrast to the protocols (‘proxy routines’167) and descriptions of staff (‘ostensive routine’167) across Barts Health NHS Trust, these observed instances of consultations (referred to by Feldman as ‘performative routines’167) enabled us to identify the social and organisational factors shaping the ways in which clinics sought to embed the Skype technology and virtual consultation services over time. It also allowed us to tease out the logistical and practical aspects of clinic work, and the technologies and artefacts supporting it, that facilitated the use (or non-use) of the technology within particular clinical contexts.

FIGURE 3.

The routine for a face-to-face consultation in the Adult/Young Adult Diabetes clinic.

FIGURE 4.

The routine for a virtual consultation in the Adult/Young Adult Diabetes clinic. ID, identification.

FIGURE 5.

The routine for a face-to-face consultation in the Antenatal Diabetes Clinic. ID, identification.

FIGURE 6.

The routine for a virtual consultation in the Antenatal Diabetes Clinic. ID, identification.

FIGURE 7.

The routine for a face-to-face consultation in the Cancer Surgery clinic.

FIGURE 8.

The routine for a virtual consultation in the Cancer Surgery clinic. ID, identification.

The work involved in embedding face-to-face and virtual consultations in clinic work broadly related to four key processes or subroutines: generating data/information (highlighted in light blue in Figures 3–8), enabling access to data/information (highlighted in blue), facilitating patient access through the clinic (highlighted in green), and tracking the patient through the clinic/care pathway (highlighted pale green). Crucially, as we come on to discuss below, each of these processes was supported by technical and material artefacts, the movement of artefacts across the (virtual and physical) spaces of the clinic/hospital and the role of multiple clinical and non-clinical actors.

As is visible in Figures 3–8, the introduction of virtual consultations significantly affected the ways in which consultations were undertaken across the three clinics. The physical presence of the patient within the clinic setting was fundamental to existing ways of identifying, scheduling, conducting, rebooking and monitoring patient appointments. For instance, in Figure 7 the physical presence of the patient at reception prompts ‘check in’ and generation of the clinic outcome form; this enables the nurse assistant to identify the patient, conduct the relevant tests and record the results on the form, which in turn enables the remainder of the consultation to take place. Embedding virtual consultations within existing services involved significant reworking of those processes in ways that accounted for the ‘virtual’ presence of the patient. The extent to which existing face-to-face consultation routines needed to be reoriented, and the ways in which this was managed by staff varied across the three clinic settings depending on the people, technologies, material artefacts, physical and spatial arrangements, clinical pathways and assessment procedures already in place. As we set out in the remainder of this section, this meant great variation in how the Skype technology and virtual consultation service were (and were not) embedded in clinics.

Work routines across the three clinics were adapted to align the Skype technology and virtual consultation service with existing roles, processes and structures. This included processes and structures related to EPR management, appointment scheduling and recording attendance. These routines and structures vary across different clinics, and so staff have collectively evolved their own roles and workarounds within their local settings.

At the outset of the study, the Cerner EPR system was used across the Barts Health NHS Trust clinic as a shared scheduling tool to book appointments and indicate when the patient had arrived at the clinic and attended their appointment. The EPR system continued to be used throughout the study in both the Antenatal Diabetes and Hepatobiliary and Pancreatic Cancer Surgery clinics, providing a facility for auditing, service management and acting as a commissioning resource. It therefore needs to accurately reflect staff time and activity (including appointment type, frequency and durations).

In the Antenatal Diabetes clinic, the EPR was used to access medical information (stage of pregnancy, type of diabetes, blood results) and confirm upcoming appointments. However, in this context, the majority of clinic data were documented and reviewed using the maternity folder held by the patient. In the Cancer Surgery context, the EPR was used to review blood test results and CT scans as part of the patient’s cancer treatment review, and to review previous GP letters generated and stored through the system.

Within the Adult/Young Adult Diabetes service, the Cerner EPR system was used at the outset of the study to review previous appointments with the consultant (via GP letters generated and stored through the system), blood/urine results (for annual review) and clinical outcomes from other clinical departments. This was used in conjunction with the Diamond diabetes management system (used by the diabetes nurse specialists to document blood sugar readings and free-text documentation of the consultation discussion), hospital notes (with past medical history not documented electronically) and pre-appointment weight and blood pressure measurements (obtained on arrival at the clinic). Towards the end of the study, a new system (EMIS) was introduced for recording nurse consultations, leading to subtle but important disruption to the virtual consultation routine – and, in particular, to difficulties in recording (in the form of coded data) that a virtual consultation had taken place (see Overview of virtual consultations in Diabetes and Cancer Surgery services above). A key challenge to the introduction of virtual consultations at Barts Health NHS Trust was the reconfiguration of the EPR system to allow the booking of ‘webcam follow-up’ appointments on the clinic appointment schedules. Each consultant has an EPR ‘profile’ through which appointments are booked by the administration teams. The available appointment types and time slots that can be booked are configured according to their existing clinic schedule. The introduction of the ‘webcam appointments’ in any clinical service involved a lengthy process of form filling, enquiries (by e-mail/telephone) and discussions and agreements among senior service managers. Prior to the VOCAL study, EPR configurations were implemented by the administrative teams within the particular clinical service. However, changes in organisational structures meant that the EPR had become managed centrally by the trust’s ICT department. Consequently, the reconfiguration of EPR for the implementation of Skype required new formal procedures, forms and authorisations that had not previously been required when EPR was managed within the clinics (where clinical and EPR management staff knew and routinely worked with each other and had a culture of collaborative troubleshooting and mutual support).

The designation of virtual consultation appointment slots needed to be aligned with existing appointment schedules, clinician availability, and clinic and office space. As noted above, these routines and structures varied across the different clinics, and so staff developed their own scheduling times and processes for virtual consultations. Within the Adult/Young Adult Diabetes service, virtual consultations were conducted alongside the outpatient clinic. This was possible because the clinic and administrative staff within the service occupied their own office and clinic space, enabling them to access Skype during the usual clinic hours. The virtual consultation service became embedded in the work of the clinic, so staff and patients repurposed the technology to enable a more flexible approach that involved more ad hoc consultations that did not require formal booking. In contrast, the outpatient clinic spaces for Antenatal Diabetes and Hepatobiliary and Pancreatic Cancer Surgery were shared with other departments and located separately from the clinicians’ offices. This meant that specific times and locations had to be established for Skype appointments. In addition, in the Hepatobiliary and Pancreatic Cancer Surgery, the booking of patient appointments required a high level of flexibility, as so much depended on postoperative recovery of individual patients and test results (e.g. from CT scans or blood tests). It was crucial to maintain this flexibility with the introduction of the virtual consultation service; hence, the nurse booked patients into face-to-face appointments (via the routine set out in Figure 7) and then offered the Skype option by telephone when the test results had been received. This created an additional layer of work, requiring the nurse to schedule both a face-to-face and a virtual consultation, and then cancel the face-to-face appointment once the virtual consultation had taken place.

It was necessary to change the process of documenting appointment attendance to accommodate virtual consultations. In face-to-face consultations in the Adult/Young Adult Diabetes clinic (see Figure 3), patient attendance to the outpatient clinic was routinely recorded by the receptionist at the clinic entrance when the patient physically appeared, which, in turn, prompted the nurse assistant to obtain the patient’s hospital record and clinic outcome form for pre-appointment medical checks – events that did not happen in virtual consultations. Instead, the ‘usual’ receptionist role was effectively bypassed (see Figure 4), and clinicians took on the role of ensuring that attendance was accurately documented; the consultant walked to the reception desk after the clinic to report on who had ‘attended’, providing the clinic outcome form (which is usually completed by hand by the consultant to indicate the next appointment and handed in to reception by the patient on their departure, i.e. exactly the same information as required for a face-to-face consultation by a different routine) for the receptionist to input onto the patient’s records and book future appointments. In the virtual consultation service, the receptionist effectively checks the patient ‘in’ and ‘out’ in the same step as booking the next appointment. This routine was additionally supported by a new role for the nurse assistant, who collated the hospital records of the patients scheduled for a Skype consultation and simply placed them on the clinician’s desk at the start of each clinic, with a handwritten note on the top saying ‘Webcam’.

In the Antenatal Diabetes clinic, the receptionist role of checking the patient in and directing them through to the first of several pre-appointment checks (see Figure 5) was similarly bypassed. Instead, the consultant contacted the service administrator by telephone following the virtual consultation so that the administrator could record the patient’s attendance (see Figure 6). In the Hepatobiliary and Pancreatic Cancer Surgery clinic, the nurse accessed the EPR system directly after the Skype consultation in order to check the patient in and immediately out again (see Figure 8), before the consultant then sent relevant documentation to administrative staff via e-mail. In all cases, recording attendance in the face-to-face clinic was considerably easier and quicker than doing so in the virtual clinic. However, there was more parallel work at the time of face-to-face consultations to generate, exchange and physically move patient data and material artefacts (the clinic outcome form, the EPR and so on) in ways that enabled the legitimate transfer of the patient from one place to the next.

A major concern when seeking to embed virtual consultations in administrative and clinical routines was the absence of recent and accurate medical information, obtained through the pre-appointment checks and the comprehensive maternity notes used across the other clinical departments treating the patient. Take the following example from a consultant in the Antenatal Diabetes clinic:

So I feel much more vulnerable, in terms of not having all their records there. Not picking up on things that may have happened to the patient . . . You can see [on the patient-held maternity record] if they have seen another professional. OK you can ask someone on Skype, but, what I have discovered is that people will not tell you, which is reasonable, something they think you are not interested in.

In face-to-face consultations in the Antenatal Diabetes clinic, the consultation is dependent upon the patient-held maternity folder, which records, for instance, blood sugar readings, insulin doses and pre-appointment checks, such as blood pressure, weight or urine ketones. This information is not routinely replicated in the hospital or EPRs. Hence, virtual consultations take place in the absence of a full medical record, and were a particular challenge when the clinician did not have an established relationship with the patient (as, for instance, with many of the patients being seen for gestational diabetes). As one of the Antenatal Diabetes consultants neatly put it:

I think the problem is, whereas with most diabetes clinics you get to know the patients fairly well, often the name and when you see the notes, you recollect them. The antenatal patient is different. They may see me once. It would be a brief, superficial appointment, rather than someone seen for the last 10 years. Because we see so many patients, I don’t have that confidence that I know the patient prior to seeing them on the Skype screen . . . And its affected by patients being so huge number, and some will only be under our care for about 6 weeks.

As the detailed maps of organisational routines in Figures 4, 6 and 8 highlight, staff developed workarounds and performed ‘hidden work’ to support the safe delivery of clinical care in the remote environment and ensure that all key aspects of the interaction were adequately recorded. They found pragmatic solutions to practical problems, such as generating their own ‘temporary notes’ (as with the handwritten ‘Webcam’ note referred to above) and assuming responsibility for aligning these with the main EPR, as described here:

So what I have been doing with Skype patients is writing myself brief notes on this and holding them here, with view to file in the hospital notes when pregnancy has concluded. This becomes part of hospital records.

Antenatal Diabetes Consultant

This development of knowledge and practice required ongoing adaptation and resolution. For instance, considerable additional time and effort was involved in reconfiguring technical systems to accommodate ‘webcam’ appointments. In the Antenatal Diabetes clinic, in particular, the lack of ‘organisational slack’ – limited resources, fixed protocols and distributed roles and responsibilities – was a major barrier to the collaborative and adaptive efforts needed to embed virtual consultations as business as usual.

Resourcing and managing the roll-out of virtual consultations

Following the insights from our meso-level ethnography and interviews, a trust-wide decision was made to establish a cross-departmental working group to assist with overcoming the numerous practical, technical and operational barriers to the smooth and efficient use of Skype for virtual consultations. This came on the back of the publication of TST,158 which set out a significant programme of improvement within Barts Health NHS Trust and with relevant health and social care agencies (see Chapter 2, Setting and context). TST is concerned with radically redesigning services, including outpatient pathways, and includes plans to speed up access to specialist advice via traditional face-to-face services, as well as the roll-out of virtual consultations. The VOCAL study came at a time when senior management at the trust was turning its attention to roll-out beyond a handful of clinics. This senior level buy-in was crucial to moving that work forward, and to establishing an outpatient project strategy group (chaired by the chief medical officer) focused on the roll-out of remote consultations beyond those Diabetes and Cancer Surgery clinics included in the VOCAL study:

And [the director of operations] was saying to me, ‘Look, we’ve got buy-in from the chief medical officer, and from the director of strategy. They want to see this happen.’ And so we sort of talked about who are the kind of key people who we need to involve. And came up with a cast list of individuals that we need to get on board with this. So really it was ICT, who are really important, finance, outpatients – as in the core service – and then each of the sites. So we, we wanted to have kind of a senior lead from each of, each of the sites, to actually drive it forward.

Programme Lead

The outpatient project strategy group (see Action research) was initially focused on supporting the three services in the study, but later extended its remit to include the wider roll-out of virtual consultations across outpatient services. Initially, the roll-out focused on developing virtual consultation services in neurology, rheumatology, haematology and endocrinology, with plans in place to extend this to include cardiology (inherited arrhythmia), urology, lipid clinics and diabetes community services. The work undertaken in the diabetes service since 2011 has been crucial in highlighting the complex challenges of embedding virtual consultation services within routine practice at Barts Health NHS Trust. Ongoing input from the diabetes consultant and programme lead, arguably ‘telehealth champions’,227 also played a role in promoting and legitimating virtual consultations as ‘business as usual’ within the trust and building key relationships.

Collaboration across clinics and staff at Barts Health NHS Trust and the sharing of knowledge and practices have been critical to the development of virtual consultation services to date. Local teams worked collaboratively with one another (e.g. in working through business cases, SOPs and clinical pathways, as they sought to pilot and then embed virtual consultation services, or in identifying unused space that might be used for private Skype clinics), taking account of the need to ensure patient safety and clinical appropriateness, co-evolve the introduction of new technologies and the development of new team roles and processes, and address IG issues and concerns.

The roll-out has also needed to accommodate competing policy priorities locally and nationally, working with national policy-makers, regulators and industry partners to find workable ways forward through close dialogue and practical problem-solving. In practice, this has required cross-departmental collaborations with a number of external partners, including the IGA and NHS England (IG), CCGs (local tariff and outpatient process) and Microsoft UK (providers of technology). For instance, as part of the action research in the VOCAL study, Barts Health NHS Trust has also been working closely with Microsoft to test various options for virtual consultations, while providing the technology provider with the information required to produce more patient- and health-care-friendly tools.

Other departments in the trust have now commenced virtual consultations with apparent success. Progress has been varied, with some ‘moving faster than others’, often driven along by ‘clinicians who are really passionate, and driving it, and they want to see it happen’ [programme lead]. Formal evaluation of this wider roll-out is soon to commence.

Description of the data set

Obtaining video and audio-recordings of both ends of a virtual consultation raised substantial logistical, technical and ethics challenges. Two researchers were often needed – one to record in the clinic and the other to attend the patient’s home (perhaps many miles away). Patients (and staff) were usually, but not always, happy to have a researcher set up the video camera and then leave the room while the consultation took place. Whereas the pilot phase consultations were relatively easy to set up to include data collection for research (because they were undertaken outside regular clinic hours), once video consultations became ‘business as usual’, it was much harder for staff to find sufficient time to recruit potential research participants and alert the research team to a planned video consultation. In addition, as noted in Evolution of virtual consultation services over the study period, a significant proportion of video consultations in the Adult/Young Adult Diabetes clinic were ad hoc (undertaken spontaneously when the patient sent a SMS Skype message when they saw the clinician was online, or fitted in informally by the clinician as a ‘quick follow-up’); these were impossible to schedule in advance for video recording.

Our target had been to obtain at least 30 high-quality recordings of complete virtual consultations (20 in Diabetes, including Antenatal, and 10 in Cancer Surgery); we also sought to collect a sample of comparator face-to-face consultations. At the end of the study, in July 2017, we had collected recordings of a total of 30 remote and 17 face-to-face consultations (Table 7).

As detailed in Chapter 2, the RIAS analysis was undertaken on a subset of the data set in which virtual and face-to-face consultations were matched as closely as possible for clinical content and (in all cases) for clinician; the subsample consisted of 12 Antenatal Diabetes consultations (six conducted remotely via Skype and six conducted face to face), 12 Adult/Young Adult Diabetes consultations (six remote and six face to face) and 10 Cancer Surgery appointments (five remote and five face to face). Patients for virtual consultations had been purposefully selected and recruited by the clinician involved, based on their assessment that a virtual consultation would be safe and appropriate. As summarised in Table 7, patients having virtual consultations within the study were more likely to be of white British ethnicity (the same ethnicity as the clinician), but were of a similar age to those having face-to-face consultations. The ethnic differences probably reflected the fact that some Diabetes and Antenatal Diabetes patients were limited English speakers, which was considered to be too risky at this early stage in the evaluation of virtual consultations.

Our initial analysis of the virtual consultation data, based on repeated watching of the merged audio and video streams (which could be viewed in parallel on screen), identified the following five overarching themes (we explore these in more detail below).

Findings from our analysis using the Roter interaction analysis system

In the VOCAL study, we found that the RIAS proved to be useful in highlighting the different kinds of talk in virtual consultations and exploring how the technology influenced, and was influenced by, the interactions and dynamics between patients and clinicians.

In order to interpret the RIAS analysis, a broad understanding of the categories is needed (see Table 2 for a detailed breakdown). In summary, guided by the RIAS and additional work applying the coding framework to telemedicine,175 all talk and behaviour in virtual consultations in the VOCAL study were classified according to the following categories:

• socioemotional (broadly capturing talk related to social etiquette, indicating interest or understanding, concerns or attempts to reassure and negative talk)

• task focused (broadly capturing talk to instruct or guide, question-asking, information-giving, checking for understanding, requests and counselling in the form of giving instruction or advice to the patient)

• process oriented (relating to verbal interactions aimed at supporting the flow and direction of the conversation)

• technology related (any of the above, but relating specifically to the technology and/or its use).

The results presented below are based on a detailed micro-level analysis of 34 consultations in three different clinical contexts to explore how the use of Skype for virtual consultations affects the interaction and dialogue of the medical encounter. This includes 12 Antenatal Diabetes consultations (six conducted remotely via Skype and six conducted face to face), 12 Adult/Young Adult Diabetes consultations (six remote and six face to face) and 10 Cancer Surgery appointments (five remote and five face to face). The consultations were all follow-up appointments after the patient had met the clinician at least once. In the Antenatal Diabetes context, that appointment took place approximately 1 week beforehand to establish a schedule for insulin dosing. In the Adult/Young Adult Diabetes clinic, the consultation formed part of a routine 6-monthly review to discuss the patient’s condition, recent test results and treatment plan. For the Cancer Surgery, the consultation formed part of the follow-up assessment of the patient’s postoperative recovery and review of medical tests (blood tests, CT scans) and symptoms (the follow-up interval varied from 3 to 6 months). Virtual and face-to-face consultations in the data set for each clinic all involved the same clinician (i.e. a consultant diabetologist specialising in the care of pregnant women in the antenatal context, a different consultant diabetologist in the adult diabetes context, and a surgeon and a nurse specialist combined in the Cancer Surgery context).

As detailed in Chapter 2, patients for virtual consultations had been purposefully selected and recruited by the clinician involved, based on their assessment that a virtual consultation would be safe and appropriate. The face-to-face comparisons were selected based on the similarity of the consultation (stage of treatment, purpose of appointment) and the clinician’s judgement that they would also have been suitable for the remote consultation option. As summarised in Table 7, patients having virtual consultations were more likely to be of white British ethnicity (the same ethnicity as the clinician), but were of a similar age to those having face-to-face consultations. The ethnic differences probably reflected the fact that some Diabetes and Antenatal Diabetes patients were limited English speakers, which was considered too risky at this early stage in the evaluation of virtual consultations.

Question 1: are virtual consultations shorter and more ‘to the point’ than face-to-face ones?

Consultation length (defined by the RIAS as frequency of utterances) was shorter for virtual consultations than face-to-face consultations across all three clinical settings. There was a small difference in consultation length in the Adult/Young Adult Diabetes service, with virtual consultations being slightly shorter in terms of the number of utterances for virtual consultations [median = 337, interquartile range (IQR) = 112.5] than face-to-face consultations (median = 366, IQR = 93.8), and in terms of duration (19:40 minutes vs. 20:10 minutes). These differences were not statistically significant for length (U = 10.0; p = 0.20) or duration (U = 18.0; p = 0.10). In the antenatal setting, there was little difference in the length of virtual consultations (median = 167, IQR = 125.5) and face-to-face consultations (median = 168, IQR = 76.3). The duration of consultations measured in minutes showed virtual consultations to be slightly shorter (7:38 minutes) than face-to-face consultations (8:13 minutes), although this difference did not reach statistical significance (U = 15.0; p = 0.63). In the Cancer Surgery clinic, virtual consultations were non-significantly shorter in duration (11:50 minutes for remote consultations vs. 19:40 minutes for face-to-face consultations, U = 6.0; p = 0.23) and had a (non-significantly) lower number of utterances (median = 192, IQR = 69.5) than face-to-face consultations (median = 217, IQR = 142.5, U = 9.0; p = 0.47).

A consultation that is ‘to the point’ would focus predominantly (although certainly not exclusively) on task-oriented clinical talk. Figures 9–11 show the overall proportion of types of talk for virtual and face-to-face consultations across the three clinics, according to the coded RIAS clusters. Taken together, they show that, aside from a small proportion of time spent on technology-related matters in virtual consultations, the proportions of talk were very similar across both kinds of consultation – although this may be partly attributable to the fact that the patient samples for virtual and face-to-face consultations were closely matched for type of condition and type of appointment. As noted above, the ad hoc consultations (which addressed a very different patient group – ‘hard to reach’, with unstable or complex needs, and/or seeking urgent input from the clinician) were not captured in our micro-level data set.

FIGURE 9.

Types of talk in the Adult/Young Adult Diabetes clinic. (a) Virtual consultations; and (b) face-to-face consultations.

FIGURE 10.

Types of talk in Antenatal Diabetes. (a) Virtual consultations; and (b) face-to-face consultations.

FIGURE 11.

Types of talk in Cancer Surgery. (a) Virtual consultations; and (b) face-to-face consultations.

These proportions are illustrated in more detail for the three clinics in Tables 8–10, which compare the frequencies of socioemotional talk, task-focused clinical talk, process talk and technology-related talk in virtual and face-to-face consultations.

TABLE 8 - Clinician and patient talk in virtual and face-to-face consultations in the Adult/Young Adult Diabetes service

N/A, not applicable.

RIAS categories	Talk in each type of consultation, median (IQR)
	Remote			Face to face
	Clinician talk	Patient talk	Total	Clinician talk	Patient talk	Total
Socioemotional	72 (26.5)	55 (34.0)	120 (51.5)	54 (30.0)	88 (71.8)	117 (71.0)
Social behaviour	15 (12.8)	11 (8.8)	26 (22.0)	19 (14.0)	18 (24.5)	36 (37.3)
Verbal attentiveness	25 (16.0)	27 (8.3)	50 (20.0)	21 (9.3)	49 (26.0)	65 (44.3)
Concern/reassurance	22 (8.8)	18 (29.8)	39 (35.5)	19 (20.0)	8 (13.5)	31 (24.0)
Negative talk	0 (1.3)	0 (0.3)	0 (1.5)	0 (2.5)	2 (6.5)	2 (8.3)
Task focused	82 (38.8)	82 (49.8)	170 (53.5)	122 (24.5)	74 (34.3)	206 (4.5)
Asking questions	24 (23.5)	1 (4.0)	25 (20.5)	30 (22.3)	1 (2.0)	33 (22.5)
Information-giving	44 (21.0)	80 (49.5)	129 (59.0)	66 (15.3)	68 (54.3)	141 (44.8)
Counselling	5 (4.0)	N/A	5 (4.0)	13 (7.8)	–	13 (7.8)
Requests	12 (13.0)	0	12 (13.0)	15 (7.0)	1 (2.3)	15 (7.0)
Process oriented	31 (21.5)	3 (4.0)	35 (21.5)	29 (8.3)	7 (9.5)	35 (11.8)
Direction	12 (8.5)	1 (3.5)	12 (7.5)	10 (11.8)	2 (2.8)	12 (11.5)
Check understanding	18 (13.5)	2 (3.3)	22 (13.3)	18 (7.3)	5 (6.8)	23 (9.0)
Technology related	1 (6.0)	1 (2.8)	2 (8.8)	–	–	–
Socioemotional	0 (1.5)	0 (1.0)	0 (2.5)	–	–	–
Task focused	0 (0.5)	0 (0.3)	0 (0.8)	–	–	–
Process oriented	1 (3.0)	0 (1.8)	2 (4.0)	–	–	–
Total	181 (42.3)	143 (84.3)	337 (112.5)	204 (38.8)	173 (82.5)	366 (93.8)

TABLE 9 - Clinician and patient talk in virtual and face-to-face consultations in the Antenatal Diabetes service

RIAS categories	Talk in each type of consultation, median (IQR)
	Remote			Face to face
	Clinician talk	Patient talk	Total	Clinician talk	Patient talk	Total
Socioemotional	35 (44.5)	38 (39.0)	74 (80.8)	43 (24.8)	36 (26.0)	83 (38.0)
Social behaviour	7 (4.8)	5 (3.3)	12 (8.3)	7 (2.3)	4 (3.3)	12 (4.5)
Verbal attentiveness	21 (25.3)	25 (5.0)	46 (57.0)	17 (16.8)	26 (19.5)	34 (49.3)
Concern/reassurance	10 (9.8)	8 (13.0)	20 (16.0)	18 (9.5)	6 (5.8)	27 (14.5)
Negative talk	0 (0.3)	0 (0.8)	0 (1.0)	0	0	0
Task focused	37 (27.5)	29 (19.0)	66 (48.5)	42 (22.3)	23 (24.2)	73 (30.2)
Asking questions	7 (7.3)	3 (2.5)	11 (5.8)	10 (3.8)	3 (1.8)	12 (4.8)
Information-giving	16 (20.3)	21 (19.5)	42 (26.0)	20 (12.3)	18 (13.8)	40 (34.8)
Counselling	5 (3.5)	–	5 (3.5)	5 (6.0)	–	5 (6.0)
Requests	6 (5.0)	0	6 (5.0)	11 (7.5)	1 (1.8)	13 (7.5)
Process oriented	6 (8.0)	2 (2.5)	8 (10.3)	11 (12.8)	1 (2.8)	14 (13.3)
Direction	2 (1.5)	1 (2.0)	3 (3.0)	8 (8.8)	0 (0.3)	8 (8.0)
Check understanding	5 (6.3)	2 (4.0)	7 (10.3)	4 (3.5)	1 (2.5)	6 (6.3)
Technology related	5 (6.3)	3 (3.5)	7 (9.0)	–	–	–
Socioemotional	2 (3.3)	2 (3.0)	4 (3.3)	–	–	–
Task focused	1 (3.3)	0 (1.0)	1 (4.0)	–	–	–
Process oriented	2 (1.3)	1 (1.3)	3 (2.5)	–	–	–
Total	89 (66)	77 (59.5)	167 (125.5)	103 (38.8)	69 (51.3)	168 (76.3)

TABLE 10 - Clinician and patient talk in virtual and face-to-face follow-up consultations in the Hepatobiliary and Pancreatic Cancer Surgery service

RIAS categories	Talk in each type of consultation, median (IQR)
	Remote			Face to face
	Clinician talk	Patient talk	Total	Clinician talk	Patient talk	Total
Socioemotional	23 (46.5)	35 (34.5)	77 (35.0)	31 (39.0)	49 (38.0)	71 (72.5)
Social behaviour	6 (12.0)	10 (16.0)	13 (26.0)	7 (14.0)	7 (24.0)	12 (36.0)
Verbal attentiveness	12 (27.5)	19 (20.0)	46 (19.0)	15 (17.0)	23 (14.5)	39 (21.5)
Concern/reassurance	7 (7.0)	9 (6.0)	12 (11.0)	5 (11.5)	6 (11.5)	9 (21.5)
Negative talk	0	0 (0.5)	0 (0.5)	0 (0.5)	1 (2.5)	2 (2.5)
Task focused	42 (40.5)	33 (26.5)	73 (39.0)	70 (38.5)	35 (44.5)	114 (63.0)
Asking questions	9 (6.5)	2 (9.5)	13 (12)	17 (10.0)	2 (11.0)	28
Information-giving	23 (29.5)	17 (28.5)	56 (32.5)	39 (29.5)	36 (46.0)	75 (45.0)
Counselling	0 (1)	–	1	1 (10.0)	–	1
Requests	4 (6.5)	0 (1.0)	4 (7.5)	4 (8.5)	0 (0.5)	4 (9.0)
Process oriented	9 (14.5)	5 (6.5)	15 (20.5)	19 (16.5)	4 (5.5)	23 (21.0)
Direction	8 (12.5)	2 (3.0)	10 (15.5)	8 (9.5)	1 (2.5)	9 (9.5)
Check understanding	2 (2.5)	3 (4.0)	5 (5.0)	8 (10.5)	3 (5.0)	10 (14.0)
Technology related	8 (8.5)	4 (13.5)	12 (22.0)	–	–	–
Socioemotional	3 (4.0)	3 (9.0)	6 (13)	–	–	–
Task focused	1 (3.0)	1 (1%)	2 (5.5)	–	–	–
Process oriented	3 (4.5)	1 (2.5)	3 (6.0)	–	–	–
Total	108 (48.5)	84 (21.0)	192 (69.5)	137 (62.5)	72 (57.5)	217 (142.5)

Across the three clinics, the most common kinds of talk for clinicians and patients in both face-to-face and virtual consultations were verbal attentiveness and information-giving.

Clinicians exhibited more socioemotional talk in virtual consultations than face-to-face consultations, which challenges the hypothesis that virtual consultations are perhaps more ‘to the point’.

The dominance of talk (i.e. who spoke more) was calculated by dividing the number of utterances for clinician talk by the number of utterances for patient talk. In the Adult/Young Adult Diabetes consultations, clinician dominance was the same in face-to-face (median = 1.2, IQR = 0.7) and virtual consultations (median = 1.2, IQR = 0.6). In the Antenatal Diabetes clinic, consultations were slightly clinician-dominated, but less so for virtual consultations (median = 1.2, IQR = 0.3) than face-to-face consultations (median = 1.7, IQR = 0.5) – a difference that was statistically significant (U = 3.5; p = 0.018). The Hepatobiliary and Pancreatic Cancer Surgery clinic showed a greater level of clinician dominance for face-to-face consultations (median = 1.4, IQR = 0.5) than for virtual consultations (median = 1.3, IQR = 1.8), but this was not statistically significant (U = 10.0; p = 0.59). The one difference that was statistically significant was probably attributable to the fact that patients in virtual Antenatal Diabetes consultations spent some time reading out their blood glucose results to the clinician.

There was no significant difference between virtual and face-to-face consultations in ‘directedness’ (what the RIAS calls the ratio of clinician-to-patient ‘control’ over the consultation, and which was calculated by dividing the patient’s biomedical questions and biomedical information from the clinician by the clinician’s closed questions, directions and biomedical information from the patient). The average directedness score was 0.82 for face-to-face consultations and 0.83 for virtual consultations in the Antenatal Diabetes setting, 0.5 for face-to-face consultations and 0.7 for virtual consultations in the Adult/Young Adult setting, and 0.9 for face-to-face consultations and 1.0 for virtual consultations in the Cancer Surgery follow-up setting (none of these differences was statistically significant). These figures suggest that differences in directedness between clinical conditions (and linked to a particular clinician’s style of consulting) appear to be far greater than any differences associated with the virtual technology.

Question 2: how do virtual consultations differ from face-to-face ones in the kind of non-technology talk that occurs?

The main finding from our RIAS analysis was that broadly similar kinds of talk occurred in similar proportions in virtual and face-to-face consultations. To explore that in finer detail, we removed ‘technology talk’ from the analysis and compared a broader range of categories of talk across patient and clinician talk in virtual and face-to-face consultations. The findings are presented in Figure 12.

FIGURE 12.

Total patient and clinician talk, with technology talk removed from remote consultations. (a) Remote consultations; and (b) face-to-face consultations.

Notwithstanding the small sample size (which could have concealed real differences between the two modes of consulting) and excluding technology-related talk, the only significant differences that emerged related to the following areas: ‘personal talk’ (constituent category for ‘social behaviour’); ‘verbal attentiveness’ (a subcategory of socioemotional talk); direction and orientation (subcategories of process talk); counselling, requests and exchanging and sharing information (subcategories of clinical task-oriented talk); pauses and silence; and achieving closure. We consider these in turn.

Social behaviour and personal talk

Across all clinical settings, the number of utterances related to the social behaviour category was slightly higher for virtual consultations (median = 17, IQR = 15.5) than face-to-face consultation settings (median = 14, IQR = 31.5), but there was no significant difference across the 34 cases (U = 138.5; p = 0.84.) Analysis of the subcategories within this cluster revealed a significantly higher proportion of utterances for ‘personal talk’ (e.g. ‘hello’, ‘how are you’?) during virtual consultations (median = 9, IQR = 4.0) than during face-to-face consultations (median = 4, IQR = 5.8) in the antenatal setting only (U = 4.0, p = 0.02). This difference may be partly explained by the fact that the initial interaction between the patient and the clinician in face-to-face encounters is done in the waiting area and during the walk to the consultation room (in other words, the difference was probably spurious). There was very little ‘chit-chat’ in either virtual consultations or face-to-face consultations in this context of Antenatal Diabetes follow-up. This may have been because of high patient numbers – a clinician typically saw 30–35 patients per clinic – and to the likely absence of an established or prolonged relationship (since most patients had recently diagnosed gestational diabetes). Personal talk in this context largely consisted of social gestures and niceties (e.g. ‘how are you?’, ‘bye’, ‘see you soon’), rather than social conversation. For instance, establishing a Skype connection (and ‘popping up’ on each other’s computer screens) would sometimes lead to laughter, greetings and waving.

Verbal attentiveness

Across the three clinics, clinicians presented more verbal attentiveness talk (in the form of agreement and back-channelling) for virtual consultations (median = 24, IQR = 20.5) than for face-to-face consultations (median = 17, IQR = 14.0), but this difference was not statistically significant (U = 115.0; p = 0.31). Conversely, patients presented more verbal attentiveness in face-to-face consultations (median = 37, IQR = 26.0) than in virtual consultations (median = 26, IQR = 14.0), which was statistically significant (U = 87.5; p = 0.02). The change in proportion of verbal attentiveness may be an artefact of the difference in talk dominance (see the discussion in Question 1: are virtual consultations shorter and more ‘to the point’ than face-to-face ones?) and also indicates (through the higher proportion of clinician back-channel talk) that the patient is more likely to take the floor and lead on information provision when communicating virtually.

Direction and orientation

Clinicians gave more direction talk during face-to-face consultations (median = 8, IQR = 9.0) than during remote consultations (median = 6, IQR = 11.5), but this difference was not found to be significant (U = 98.0; p = 0.11). However, analysis of the different clinic settings showed that the clinician in the antenatal setting did present significantly more direction talk during face-to-face consultations (median = 8, IQR = 8.8) than during virtual consultations (median = 2, IQR = 1.5, U = 0.05; p = 0.01). Analysis of the subcategories showed that this difference was attributable to a significantly higher occurrence of ‘transition’ statements by the clinician during face-to-face consultations in the Antenatal Diabetes setting (median = 7, IQR = 2.8) than that in remote consultations (median = 1, IQR = 1.8, U = 0.5; p = 0.01). For example, during face-to-face consultations, the clinician played a more active role using placeholder statements (e.g. ‘Right, let’s take a look [at your maternity folder]’, ‘Hang on, I’m just going to get the nurse to take a look at your rash’). These differences were unsurprising, and probably relate to the different physical environments, material artefacts and staff availability in the different types of consultation.

During face-to-face consultations in the Antenatal Diabetes clinic, patients did not use any talk that could be coded as giving direction, orientation or instruction. In contrast, in some virtual consultations, the patient did give direction or instruction. Much of this talk related to the patient’s role in communicating their blood glucose readings to the clinician over Skype and writing the details in their maternity folder (e.g. referring to their own notes or getting a pen to write things down). However, this difference in ‘direction’ talk was not statistically significant (U = 11; p = 0.18).

Counselling

Although not statistically significant (U = 108.0; p = 0.22), the proportion of ‘counselling talk’ from clinicians was greater during face-to-face consultations (median = 5, IQR = 11.0) than during virtual consultations (median = 4, IQR = 5.5), across all clinical settings. The reasons for this difference varied depending on the clinical context. In the diabetes setting, the clinician often had direct access to the patient’s notes on blood sugar readings and their insulin pumps, which may have facilitated a dialogue on managing their condition. During some of the Cancer Surgery clinics, the clinician conducted a physical examination of the patient if they reported any pain or symptoms, which may explain the higher levels of education and counselling talk in that setting.

Requests

In the Antenatal Diabetes setting, requests (asking for something) were significantly more frequent during face-to-face consultations (median = 13, IQR = 7.5) than during virtual ones (median = 6, IQR = 5.0, U = 4.0; p = 0.02). This difference was accounted for almost entirely by a higher rate of the clinician asking the patient to confirm their understanding in face-to-face consultations (median = 8, IQR = 3.0) than that in virtual consultations (median = 2, IQR = 4.5), which was found to be statistically significant (U = 2.0; p = 0.01). The majority of this talk was characterised by an inflected word or phrase (e.g. ‘Okay?) that invited the patient to indicate their understanding following an ‘information-giving’ or ‘counselling’ statement. Some face-to-face consultations involved the patient requesting a prescription (note that unless the patient and clinician were co-located, it would not be possible for the patient to obtain a printed and signed prescription, and electronic prescriptions were not in use in this setting).

Asking questions

The proportion of question-asking from clinicians in each of the clinics was similar for virtual and face-to-face consultations. However, in the Antenatal Diabetes setting, there were significant differences with regard to open versus closed medical questions. Closed questions included direct questions seeking specific information, whereby short responses are generally the only response options (e.g. ‘Are you just taking one tablet of the metformin?’). Open questions were characterised by their non-specificity and/or probing intent (e.g. ‘How was the scan?’). Most clinician questions in virtual consultations were open ended (median = 2, IQR = 3.8) as opposed to closed (median = 1, IQR = 3.0). In face-to-face consultations, the opposite was the case; the median number of open-ended questions was 2 (IQR = 1.3), and the median number of closed questions was 4 (IQR = 2.5). The higher occurrence of closed medical questions in virtual consultations was statistically significant (U = 5.5; p = 0.04). These differences in the occurrence of open and closed questions appeared to be largely attributable to the differences in the spatial and material aspects of consultations. In particular, the maternity folder could be shared and viewed by the clinicians during face-to-face appointments, whereas virtual consultations relied on the patient to verbally represent and summarise this information – often in response to questions from the clinician.

Sharing information

Tables 11–13 show a breakdown of the information shared between patients and clinicians in virtual and face-to-face consultations across the three clinical settings. In the RIAS, medical information relates to statements of fact or opinion relating to the medical condition, symptoms, diagnosis, prognosis, past tests, test results and medical background of the patient. Therapy information relates to statements of fact or opinion regarding the ongoing or future (beginning with this visit) treatment plan, such as information relating to medications used or drug regimen. Lifestyle information relates to statements of fact or opinion relating to lifestyle (smoking, diet, sleep, alcohol and exercise habits), family and home situations, work or employment, health habits and self-care issues. Psychosocial information relates to concerns or problems such as stress, feelings and emotions, values and beliefs. Other information relates to talk that does not fall into the other categories, and includes neutral statements about the consultation (e.g. ‘today’s date is . . .’).

TABLE 11 - Median (IQR) for information-giving talk in face-to-face and remote consultations in the Adult/Young Adult Diabetes service

Information	Type of consultation
	Virtual	Face to face
Medical condition
Clinician	9 (10.0)	16 (16.0)
Patient	26 (27.0)	31 (23.0)
Therapy
Clinician	24 (17.8)	39 (10.0)
Patient	18 (14.5)	26 (15.0)
Lifestyle
Clinician	0 (1.25)	1 (2.3)
Patient	10 (20.3)	8 (7.0)
Psychosocial
Clinician	0 1 (1.5)	1 (2.0)
Patient	5 (17.3)	5 (7.0)
Other
Clinician	1 (4.8)	5 (6.0)
Patient	6 (8.5)	6 (14.5)

TABLE 12 - Median (IQR) for information-giving talk in face-to-face and remote consultations in the Antenatal Diabetes service

Information	Type of consultation
	Virtual	Face to face
Medical condition
Clinician	5 (1.8)	4 (6.8)
Patient	16 (16)	12 (8.8)
Therapy
Clinician	15 (19.0)	11 (11.3)
Patient	2 (4.0)	4 (3.5)
Lifestyle
Clinician	1 (2.3)	1 (2.0)
Patient	1 (2.3)	3 (9.5)
Psychosocial
Clinician	1 (1.3)	1 (1.0)
Patient	1 (1.3)	0 (1.0)
Other
Clinician	1 (2.5)	0 (2.0)
Patient	1 (2.3)	0 (1.3)

TABLE 13 - Median (IQR) for information-giving talk in face-to-face and remote follow-up consultations in the Hepatobiliary and Pancreatic Cancer Surgery service

Information	Type of consultation
	Virtual	Face to face
Medical condition
Clinician	11 (14.0)	12 (11.5)
Patient	8 (21.5)	13 (19.5)
Therapy
Clinician	12 (19.5)	21 (22.5)
Patient	6 (13.5)	7 (12.5)
Lifestyle
Clinician	0	0 (1)
Patient	1 (2.5)	2 (6.5)
Psychosocial
Clinician	0 (0.5)	1 (3.0)
Patient	2 (5.0)	1 (3.5)
Other
Clinician	0	2 (4.0)
Patient	1 (1.5)	2 (6.0)

Although there was much information shared in virtual and face-to-face consultations, the kinds of information shared differed across clinics. In the Antenatal Diabetes service, there was a higher proportion of information-giving talk from the clinician in virtual consultations (88%) versus face-to-face consultations (65%) about therapeutic regime – perhaps because instructions and advice that could remain implicit in a face-to-face interaction were made explicit (or explained in more detail or repeated) in the virtual interaction. A far smaller proportion of clinician information-giving related to lifestyle (4% compared with 0.3%) or psychosocial issues (4% compared with 0.2%) in virtual consultations. This may have been a result of the clinician having direct access to the patient’s medical information (blood glucose readings, insulin doses) and being able to direct questions to broader health-related issues in face-to-face consultations. In contrast, virtual consultations often required the verbal exchange of medical information, with the patient providing more information about their medical condition.

Patients took on a greater role in providing medical information during virtual consultations, mainly in order to convey the data they had collected to the clinician. The verbalisation of patient-generated data resulted in a greater level of affective talk on the part of the patient (e.g. optimism or concern about their blood glucose readings) in the virtual setting than in the face-to-face setting, although the differences did not reach statistical significance.

These differences in the types of talk across the different clinical settings were not statistically significant.

Examples of differences in information-sharing are given in Tables 14 (for a face-to-face consultation) and 15 (for a virtual consultation). In the exchange in Table 14, the clinician views the patient’s blood glucose results and insulin dosages (which are not verbalised by either party), as well as the result of an ultrasound scan. Her talk is mostly focused on commenting on these results and suggesting a change in treatment. In contrast, much of the talk in the virtual consultation (see Table 15) consists of the patient reading out both her blood glucose results and her (varying) insulin dosages from her own record and expressing concern about the former, while the clinician records the patient’s readings on paper in the clinic before commenting and suggesting a change in treatment.

TABLE 14 - Example of questioning and information-sharing during a face-to-face Antenatal Diabetes consultation

Speaker	Questioning/information-sharing
Clinician and patient sit together with the maternity folder open on the clinician’s desk, including written blood glucose readings and insulin levels
Doctor	Great, so . . . So we’ve met before
	And you are, 30 weeks?
Patient	Yes, 30 weeks
Doctor	Great
	And you are taking insulin and metformin?
Patient	Yes
	[0.19]
	And you had a scan 2 weeks ago?
Patient	Yes, 2 weeks
Doctor	It looks good
Patient	Yes, this month
Doctor	Fine
	[0.16]
	Brilliant. How have the sugars been?
Patient	Sometimes controlled, sometimes a little bit up
Doctor	Ok, all right
Patient	I increased, a little bit, my insulin
Doctor	Perfect. In fact I wonder if you can go up just a little bit more
	I wonder about 8. How would you feel? Or even 9

TABLE 15 - Example of questioning and information-sharing during a virtual Antenatal Diabetes consultation

Speaker	Questioning/information-sharing
Doctor	Now how have your sugars been since I saw you on Friday?
Patient	Not great to be honest
Doctor	OK
Patient	So, what was Friday’s date, it was 14th wasn’t it
	And you [. . .]
	To be fair, after I left you we had Chinese and we had dim sum and I thought we made some fairly OK choices but apparently not because my sugars was 11.9
Doctor	Yep
Patient	And then after dinner it was 10.4
Doctor	Yep
Patient	Erm, and then
	The following day, it was 28
	So the insulin I took was 28, 2834, 34
Doctor	Yep
Patient	And I misread the . . . It was supposed to be 30 after dinner, or 32 after dinner, and I misread it as 34 so I just took that
	OK
	And my numbers were 7.2, 7.2, 8.3 and 6.9
Doctor	Lovely, OK
	Yep
Patient	And then the following day
	I took 28, 30, 30, 34
Doctor	Yep
Patient	And my numbers were 6.8
Doctor	Yep
Patient	8.4, 7.8 and 8.5
Doctor	Yep
Patient	So again, not ideal

Pauses and silence

There were significantly more pauses (a silence of 2 seconds or more) during face-to-face consultations (median = 3, IQR = 4.5) than during virtual consultations (median = 0.0, IQR = 1.0, U = 58.5; p = 0.002). The reasons for this were not immediately apparent. However, it is noteworthy that many of the pause events involved the clinician interacting with paper materials (e.g. reading and writing clinical notes) and interacting with the computer (e.g. accessing the EPR, checking appointment schedules). It is possible that the shared physical presence in face-to-face consultations allows such activities to be implicitly understood and made aware of (e.g. a patient may stop talking if the clinician turns to the computer). In virtual consultations, in contrast, it appears that the clinician may need to indicate their action through transition talk (e.g. ‘I will just check your appointment time on the computer’) or the patient may continue to talk while the clinician views the information on their screen.

Closing the consultation

The closing phase of virtual consultations was usually clearly defined and always marked the end of the appointment. In contrast, in face-to-face consultations, the clinician often directed the patient to the reception desk or asked them to wait for one of the nurses (e.g. to address additional clinical concerns). Hence, the end of a face-to-face appointment was often less clearly defined and ‘closing’ comments or social gestures were less appropriate.

Because the virtual Antenatal Diabetes service was still at a pilot stage during this research, the closing phase typically also included making a new appointment, as illustrated in Table 16. Such interactions (which contrasted with appointment-booking in the main diabetes clinic) illustrate how the remote service was not fully embedded in the ‘business as usual’ of the Antenatal Diabetes clinic.

TABLE 16 - Example of discussion about the use of Skype during the closing stages of a virtual consultation

Speaker	Discussion
Doctor	I’ll tell you what, do you want – cause I’m away next week would you like to Skype again on Friday?
	Before I go away
Patient	On Friday? OK. Yep
	What –
Doctor	Would you be free on Friday afternoon by any chance?
Patient	That’s fine, what time?
Doctor	Do you have a preference?
Patient	I don’t
Doctor	Um, OK well I’ll be doing clinic in the morning
	I’m just checking my diary
Patient	OK
Doctor	Um
	[0.7]
	So what I could do is aim to Skype you at maybe 2 o’clock, or something?2 o’clock, OK
	Does does that sound reasonable? And if I have a problem with that I’ll message you on Skype
Patient	OK. Alright, that sounds good

Summary

In summary, and based on a small subsample of consultations analysed in depth, virtual consultations appeared to be slightly shorter and slightly less clinician dominated than, but as equally clinician controlled as, face-to-face consultations. Apart from technology-related utterances, the kinds of talk were broadly similar in the two settings, but some subtle differences in certain categories of talk were evident.

Our analysis suggests that the main differences emerging between face-to-face and virtual consultations were a consequence of social, physical and material differences between the two settings (e.g. the absence of shared maternity notes had an impact on the direction of questioning and information-sharing in Antenatal Diabetes consultations). The physical setting of the clinic also provided social and physical arrangements that shaped interaction. Differences in patient–clinician interaction were rarely attributable to video-mediated interaction per se, but rather to contextual differences beyond the video link (e.g. whether or not a nurse was at hand, whether or not the clinician could see the patient’s blood glucose readings). The higher proportion of pauses and mutual silence during face-to-face consultations is one area that requires further investigation. Such moments in a clinical encounter may play an important role in opening up opportunities for the patient to raise concerns.

Theme 1: embedding virtual consultations in a clinical service

It is often stated that new technologies need to be ‘embedded’ in organisational infrastructures and routines. Despite this wide acknowledgement of the need for embedding, our macro-level data suggested that little consideration had been given to this issue by either policy-makers or those in industry. Rather, the use of Skype for virtual consulting was assumed to be a straightforward matter of downloading the software onto the relevant computers – an illustration, perhaps, of the ‘plug and play’ mindset we have described previously in telehealth discourses. 81

A significant storyline in our macro-level data was the financial pressures faced by the public sector and the resulting pressures within the health-care sector to work with fewer resources. Technology was often depicted as a logical route towards achieving cost-savings (e.g. by reducing staff time, supporting caring for people at home) and increasing quality (e.g. through improved access). This powerful overarching narrative (evident in both industry and policy discourses) rarely acknowledged the need for technologies to become embedded within existing organisational structures, processes and routines (and the corresponding need to allocate time and resources to this early phase), nor did they acknowledge that once embedded, technologies can and must continually evolve and adapt over time.

In contrast, it was evident from our meso-level data that the introduction and use of Skype technology for virtual consulting required much time and effort, and many resources (and, in some cases, interpersonal or interdepartmental conflict) to achieve alignment with existing management and administrative structures, processes and systems. This is described in detail in Chapter 4, and included addressing the concerns of the ICT department about IG and workload, reconfiguring the EPR (e.g. to create new appointment slots that would indicate that they were ‘webcam’ appointments), and aligning with the routines of appointment scheduling and recording attendance.

Embedding of the technology within existing work routines and practices required high-level commitment across the team and mutual awareness of what other collaborators were doing in order to provide a context for each party’s own activity. 228 Strong clinical leadership and ‘championing’ the new technology and mode of service delivery appeared to be key to overcoming the multiple and interacting barriers on both a small and larger scale. Clinical leaders offered a meso-level ‘organising vision’229 for the introduction and roll-out of the remote approach that provided a shared rationale for why Skype technology should be adopted/used and activated, and co-ordinated stakeholders to promote adoption and diffusion. At the clinic level, in particular, this input helped to ensure that the team functioned collaboratively and effectively towards the shared goal of making virtual consultation services business as usual. Our data illustrated starkly that a high workload and lack of organisational slack (time, space, people, expertise) severely limited front-line teams’ capacity to initiate and successfully maintain the embedding of virtual consultations in clinic work.

Theme 2: selecting/inviting patients to use the virtual consultation service

At the macro level, the prevailing policy assumption was that once a virtual service was established as business as usual, it would be a viable, affordable (indeed, cost-saving) option for a high proportion of patients in any given service. For example, the National Information Board stated that:

Better use of data and technology has the power to improve health, transforming the quality and reducing the cost of health and care services.

National Information Board (p. 3). 5

Contains public sector information licensed under the Open Government Licence v3.0)

This was also the case for industry strategists. For example, in an interview with a leading IT provider, the focus was largely on particular products and their potential to be bought into the NHS:

And for, you know, videoconference and instant message functionality generally you think, ‘oh my goodness, it would be so easy to do’. And everyone would be on the same system so it would work. Instantly. There would be no interoperability issues.

The above quotations illustrate that, for national-level stakeholders, the strategic terrain (i.e. the different aspects of the external world as viewed by these actors) was shaped by what we have described elsewhere81 as a ‘modernist’ organising vision for technology-enabled care delivery. A modernist vision has the following features: technological innovations are rational solutions to the challenges facing the health service (especially rising rates of non-communicable diseases and escalating patient demand) and they will help to achieve efficiency and cost-effectiveness of services, and have the potential to ‘empower’ both clinicians and patients.

The quotations above – and a modernist vision for technology-enabled care more generally – implies three assumptions that appear to be relatively unexamined by both national policy-makers and industry providers. First, that new technologies (specifically in this case study, virtual consulting) are an effective way of driving new and more efficient models of care into practice at scale, thereby reducing the high and rising burden of outpatient appointments. Second, that when planning the provision of a technology-supported service in a particular condition, the clinical condition can be considered in broad-brush terms (i.e. as a textbook ideal type, as in ‘patients with diabetes in pregnancy’). Third, that the patient population can generally be treated as more or less homogeneous in terms of their wider characteristics (or, perhaps, that any differences can be construed as ‘gaps’ – such as digital literacy – that can be ‘filled’ by education).

At the meso level, the original introduction of virtual appointments for diabetes at Barts Health NHS Trust back in 2009 was not unrelated to the growing pressures on outpatient services and the year-on-year reduction in budgets. But it was driven more immediately by the high DNA rate among young adults (around 50%), and by a belief of the lead consultant and diabetes specialist nurse that large numbers of young adults were becoming disengaged from the service and defaulting from follow-up. In particular, a virtual service was considered to be a potential way to retain ‘harder to reach’ patients, such as black and minority ethnic and lower socioeconomic groups (who were, for example, less likely to be allowed paid time off work for appointments, and who struggled to meet the cost of transport to hospital). The clinicians’ view was that the use of technology would make the service more accessible to a group of patients whom they viewed as particularly vulnerable. For example, and as part of a general strategy of making the service more accessible to patients who were reluctant to engage, they encouraged patients to message them on Skype and spent time responding to these messages; the virtual service included a growing proportion of patient-initiated consultations, in which patients would see that the doctor or nurse was online and send a Skype message asking for a non-booked consultation. Earlier publications on the virtual diabetes service at Barts Health NHS Trust emphasised how a ‘disengaged’ clinic subpopulation had been successfully re-engaged by offering the service and making it possible for such patients to contact the doctor or nurse at their convenience, whenever and wherever they sought advice or support. 9,11

The extension of the virtual consulting model to Antenatal Diabetes services was driven partly by the need to address steadily rising patient numbers in the Antenatal Diabetes outpatient clinic (the incidence of gestational diabetes is increasing in parallel with growing levels of obesity, for example). However, clinicians were also motivated to try to reduce the burden on pregnant women of making frequent visits to a busy hospital clinic, perhaps with toddlers in tow. Gestational diabetes can be labile and need daily adjustments in insulin dose, and hence become burdensome and potentially difficult for the (otherwise well) pregnant woman. In keeping with the idea that pregnancy is a normal physiological state, not an illness, both midwives and doctors seek to avoid ‘medicalising’ the condition if at all possible.

Similarly, the clinical lead of the Hepatobiliary and Pancreatic Cancer Surgery service was keen to reduce the burden on his patients of making long journeys for a clinic check-up in the postoperative period. As long as there was no clinical reason for the patient to be seen face to face, it was viewed as clinically more appropriate for the patient to stay at home and be seen remotely.

Thus, when making the (meso-level) decision to introduce a virtual consultation service, clinical leads saw the strategic terrain at least partly in terms of improving the quality and accessibility of the service for patients, for whom a journey to the hospital was – variously – unnecessary, inconvenient, unaffordable or clinically inadvisable. Such drivers were rarely mentioned by national-level stakeholders (whose economic models did not generally include patient-borne costs or patient time).

Although the doctors (and to some extent, nurses and midwives too) in participating clinical teams viewed the virtual consultation service as improving accessibility and increasing choice for patients in general, they did not view this service as automatically available to all clinic patients. Rather, and in every particular case without exception, a clinical judgement was made on the ‘suitability’ and ‘appropriateness’ of the patient for the virtual services.

Decisions around suitability were made on a case-by-case basis and informed by a range of factors that varied across the different clinical contexts, and were discussed and mutually agreed with the patient. Such decisions appeared to be difficult to anticipate in advance and to require three things: (1) clinical experience in the relevant specialty, (2) tacit knowledge about how virtual consultations play out in this particular condition (e.g. content to be discussed, materials/resources needed, likelihood or not that a physical examination will be needed) and (3) personal knowledge of the patient. The last of these appeared to involve a subjective assessment of the patient’s health literacy (in general, and specifically, in relation to the clinical condition), IT literacy, the ability to communicate (e.g. across a language barrier) and confidence. The main question driving the decision appeared to be ‘would a virtual consultation be clinically safe for this patient, with this condition, at this time, for this aspect of their care?’ (see the quotation in Chapter 4, New clinical roles and practices: triage, technical support and direct access).

Applying SST to the decision of whether or not to offer the option of virtual consulting to a particular patient, the strategic terrain looks somewhat different from what it looked like for the macro- and meso-level stakeholders. At this micro level, and notwithstanding that this was an experimental service that had not yet been fully tested or incorporated into business as usual, the over-riding influence was a professional norm: primum non nocere (first, do no harm). Our interviews with doctors in particular highlighted the strong sense of clinical responsibility for the patient, along with a sense that they would be held accountable (both professionally and legally) if harm resulted from a virtual consultation. This theme, which was very strong in our micro-level data, was not in evidence in the meso- or macro-level data sets.

Expressed in the language of SST, our data support the conclusion that the clinicians’ general dispositions (based on education, experience, professional norms, values and so on) were oriented to providing a virtual service that they saw as accessible and convenient for patients, and which may also save money and help to ease the pressures on the outpatient clinic. But these general dispositions were often (and in the case of Antenatal Diabetes patients, almost always) over-ridden by the clinician’s conjuncturally specific assessment of whether or not a virtual consultation was appropriate for this patient in this situation, given these contingencies. Furthermore, it was only when the service had been introduced on a pilot basis and the clinicians began to make these fine-grained decisions that the non-suitability of virtual consulting for many patients became apparent.

The high barrier in some specialties to offering the virtual consulting option contrasted starkly with the view of our patient advisory panel that all patients should be offered the virtual option so that the service was available to all patients who chose to use it. This view was strongly and universally held by those participating in discussions on the topic. Implicit in this view was the assumption that it was the patient, not the clinician, who was best placed to make the judgement about the trade-offs between clinical safety and convenience of access. One patient made the point that, although the clinical grounds for ‘suitability’ may be best assessed by the clinician, there were many wider influences (such as the impossibility of getting time off work or the need to collect children from school) that were best assessed by the patient and/or carers.

Using SST, we can say that the strategic terrain for the patient is characterised by a host of practical, material and sociocultural influences, along with various commitments and accountabilities (to employers, other family members and so on). For the patient, the care of the clinical condition cannot be considered in isolation from the wider landscape of things at stake.

Theme 3: doing a remote consultation

‘Doing’ a remote consultation involved three phases: first, establishing connection (technical); second, the actual encounter (clinical); and third, arranging follow-up (administrative). Our macro-level data suggest that neither policy-makers nor industry stakeholders paid much attention to the detail of the virtual consultation. Apart from recognising the obvious fact that certain aspects of the physical examination would be impossible in the remote encounter, they appeared to assume that the latter would be unproblematic and unfold in a more or less similar way, once a successful technical connection had been established; the policy-makers we interviewed did not consider how the clinical encounter would link with the administrative task of booking the next appointment.

Our industry participants were keen to ensure that the technology was appropriately designed and adapted to support a high-quality connection, and that its functionality supported the nuanced affordances that were likely to be needed clinically. For example, they showed interest in reviewing (with consent) selected video-recorded consultations, better understanding how different clinicians and patients use virtual consultations and adapting the Skype software for different health-care settings and clinical contexts (e.g. improved video quality, the use of multiple screens, developing ‘virtual waiting rooms’). Their main focus, however, was on advanced design and functionality.

At the meso level, ‘doing’ a virtual consultation was far from straightforward. As noted in Chapter 4, ‘getting the patient set up’ for such a consultation involved considerable ‘hidden work’ around practical issues, especially providing flexible (and unofficial) IT support, undertaking ‘test calls’ with patients (in which no clinical discussion was held), following-up on ‘contact requests’ (in which the patient had sent a Skype message asking for a contact) and messaging the patient to inform them of their appointment time (or confirm or amend such an appointment). As Figures 3–8 illustrate, even when the virtual consultation service was relatively well routinised (as in the Young Adult Diabetes Clinic), much of this hidden work fell to the clinician.

Our micro-level analysis of the use of Skype to support actual consultations (described in detail in Chapter 5) revealed that consultations unfolded smoothly only when – and to the extent that – both parties engaged both in the technical set-up phase and in an ongoing process of adaptive troubleshooting to ensure that the technology ‘worked’ and the encounter mirrored, as far as possible, the conventional clinical encounter in that specialty. In the VOCAL study, the clinician was the main (and usually the only) staff member involved in the technical set-up phase before the consultation could begin. His or her role included fiddling with equipment and making adjustments to technology settings in order to resolve problems as and when they arose. Often, clinician and patient worked together to troubleshoot and perform one-off workarounds for unique technical and/or logistical problems.

The introduction of virtual consulting in this particular organisation137 (i.e. this does not necessarily mean that it must happen this way in other organisations) thus required a significant element of ‘reskilling’ and development of know-how among clinical staff to resolve unique technical problems and improvise in new situations. Our ethnographic observations affirmed the previous observation of Brown and Duguid230 that such knowledge is typically exchanged among staff on the clinical team through social interaction, and, in particular, via the exchange of stories (sometimes with humour when describing a technical difficulty and efforts to resolve it).

Arguably, the extension of the clinician role need not necessarily be a prerequisite for delivering virtual consulting (private-sector virtual consulting services, for example, appear to be characterised by a dedicated administrative and technical team who take responsibility for the initial technical set-up phase and ongoing technical/logistical support, while the clinician sticks to a more traditional clinical role231). As Barley137 observed, technology introduced into health-care settings is ‘an occasion for structuring’ – that is, it offers the opportunity for developing new roles and competences, but because of its inherent interpretive flexibility (i.e. technology can be used differently by different agents), it does not produce these new roles in a deterministic way. Whether the tacit knowledge for technical set-up and support is developed by clinicians, administrators or both, it is clear from our findings that organisations should consider how to create the conditions for social interaction that will cultivate and support the sharing of such tacit knowledge and know-how.

The patient was also required to develop and apply a set of technical skills, including the ability to troubleshoot in real time and interact with the clinician to deal with problems as they emerged. Different patients were more or less willing and more or less able to do this. As well as general IT literacy and specific experience and understanding of the Skype technology, factors influencing their capacity to consult remotely included the material properties of their hardware and software and people available to help out with set-up and troubleshooting.

In relation to the clinical aspects of virtual consultations, our study demonstrated that the interpersonal interactions and dynamics had subtle, but potentially important, differences when undertaken remotely compared with face to face (see Chapter 5). Although our sample was small and the findings provisional, it appears that for equivalent clinical interactions, virtual consultations are generally shorter, somewhat less clinician dominated (i.e. the patient had a higher proportion of talk time) and include considerably more technology-related interaction. But these differences were likely to be explained largely, if not entirely, by the physical and material differences between remote and face-to-face consultations, which necessitated the verbalisation of certain aspects of communication in the former, but not the latter. In addition, of course, some consultations (requiring a physical examination) cannot take place virtually.

The clinical aspects of the virtual consultation could not be separated cleanly from the technical and practical aspects. During the virtual consultation, for example, both clinician and patient made extensive use of other artefacts (including paper and electronic records) to support their own input to the interaction and make the other party aware of information that would have been more readily evident in a face-to-face encounter. Workarounds (often invisible or subtle) were frequently needed to adapt roles and processes to the contingent detail of an unfolding consultation. As detailed in Chapter 4, considerable ‘articulation work’ (i.e. the ability to co-ordinate resources, actions and people) was sometimes needed to deal with local contingencies and unanticipated situations.

We also found that an existing relationship and ‘common ground’ between the patient and clinician appeared to be a positive precondition (although perhaps not an absolute requirement) for effective virtual communication. An awareness of these social dimensions of video-mediated communication helped practitioners to evolve their approach to service delivery and relationship formation (e.g. utilising Skype for routine follow-up, developing an initial rapport with patients face to face, inviting patients to message them with questions).

Finally, as described in Chapter 5, the third phase of the virtual consultation – booking a follow-up slot – appeared to depend on the clinician’s willingness to engage with this administrative task, their familiarity and skill with the booking system and their ability and confidence to judge between a face-to-face encounter, a Skype video consultation and a Skype message interaction.

Theme 4: privacy and security

At the macro level, national-level stakeholders from government and industry were concerned about IG only insomuch as they recognised the need to ensure that regulatory requirements were fully met and that public perceptions of data security and privacy were positive. These concerns were generally expressed at a generic level (all digital health initiatives were viewed as raising similar IG challenges) and policy-makers did not appear to be aware of the detail of the specific privacy and security issues around remote video consulting. The national-level policy-makers in our steering group were keen to work with us in producing IG guidance on this topic, with a view to ensuring that the introduction of virtual consulting in other NHS organisations followed a more or less standardised and approved process.

At the meso level, one of the major barriers to the implementation of virtual consulting services within the trust was strong concerns around IG, particularly from the IT department. These concerns appeared to be driven by a perception of strict regulatory and legal pressures, and reinforced by a lack of national consensus on the appropriateness of Skype (and similar virtual media) for use within the NHS. As set out in Chapter 4, this barrier formed a significant part of the action research component of our study, in which we sought to address the prevailing lack of consensus through the creation of guidance and protocol documents (both local and national).

These documents appeared to act in some sense as ‘immutable mobiles’228 or ‘boundary objects’232 that served to support dialogue and co-operation across clinic and administrative departments (and between local teams and national policy-makers). Indeed, the way in which the documents (and the writing of them) brought people together for further discussion was at least as important as the actual list of procedures to be followed. Indeed, although protocols and SOPs certainly provided guidance that staff found helpful, it was also widely believed to be necessary to ‘work around’ official procedure when in the best interests of the patient.

Interestingly, our micro-level analysis showed that although patients and their carers were aware of IG issues at some level, they were generally relaxed about these and rarely, if ever, cited such issues as a reason not to consult remotely. This finding resonated strongly with our previous work on patient concerns about the summary care record, which showed that patients considered the trade-off between potential security breaches (a theoretical possibility, but in practice unlikely) and accessibility and quality of clinical care (potentially enhanced through the use of the new technology), and if they trusted the clinician, they tended to trust the clinician’s use of the technology. 233 Indeed, our interviews with patients revealed that some placed high value on the fact that a trusted clinician was ‘visiting’ their home and sometimes took time to show the clinician aspects of their home (such as posters on a bedroom wall) as a way of developing the established relationship further.

From the clinician’s perspective, the data security aspects of the consultation were taken very seriously. Clinicians were aware of procedures to maintain patient privacy and confidentiality; they applied these flexibly – and sometimes used workarounds in a way that prioritised the patient’s best interests (e.g. using the researcher’s login to ensure access in the face of technical problems).

Theme 5: roll-out and scale-up

Our macro-level analysis of policy documents and interviews with policy-makers (described in detail in Chapter 3) showed that, since at least the early 2000s, there has been a strong policy emphasis on the need for ‘transformation’ of health and social care provision. In these policy narratives, virtual consultations are depicted as having a role in delivering this transformation ‘at scale’ and ‘at pace’. The implication is thus that following an initial set of pilot studies and demonstration projects, the introduction of virtual consultations will gather speed rapidly. In reality, the literature on national roll-outs of ‘transformational’ IT projects in health care suggests that this assumption may be overoptimistic. 69 Indeed, just because a technological innovation has been successfully implemented and embedded in one locality, each new setting will continue to face significant implementation challenges, and there may be a considerable time lag between the adoption of technology and the realisation of any productivity gains.

Interviews with industry representatives from both large and smaller companies indicated a similar conviction in the potential of virtual consultations to deliver transformation at scale and pace. However, although industry representatives welcomed the shift away from the centralised procurement characteristic of the National Programme for IT (2007–11) and the extension of NHS contracts to a wider range of providers, they continued to struggle to engage with health-care organisations and systems. The relationship between the technology industry and the public sector was repeatedly cited as complex, with procurement processes in particular presenting challenges to potential scale-up. Interviewees talked about investment in industry–NHS partnerships as a means of working together to develop and adopt technologies, but recognised this a resource-intensive process with little guarantee of significant (i.e. national) roll-out in the short term.

At the meso level, much of our ethnographic fieldwork and action research consisted of observing and supporting the trust as they strove to overcome the many barriers to mainstreaming virtual consultation services. This initiative was explicitly driven by a strategic focus on transforming care and, as such, had the support of senior executive teams and dedicated resources and staff.

Specifically, the virtual consultation service that had initially developed as an ad hoc project led by interested clinicians within the Diabetes service was aligned with TST, a trust-wide initiative aimed at reducing the number of face-to-face consultations across all clinical services as appropriate. The TST initiative was associated with a significant shift in logistical and technical support, from bespoke support to the innovator department (Adult/Young Adult Diabetes) and two ‘early adopter’ departments (Antenatal Diabetes and Hepatobiliary and Pancreatic Cancer Surgery) to a service that now cuts across (at the time of writing) four local departments and with another four planned. As described in Chapter 4, Trust policy and service-level agreements, the provision of SOPs and central support undoubtedly helped to ‘oil the wheels’ for departments seeking to introduce and embed the new service. However, these central initiatives and resources inevitably also meant a shift towards the centralisation of control, which could potentially conflict with the ‘bottom-up’ and grounded approach we observed in the early adopter departments.

At the micro level, and as described in detail in Chapter 5, our in-depth analysis of virtual consultations has illuminated the numerous context-specific interactions and local contingencies that are integral to providing a successful technology-supported remote consulting service. The implication of these micro-level data are that imposing a centrally driven, standardised solution for implementation and embedding as part of the wider roll-out is unlikely to go smoothly. Indeed, we have recently shown, in using different types of technology as an example, that ‘scale-up’ and ‘roll-out’ should be considered to be social practices (i.e. the uptake of the technology should be studied in the context of what human actors find salient, meaningful, ethical, legal, materially possible and professionally or culturally appropriate in particular clinical or social situations). 136

Conclusions

One clear conclusion from this study is that setting up a virtual consultation service is no bowl of cherries. ‘Embedding’ Skype (or comparable technologies) at the meso level into organisational infrastructure and routines is likely to require not only substantial set-up resources, but also ongoing human effort, as well as time and resources. The lack of recognition of the magnitude of this challenge at the macro level by policy-makers (and, to a lesser extent, industry representatives) is one of the main findings of this study.

A second conclusion is that the adoption, implementation and day-to-day use of virtual consulting is necessarily both adaptive and contingent, involving what are often complex clinical decisions and with a strong over-riding framework of (for the clinician) clinical safety and (for the patient) practicality and personal accountability. In the Young Adult Diabetes service, in particular (which had been running remote consultations for several years when this study was commenced), clinicians had used their autonomy to ‘repurpose’ the Skype technology to facilitate access to care in a way that had not been anticipated at the national level. This adaptive and selective use of Skype at a local level aligns with the classic work by DeSanctis and Poole234 on adaptive structuration theory; it challenges the assumptions (particularly at the government and industry levels) that a ‘digital innovation’ is a fixed and clearly definable option, used in a particular way, that needs to be ‘engaged with’ in a standardised way across the public sector.

Our third conclusion is that the model–reality gap for ‘doing’ a virtual consultation is wide. The virtual consultation was depicted at the policy level as a straightforward and unproblematic task using a reliable ‘plug and play’ technology. The logistical reality of such consultations at the level of the organisation or clinical team (meso) and at the level of the actual interaction (micro) told a different story; clinical work tended to unfold in a unique and messy way, with multiple subtle contingencies (both clinical and technical). Dealing with these contingencies required work and interaction from both clinician and patient (only some of whom had the inclination and capacity to engage with this).

Fourth, although the governance and regulation of virtual consulting is widely perceived as something that can be ‘fixed’ through the development of standards and guidance (and while we ourselves contributed to the development of such guidance as part of this study), our findings suggest that ‘articulation’ (including unofficial workarounds) is part of the reality of virtual consulting. Goorman and Berg235 define articulation as follows:

All tasks involved in assembling, scheduling, monitoring and co-ordinating all of the steps necessary to complete a production task (patient trajectory). This means carrying through a course of action despite local contingencies, unanticipated glitches, incommensurable opinions and beliefs or inadequate knowledge of local circumstances. Every real world system is an open system . . . No formal description of a system (or plan for its work) can thus be complete . . . every real world system thus requires articulation to deal with the unanticipated contingencies that arise. Articulation resolves these inconsistencies by packaging a compromise that ‘gets the job done’ that is, that closes the system locally and temporally so that the work can go on.

Reproduced with permission from Goorman E, Berg M. Modelling nursing activities: electronic patient records and their discontents. Nurs Inq 2000;7:3–9235

What this means is that the generation, endorsement and dissemination of SOPs (a tangible achievement of this study) will not, in and of itself, make virtual consultations happen. Attention also needs to be paid to the messy reality of implementing these standards, in particular, in ‘imperfect’ real-world settings. To help with this, we have provided a set of practitioner resources alongside this report that have been informed by emerging findings, been actively used and refined as part of our action research and are intended to support service providers, professional bodies, policy-makers and patients and carers who are interested in or actively pursuing the development of virtual consultation services.

In relation to the prospects for extending the virtual consultation model to other clinical specialties and other NHS organisations – and notwithstanding the powerful prevailing policy discourse on ‘scale-up’ and ‘roll-out’, which is assumed to be achievable rapidly and ‘at scale’ – our findings indicate that mainstreaming virtual consulting across multiple departments in multiple organisations will be neither smooth nor quick. The clinical and logistical realities will play out differently for different clinical specialties and different hospital departments (not to mention primary care). On the basis of empirical data from the three departments we have studied in detail, our brief exposure to four further departments in the same trust and contact with various groups across Barts Health NHS Trust involved in virtual consultations (e.g. the outpatient project strategy group), as well as on the theoretical literature cited above, we anticipate that each and every department in every organisation that seeks to embed virtual consulting will need to take a sociotechnical systems approach236 to organisational change. Specifically, they will need to emphasise ‘minimal critical specification’ (no more specified than absolutely essential), the ‘sociotechnical criterion’ (problems should be resolved by groups that directly experience them) and ‘incompletion’ (the need for continual revision of objectives and structures). 236

Strengths and limitations of this study

The main strength of this study is that it is, to our knowledge, the first major study of remote video consultations from a sociomaterial perspective that has attempted to link different levels of data collection and analysis across the macro level (the historical and current policy drivers and wider sociocultural, technological and economic context), the meso level (the practicalities of organisational and service change) and the micro level (the fine-grained detail of the actual consultation). In contrast to trials of virtual consultations, which have provided little insight into the organisational complexities of implementing a new technology-enabled service or how such services evolve and (may or may not) become embedded in health-care settings, the naturalistic, mixed-methods and multilevel design of the VOCAL study has enabled the study of the emergence of video consultation services and the challenges faced in this process.

We succeeded in our goal of collecting rich qualitative data that allowed us to go beyond the more usual approach of making simplistic ‘technology on versus technology off’ comparisons with narrow, predefined outcome measures. In short, and notwithstanding the impressive features of the technology that allowed a clear video link to be set up with the patient at home, our ethnographic approach exposed the ‘messy reality’ of establishing a virtual consultation service and illuminated the pros and cons of using this medium for clinical interaction in different settings.

Another advantage of our study design was the prominent action research component. This operated most obviously at the meso level, allowing (indeed, requiring) the research team to work collaboratively with the clinical teams, as well as the trust managers and the ICT department, to help align (for example) the service development, technological development and IG aspects of the ‘embedding’ of Skype consultations. In addition, one member of our steering group was a member of NHS England’s Digital Health team; along with other key informants in our macro-level interviews, he was able to help us to establish a policy dialogue at the national level – particularly in relation to a possible national tariff for virtual consultations. Finally, our key informant interviews with industry representatives also fed back into discussions both locally and nationally about how the industry–NHS partnership over virtual consultations could be strengthened.

The main limitation of the study is its focus on a single NHS organisation, and the sample size for the detailed analysis of virtual consultations was small. In some ways, Barts Health NHS Trust faced a similar set of issues and challenges as any other acute hospital trust during the study period. In other ways, it was unique – or at least, an ‘outlier’ – in that it faced more severe financial pressures than most comparable organisations (as reported in Chapter 2, it was put on ‘special measures’ by the Care Quality Commission part way through our fieldwork). Thus, despite having many cutting-edge technologies and a state-of-the-art electronic record system in use (Barts Health NHS Trust won the ‘Digital Trust of the Year’ award in 2014), the reality on the ground was that there was zero ‘organisational slack’ to support new projects.

As organisational slack is a prerequisite for innovation in any organisation,237 it is likely that organisations under less tight financial and staffing pressures would have found it easier to introduce a remote consulting service. On the other hand, Barts Health NHS Trust was an innovator organisation in introducing such a service on a pilot basis back in 2009, and the innovator clinician (SV) was a co-applicant on this study and a passionate champion across Barts Health NHS Trust for the use of virtual consultations to reduce health inequalities. Organisations lacking ‘innovators’ and ‘champions’ are generally less likely to succeed in complex change projects,237 so it cannot be assumed that our own case study represented the most challenging context.

Although virtual consultations are relatively new to health care, a number of VoIP products have been available for years, and their numbers continue to grow. Our study focused on Skype as the VoIP product of choice within our case study site, as a result of its popularity (many people attached the idea of calling via the internet to Skype) and a good level of familiarity with Skype on the parts of patients and their families. As we set out in our findings chapters, the use of Skype led to accommodations in the social structure and processes of each of the clinical services and the wider trust. However, as Skype is one of a number of services that can support virtual consultations (others including, for example, FaceTime or GoogleTalk™; Google LLC, Mountain View, CA, USA.), it is possible that some of our conclusions are a product of Skype usage, rather than being common to all VoIP applications. It is also worth noting that a researcher was present in either the clinic or the patient’s home when each Skype consultation was set up (primarily to video-record the consultation as data for the study, but simultaneously being viewed by some patients and clinicians as a source of advice and support when seeking to set up a virtual consultation), which may have influenced the set up and progress of that consultation.

Some critics will view it as a limitation of the study that we did not set out to generate an ‘effect size’ or a cost-effectiveness analysis on the use of virtual consultations. This was deliberate. As we believe our findings have demonstrated, virtual consultations cannot be treated like a drug or even as a complex behavioural intervention to be tested ‘on’ patients. Rather, they are the result of a hugely complex sociotechnical system, in which ‘successful’ virtual consulting is contingent on multiple factors at multiple levels. If we appear to have produced ambiguous findings, this is perhaps because ambiguity and tension are inherent to complex sociotechnical systems. To questions such as ‘do virtual consultations work?’, ‘are virtual consultations safe?’ and ‘are virtual consultations cost effective?’, we suggest that the answer will always be ‘it depends’.

Use of findings from this research

The VOCAL study deliberately adopted an action research approach that involved working with national and local stakeholders, helping to identify and address potential blocks (e.g. developing new ways of working to enable Skype to be loaded onto clinic computers), develop processes (e.g. SOPs for those seeking to develop virtual consultation services) and feed into ongoing discussion (e.g. regarding the potential roll-out of virtual consultations via the outpatient project strategy group). This ‘action’ part of the study has generated a number of materials that are already in use within Barts Health NHS Trust and, in some cases, more widely. We have included relevant documents on the NIHR Journals Library website to support those thinking about or setting out to develop virtual consultation services. These resources are evidence based (in terms of both having been developed from research evidence in this study and, where relevant, drawing on documented best practice elsewhere).

The Diabetes service in Newham began conducting virtual consultations in 2011 and has generated considerable experience in developing the service (and related organisational routines) in the years since, as well as working with patients and carers to evaluate and evolve that service. Following publication9–11 and presentations across the UK about the service, the team in Newham have received over fifty requests for further information (often accompanied by requests for visits and assistance) from trusts across the UK. With a focus on making the best use of the accumulated experience at Barts Health NHS Trust, and being keen to ensure the application of our findings in practice, we have secured an Economic and Social Research Council Impact Acceleration Award and a Health Foundation Scaling Up Improvement grant to help us to extend and apply the research that we have undertaken in the VOCAL study. The grant is an important outcome of the VOCAL project, as it will help to ensure that our findings have the best chance of having an impact on practice by continuing to support and evaluate the wider roll-out of virtual consulting services across Barts Health NHS Trust, adapting and extending this roll-out and evaluation to two other NHS trusts, and collaboratively creating online resources to support the introduction and sustained use of virtual media tools for virtual consulting nationally.

Building on the relationships established as part of the VOCAL study, we are also currently exploring opportunities to work with Microsoft to develop more patient- and health-care-friendly tools via Skype (and related) technology.

Suggestions for further research

There is much scope for pursuing further the basic study design and methodological approaches used in this research to add depth and contrast to the findings presented here. The policy context is changing rapidly and (at the time of writing) the UK is entering a period of unprecedented environmental turbulence in terms of the economic, technological and regulatory context for innovation in the public sector. Notwithstanding a prevailing squeeze on finances and uncertainty about regulatory structures, there is much policy interest in digital ‘solutions’ and there is likely to be money available for pilot projects. We hope that those who draw on such funding to pilot and mainstream the use of virtual consultations will use a similar methodology to that described in this report to evaluate and document the multiple interacting influences on ‘success’. As Flyvbjerg238 has noted, ‘[A] scientific discipline without a large number of thoroughly executed case studies is a discipline without systematic production of exemplars, and . . . a discipline without exemplars is an ineffective one’.

There is currently significant interest (from policy-makers, industry representatives and those working in NHS management, as well as patients and carers) in the potential roll-out of digital ‘solutions’ across the NHS. Previous studies have highlighted how digital technologies, including Skype and similar virtual media, often succeed on a small scale, but fail to achieve widespread use (scale-up) or become routine practice in other settings (spread). 136 Given the current impetus from national policy for NHS organisations to rapidly adopt such technologies (i.e. at scale and pace), combined with the level of work involved in implementing them, both spread and scale-up are potentially fruitful areas for further study. This includes understanding professional and organisational drivers for adopting (and not adopting) digital technologies, such as Skype and other virtual media, into routine patient care; supporting decision-makers in designing policies that can support spread and scale-up and NHS organisations already attempting to incorporate digital technologies; and evaluating the impact of virtual consultations on NHS organisations, as well as patients and professionals. Current research in this field tends to undertheorise the process of scale-up and spread. A potentially fruitful theoretical approach would be to consider the adoption and use of technologies as social practices. 136

Our study has focused exclusively on Skype as the virtual media platform supporting virtual consultations. There is significant scope for exploring the use of other similar media (e.g. FaceTime) both in the public and private sectors, and the interaction with other digital technologies (e.g. virtual waiting rooms). The VOCAL study has focused largely on mediated consultations using a computer, with some sessions being reliant on other devices (mobile phones and tablets). This raises the question of whether or not there are any differences in interaction and perceived success as a function of the device (or combination of devices) being used.

Only a handful of previous studies (and none from the UK) have attempted to apply the RIAS (or a similar) analytic approach to the study of remote video consultations. As noted in Chapter 5, our sample size was too small to draw confident conclusions about the differences in the kinds of talk that occurred in the remote environment compared with the face-to-face environment. We did demonstrate that differences between clinical specialties were more significant than differences between remote and face-to-face consultations within a particular specialty and that many, if not all, differences in the latter were readily explained by practical and material factors. Much additional research could be done, however, to extend the empirical data set on talk in virtual consultations, generate further hypothesis-driven questions that could be addressed using the RIAS or a similar approach focused on categorising different kinds of talk across different genres of consultation and inform the development of guidance on any suboptimal communicative practices.

The micro-analysis of consultations could also take a more qualitative approach from the perspective of sociolinguistics (e.g. using conversation analysis or discourse analysis). For instance, as noted in Chapter 5, the higher proportion of pauses and mutual silence during face-to-face consultations is one area that requires further investigation. A detailed description of, and rationale for, such an approach is beyond the scope of this report (see The Handbook of Conversation Analysis239 for an introduction), but, briefly, this technique sees communication as a dynamic interaction that emerges moment by moment and allows a deeper and perhaps more sociologically critical analysis of the power dynamics of a consultation than is made possible through the RIAS approach. Such an approach would allow for a finer-grained analysis of spoken interaction that could yield original and important findings about the communication strategies that make up a ‘good’ remote consultation. We have applied for further funding to undertake a detailed interactional analysis of our data set and would be keen to collaborate with other researchers to pool empirical data and develop a common analytic approach.

Finally, our findings suggest that there is much further research to be done that looks at the interactions between the macro, meso and micro levels in the regulation and governance of virtual consultations. For example, our own team is currently undertaking a theory-driven, multilevel analysis of the tensions between risk, governance and patient-centred care, focusing on how institutional regulations impact everyday work practice and interaction with patients, using a theoretical model developed previously by Hillman et al. 240 There is much scope to apply such a framework to different kinds of virtual and technology-supported consultation.

Contributions of authors

Sara Shaw (Senior Researcher, Sociology/Policy) was involved in the project design and led on the macro element of the study. She was also involved in the wider data collection, analysis, dissemination and publication. She led on co-ordinating the final report to NIHR.

Joseph Wherton (Senior Researcher, Health Psychology) was the lead researcher for the project, led on the ethics approval for the overall project and on the micro- and meso-level data collection and analysis, and was involved in the project design, analysis and write-up.

Shanti Vijayaraghavan (Consultant, Diabetes and Endocrinology) had the initial idea for studying the feasibility of virtual consulting in an outpatient setting and developed the idea into a research protocol (with TG). She led on establishing clinic processes for remote consultations in Diabetes and contributed to similar processes more widely in Barts Health NHS Trust, facilitated data collection and contributed to the dissemination, report writing and publications.

Joanne Morris (Research Manager) was the project manager for the study, and contributed to setting up virtual consultations in the Diabetes and Cancer Surgery clinics and to similar processes more widely in Barts Health NHS Trust, was involved in the project design, facilitated data collection and contributed to the dissemination, report writing and publications.

Satya Bhattacharya (Consultant in Hepatobiliary and Pancreatic Cancer Surgery) was involved in the project design and led on establishing clinic processes for virtual consultations in Pancreatic/Liver Cancer at the Royal London Hospital. He facilitated data collection and contributed to the dissemination, report writing and publications.

Philippa Hanson (Consultant, Endocrinology) was involved in the project design and setting up virtual consultations in Diabetes at Newham University Hospital and Mile End Hospital, facilitated data collection, and also contributed to the wider dissemination and publication from the study.

Desirée Campbell-Richards (Research Nurse) was involved in the project design and setting up virtual consultations in Diabetes at Newham University Hospital and Mile End Hospital, and also contributed to the PAG and the wider dissemination and publication from the study.

Seendy Ramoutar (Cancer Research Nurse) was involved in the project design and setting up virtual consultations in Pancreatic/Liver Cancer at the Royal London Hospital, facilitated data collection, and contributed to the wider dissemination and publication from the study.

Anna Collard (Freelance Researcher, Anthropology) was involved in the project design and led on the setting up and running of the PAG. She was also involved in the wider dissemination and publication.

Isabel Hodkinson (GP Principal/CCG Clinical Lead) acted as the lead contact with local CCGs, fed into the project design and data collection (particularly for the macro and meso levels of the study) and contributed to the report writing and publications.

Trisha Greenhalgh (Professor, Primary Care) was the chief investigator of the VOCAL study and (as such) its guarantor. She acted as the University of Oxford lead and was responsible for the overall project co-ordination and management, project design and supervision of the researchers and administrative staff. She was also involved in data collection, analysis and dissemination events, co-ordinating the study steering group, report writing and publications.

Publications

Greenhalgh T, Vijayaraghavan S, Wherton J, Shaw SE, Byrne E, Campbell-Richards D, et al. Protocol: the virtual online consultations – advantages and limitations (VOCAL) study. BMJ Open 2016;6:e009388.

Greenhalgh T, Shaw SE, Wherton J, Vijayaraghavan S, Morris J, Bhattacharya S, et al. Video outpatient consultations: a case study of real-world implementation at macro, meso and micro level. J Med Internet Res 2018;20:e150.

Data sharing statement

All data requests should be submitted to the corresponding author for consideration in the first instance. Data relating to establishing and monitoring virtual consultation services within Barts Health NHS Trust may be available from the trust, but cannot be distributed via the study team. Access to anonymised data drawn from consultations, interviews and observations with patients, staff and wider stakeholders may be granted following review.

Patient data

This work uses data provided by patients and collected by the NHS as part of their care and support. Using patient data is vital to improve health and care for everyone. There is huge potential to make better use of information from people’s patient records, to understand more about disease, develop new treatments, monitor safety, and plan NHS services. Patient data should be kept safe and secure, to protect everyone’s privacy, and it’s important that there are safeguards to make sure that it is stored and used responsibly. Everyone should be able to find out about how patient data are used. #datasaveslives You can find out more about the background to this citation here: https://understandingpatientdata.org.uk/data-citation.

Disclaimers

This report presents independent research funded by the National Institute for Health Research (NIHR). The views and opinions expressed by authors in this publication are those of the authors and do not necessarily reflect those of the NHS, the NIHR, NETSCC, the HS&DR programme or the Department of Health and Social Care. If there are verbatim quotations included in this publication the views and opinions expressed by the interviewees are those of the interviewees and do not necessarily reflect those of the authors, those of the NHS, the NIHR, NETSCC, the HS&DR programme or the Department of Health and Social Care.