Description of data set and overview of findings

Our final data set consisted of 78 texts (68 documents and 10 sets of field notes). Documents comprised 22 academic papers (11 empirical and/or theoretical, 11 reviews or commentaries), three guidelines or standards, three local protocols, five European Union and 12 UK policy documents, six lay and four business media articles, five reports from independent consultancies or think tanks, three conference brochures, three industry brochures and two patient organisation ‘factsheets’. Field notes covered 10 conferences and other knowledge-sharing events. Of our 78 texts, 58 were from UK, 10 from elsewhere in Europe, nine from North America and one from Australia.

Analysis identified four overarching discourses (summarised in table 1): modernist, humanist, political economy and change management. Modernist discourse, often written in the future tense, depicted high-technology innovations designed by experts as offering reliable, cost-effective and ethically benign solutions to complex clinical and social problems. Humanist discourse focused on individuals and depicted present-day technologies (chiefly alarms and monitoring devices) as potentially disruptive (stigmatising and unwelcome in the home), distantiating (separating people from their families and carers) and disempowering (surveillance as control), and as offering at best a partial solution to the age-related decline of the human body and mind. Political economy discourse depicted telehealth/telecare as a lucrative business market being manipulated by commercial vested interests. Change management discourse depicted adoption, spread and sustainability of these technologies as dependent on good project management and organisational processes. We describe these contrasting discourses in more detail below.

Modernist discourse: technology as solution, technology user as consumer

The problem for which telehealth and telecare technologies were considered to be the solution was expressed thus by one academic review:The over burdened health care system will face a worldwide wave of retirees who will live longer, cost more to treat, and demand new goods and services to help them stay healthy, active, and independent. (page 171)23

This extract illustrates several features of modernist discourse. The population of older adults was depicted in vague, collective and homogeneous terms (‘retirees’, ‘elders’, ‘baby boomers’); expanding in an uncontrolled and threatening way (‘wave’, ‘tsunami’, ‘time-bomb’); and as resource-hungry, demanding and chronically sick (hence in need of preventive input, monitoring and treatment). Their longevity was depicted as a problem rather than a positive marker of health or the success of the healthcare system. The healthcare system, correspondingly, was depicted as operating at near-maximum capacity and in danger of collapse.

Telehealth and telecare technologies were presented as the rational and cost-effective solution to these problems, as tidy, precise, ‘in control’, (near-) invisible, freestanding, mobile, portable, autonomous and ethically benign. Imaginary future scenarios of the ‘smart home’ (spacious, clean, connected, safe, watched-over and free of technical or emotional trouble) and digitally empowered older citizen (literate, skilled, technically and socially engaged and with high quality of life despite having illnesses in need of monitoring) were common (see, eg, the Smith family in a ‘roadmap’ prepared for the European Commission24).

‘Empowerment’ was equated in this discourse with compliant use of telehealth technologies for self-monitoring and education. Through ‘persuasive technologies’ (defined as ‘a computing system, device, or application intentionally designed to change a person's attitude or behaviour in a predetermined way’23), medical intervention would be possible at an increasingly earlier stage in the chronic disease process, allowing the person (assumed to be a rational actor) the opportunity to ‘choose’ a healthier lifestyle.

Technological and human infrastructures were depicted, like the technologies themselves, as ubiquitous, error-free, ‘automatic’ and always available:… electronic sensors or equipment that monitors vital health signs remotely, e.g. in your own home or while on the move. These readings are automatically transmitted to an appropriately trained person who can monitor the health vital signs and make decisions about potential interventions in real time, without the patient needing to attend a clinic (page 4)Headline findings, Whole Systems Demonstrator Trial25

Modernist discourse sometimes acknowledged material limitations of current technologies, including periodic breakdown, ‘clunky’ functionality and interoperability and governance challenges, but depicted a time in the near future when such limitations would be overcome. Some texts offered future scenarios of automated ‘placeless’ data capture occurring without disruption of people's lives or lifestyles.

Modernist discourse anticipated significant reductions in costs as technologies generated various (unspecified) efficiency savings, typically using an eclectic selection of references and conditional language (eg, ‘(a maximally-saturated assistive technology market) may yield up to around $75 billion annually of savings, or about 5% of the projected $1.7 trillion spent on United States healthcare’26).

Telehealth and telecare technologies were depicted as a business opportunity, fuelled by a lucrative ‘silver pound/dollar’. Moral worth was located in the technological and business sphere, aligned with innovation, financial and technical efficiency and clinical effectiveness. Modernist discourse tended to equate ‘ageing well’ with becoming adept in the use of technologies, while depicting human contact between professional carers and vulnerable older people as unnecessary, inefficient and wasteful of resources (technologies were seen as ‘saving’ resources by reducing contact time). Technology development was framed as a design-as-engineering challenge, undertaken by specialists in a setting removed from front-line healthcare or the lived experience of illness.

In modernist discourse, agency was sited in the technology itself (eg, ‘The devices gather health information’). Technology users were depicted as consumers of technological products, passively accepting of whatever technologies were provided, skilled (and/or willing to develop skills) in how to use them, gaining enhanced health status and social experience (telepresence) from them and having only positive emotions towards them:Ms EW, 64, a retired technologist who keeps a laptop on her dining room table near a crocheted doily and a scented candle, often uses the [video chat] service to drink her morning coffee with one of several friends who live in the same town.New York Times feature article27

Delays or barriers in the adoption of telehealth or telecare were explained in terms of a deficiency in the technology user, who was described as ‘resisting’ or ‘unengaged’; the solution to this was framed in terms of user education or modification to the technology. Negative emotions and interpersonal conflict were either not addressed or presented with a potential technological fix.

A number of elements of modernist discourse helped to construct telehealth and telecare technologies and their use in the home as legitimate (reasonable, ethical and examples of good professional practice, good citizenship and so on). These included medicalisation (ageing is a disease or predisease to be ‘monitored’ and ‘treated’), rationalisation (technologies represent progress towards a more ordered and rational society), risk (the autonomous, unmonitored older person is ‘at risk’ and ‘unsafe’), deficiency (older people lack something that technology can replace), rational choice (people can be persuaded or ‘nudged’ to use technology for their own good) and empowerment (technology is liberating and provides comfort, quality, convenience and value for money).

A significant component of modernist discourse was mobilisation. For example, the Technology Strategy Board and the European Union both talked of a ‘4-point plan for mobilisation’, which comprised: (1) raise awareness, (2) create the ‘right’ conditions (eg, remove barriers to free trade), (3) accelerate investment in and take-up of proven solutions and (4) prepare for the future through further research. Technology development was depicted as necessarily progressive, occurring in phases (‘prototype’, ‘beta’, ‘launch’, ‘1st/2nd/3rd generation’), each phase producing a more sophisticated, more fit-for-purpose and more efficient (but no more expensive) product.28 This progression was depicted as occurring at speed and accelerating; stakeholders were exhorted to board the train before it left the station or risk getting left behind.

Modernist discourse was particularly apparent at knowledge-sharing events, at which it was enacted as well as written. The focus in most of these events was on showcasing advanced technological ‘solutions’ (eg, robotic products) that were not yet on the market but were presented as ready to be deployed, typically accompanied by speech acts such as the award of a prize for the most promising innovation. The argumentative work of modernist discourse was achieved at these events partly by the mutually reinforcing activities of stakeholders from different sectors as industry representatives demonstrated their products, academics offered research findings, policymakers launched new documents or themes and journalists asked questions.

Since the problem that telehealth and telecare technologies were intended to solve was considered to be extensive, urgent and rapidly worsening, the solution was depicted, correspondingly, in terms of partnerships between multiple institutional stakeholders (see Introduction section), supported by frequent knowledge-sharing events and ‘digital economy hubs’. The preferred research design was the large-scale experiment (to show ‘what works’), as exemplified by the £31 million government-funded randomised Whole System Demonstrator trial.25 National policies and business strategies in the UK were nested in a wider European vision and infrastructure for knowledge and economic growth.28

Substantial upfront allocations of resources to build inter-sectoral links and facilitate research and development were justified via a discourse of investment, payback and economies of scale, but potential conflicts of interest between these stakeholders went largely unexamined. The social change predicted by modernist discourse was dramatic, far reaching and technology driven: new technologies would not merely enhance current approaches to supporting people in their homes; they would fundamentally alter them.

Humanist discourse: technology user as active moral agent

Some texts in our data set, notably academic papers by social scientists29–33 and patient organisation advice documents34 ,35 revealed a very different discourse which contrasted with modernist discourse in five key ways. First, it centred not on technologies but on the lived experience of illness, ageing and disability; the spaces and places where people lived; the care relationships and accountabilities (formal or informal) which supported them and the positive human qualities of technology users. In particular, older people, whether physically disabled or cognitively impaired, were presented as active, resourceful and moral agents who struggled to live meaningful lives, fulfil social obligations (such as not troubling their professional carers or relatives too much) and look after the technologies given or lent to them. They were presented as the subjects rather than the objects of technology design and to that end, research studies typically had participatory and qualitative designs and studied a small sample of participants in depth.

Second, humanist discourse focused not on an imagined high-technology future but on current low-technology reality—especially on the (fairly limited) range of alarms and monitoring devices available at the time of our study. While technologies were not necessarily depicted negatively (eg, one study showed how people with dementia used mobile phones when out walking,29 and voluntary sector advice often emphasised benefits of technologies34 ,35), these were seen as offering, at best, partial solutions to the complex challenges of chronic illness, ageing, rejection and various forms of exclusion and isolation. Health and social care staff and users represented in research studies in our sample, typically saw assistive technologies as prone to failure, with limited interoperability, not always fit for purpose, creating as well as solving problems, distracting carers from the human needs of the technology user and becoming the ‘patient’ as users or carers took on responsibility for maintaining and repairing them.

Third, humanist discourse emphasised symbolic meaning. Qualitative studies that explored participants' reluctance to use alarms and tags found that some people viewed them as intrusive, disempowering, stigmatising, isolating and exposing them to surveillance by relatives and/or the state.29 Metaphors included medicalisation of the lifeworld and technology as intruder. These texts emphasised that different parts of the home had different social and cultural meaning. The living room, for instance, typically served as a site of social interaction, leisure and public display (of flowers, photographs, trophies and so on); the bedroom was a site of privacy and intimacy. Technologies placed in these spaces acquired altered significance and in turn altered the significance of the spaces. The behaviour of participants who refused to install, turn on or use telehealth or telecare technologies could sometimes be understood in terms of protecting the boundary between the home and the clinic.30

Fourth, humanist discourse reframed technology design as a personal and interactive activity, in which the intended user was an equal partner and expert in their own needs and preferences. One group of researchers, for example, described a person with early dementia who still enjoyed jogging but sometimes could not find his way home; designers worked with ‘The Runner’ to understand and incorporate his values and priorities; the result was a lightweight satellite positioning device in a Velcro-fastened striped armband that conveyed a ‘sporty’ rather than ‘tagged’ identity.31

Finally, humanist discourse emphasised the delicate and situated balance between benefit (eg, protection) and harm (eg, stigma, intrusion, distantiation). Inscribed in technology, deliberately or inadvertently, are the values, norms, world-views and assumptions of its designers and sponsors, which may include age, gender, class, cultural and other stereotypes. It is a strange form of ‘empowerment’, for example, that requires an individual to stay at home, undertake bodily rituals several times a day and enter data into a machine for distant processing. Video technology could increase social isolation by enabling professional carers and relatives to monitor them at a distance, avoiding getting close to human mess (smells, emotional needs, suffering, interpersonal conflict, body fluids). As the Royal College of Nursing put it, ‘Healthcare and nursing practice has traditionally demanded face-to-face contact between patients and professionals. The ethics of remote assessments and treatment have yet to be fully explored’.36

Political economy discourse: the technological–industrial complex

Political economy discourse was articulated mainly though not exclusively by critical academics. It sought to identify and challenge the techno-economic regimes created and perpetuated by vested interests around assisted living technologies.37 Matheson, for example, draws explicit parallels between the pharmaceutical and medical device industries.38 He argues that scientific and technological innovation is far from a politically neutral exercise in building a knowledge base. Rather, research and marketing activities combine to produce ‘product canons’ (master narratives depicting a large growing need and a drug or technology that will meet it) that integrate scientific truth claims and commercial positioning, thereby generating knowledge with implicit commercial functionality.

Texts in political economy discourse challenged the ‘efficiency narrative’ (ie, the assumption that technology, however expensive, will increase efficiency and hence bring return on investment and progressive cost-savings) put forward by techno-economic regimes. They predicted that the introduction of telehealth and telecare technologies would be a conflict-ridden process occurring at a slow pace since negotiations would be heavily constrained competing vested interests and structures such as organisational boundaries, budgets, professional accountabilities and legal jurisdictions:The current climate of strong desire to control healthcare costs, to introduce fundamental reform of NHS commissioning and to support technological solutions together with powerful commercial interests from equipment suppliers and uncertainties about efficacy could well result in expensive and inappropriate implementations being initiated. […] It would be extremely unfortunate if investment in telemedicine is made and schemes developed at the expense of more low-tech interventions that are proven to be effective (in particular, Pulmonary Rehabilitation programmes). (page 18)39

A key argument presented within political economy discourse was that the symbolic meanings we give technologies are shaped and constrained by institutional forces and by our awareness (or lack of awareness) of alternatives. Policymakers seduced by the rationalist appeal of technology may conflate the management of illness (ie, everything a person needs to do in order to live well with a condition) with the much narrower set of tasks linked to the management of disease (collection of a more limited data set of biometric data such as blood pressure, weight or peak expiratory flow rate). Few intended users achieve the insight and confidence to recognise and challenge this framing.

Finally, political economy discourse proposed that the techno-economic complex linked to telehealth and telecare technologies reflects a shift away from traditional social values in which healthcare is a public good, and society's duty to the vulnerable is defined in terms of human contact and the logic of care, towards neoliberal values in which healthcare is a commodity and society's duty to the vulnerable is defined in terms of monitoring and instrumental rationality (decision-making). Poverty, poor health and social exclusion have complex determinants which are little if anything to do with the so-called ‘digital divide’ and which do not have a simple technological or behavioural fix. Far from being a benign transformative force that will produce a healthier, more equal and more cohesive society once the information-poor have been given training and behaviour modification advice, ‘assistive’ technologies are an invidious form of social control, potentially increasing inequalities by transferring responsibility and workload from the system onto the patient and family. Political economy texts ask, on behalf of the intended user, ‘Whose interests does this technology serve?’.33

Change management discourse

Some texts depicted the introduction of telehealth and telecare technologies in terms of change management.40–44 For example, a monograph by the Kings Fund, an independent think tank which was a partner in the Whole Systems Demonstrator trial, hints at forthcoming trouble as technologies that have been shown to ‘work’ in generously funded randomised controlled trials encounter real-world issues such as staff motivation and skill base; professional roles and relationships; organisational routines and inter-organisational care pathways and the need for a substantial recurrent budget for purchasing and maintaining technologies.40

Such notes of caution have tended to come from senior individuals who were actively involved in implementing telehealth and/or telecare technologies and who had initially positioned themselves within the dominant modernist discourse, but who subsequently sought to explain low uptake, patient or staff resistance or unexpected consequences (eg, hidden costs). This typically required a re-engagement with the ‘mess’, which technologies had originally been expected to resolve, as this academic paper illustrates:the failure of e-health systems to become routinely used within the health services is not necessarily a consequence of design or technological flaws. […] Many factors could be at play: disruption to services, a mismatch between functionalities of the system and work practices, or resistance or even rejection by end-users for reasons including: concerns regarding patient/practitioner relationship, the need to redefine structures, roles and hierarchies, perceived threat to status within the organisation, fear of a lesser role, becoming redundant, etc. (page 301)45

Commentators in this discourse continued to view telehealth and telecare technologies as potentially beneficial, transformative and cost saving. But in an explicit departure from the technological determinism of modernist discourse, they sought to emphasise that transformation would require a great deal of additional work and to reframe technology as opportunity for rather than driver of organisational change.

Two research studies in our sample used May's normalisation process theory to frame a series of questions to guide data collection and analysis in a telehealth implementation project: (1) ‘what is the (new) work that has to be done?’; (2) ‘who does the work?’; (3) ‘how does the work get done?’ and (4) ‘how is the work understood (accounted for)?.45 ,46 Most studies in this literature, however, lacked an explicit theory of change.

While many texts used the term ‘complexity’, organic metaphors that reflected a complex adaptive system perspective (‘co-design’, ‘evolution’, ‘dynamic system’, ‘interdependence’) were rare. More commonly, the term was used to depict complicatedness, that is, multiple stakeholders and components for which the key challenge was essentially operational: planning, project management, clarity of roles and responsibilities and so on. One management consultancy, in a monograph entitled ‘Healthcare without walls’, for example, offered a step-by-step approach to developing telehealth services. In one example from an NHS Trust (page 33), they suggest step 1: Identify needs; step 2: Establish buy-in; step 3: Consider which technologies are needed; step 4: consider workforce requirements such as training and change management; step 5: Design evaluation; step 6: plan implementation; step 7: write business case and step 8: Share best practice.41

This example illustrates a key difference between political economy and change management discourses. Whereas the former viewed inter-organisational and inter-sectoral conflict as inherent and linked to historical and social forces, the latter anticipated that at some stage in the future (if the correct steps were followed, relationships built, groundwork done, knock-on effects mitigated and so on), they would be resolved. Indeed, change management discourse engaged only superficially with the issue of conflict between stakeholders—presenting it either as a straightforward project management challenge or as a ‘difficulty’ that was raised but not analysed further.

How discourses mapped to stakeholders and interacted

While the different stakeholder groups were not entirely homogeneous, it was broadly the case that the four discourses described above reflected the position of one or more key stakeholders. Modernist discourse largely reflected that of policymakers, the technology industry and researchers from biomedical and health informatics disciplines. In the UK, these three groups frequently came together in government-funded knowledge-sharing events and ‘digital economy hubs’. Humanist discourse tended to reflect the perspective of social scientists and related academic disciplines (eg, nursing); these stakeholders were represented at the large knowledge-sharing events but their work was less prominent and less highly valued (eg, research in the humanist tradition did not feature in keynote presentations or win prizes for ‘best innovation’). Political economy discourse was aligned with the position of critical academics and sceptical clinicians and again was a minority voice except at ‘fringe’ meetings. Change management discourse was articulated mainly by those at the front line of implementation (in particular, management consultancies and think tanks).

The four discourses were readily discernible and distinguishable from one another. However, few individual texts restricted themselves to a single discourse. Rather, most texts reflected one (most commonly, modernist) discourse but, to a greater or lesser extent, acknowledged elements of the other three. With few exceptions, however, different discourses tended to ‘talk past one another’ in documents (eg, with the minor discourse appearing only in a footnote or a separate subsection) rather than informing and challenging one another in effective interdisciplinary dialogue. Similarly, conferences held to showcase high-technology research often included one or two sessions on ‘the user experience’, ‘change management’ or ‘cost-effectiveness’, but the main presentations were futuristic, technology focused and strikingly divorced from considerations of real-world implementation. In the academic world, modernist discourse dominated health informatics and computer science journals and conferences, while humanist and political economy discourses were sited in the social sciences, with little cross-fertilisation between these.