Aim

To study assisted living technologies in their organisational, social, political and policy context, using a complex systems approach that considers interdependencies, emergence and conflict.

Objectives

1. To recruit a maximum variety sample of health and social care organisations seeking to improve services to older people with complex health and social care needs with the aid of technology.

2. To support these organisations in developing, delivering and evaluating their chosen service using an action research design, informed by ethnography and contextual analysis.

3. To generate wider empirical lessons about the introduction of technology-supported services in health and social care.

4. To build theory and method about fourth-generation research into assisted living.

Research questions

1. How can we improve the development of assisted living technologies by and for people with multimorbidity and declining health?

2. How can we better promote the customisation and use of such technologies in the home and the community by individuals, their carers and support services?

3. How can we ensure that ethical and existential concerns (What matters to people? What are homes for? How should we live? What are society's responsibilities towards its sick?) inform and shape the development, introduction, adaptation and use of assisted living technologies?

Overview of study design

The research has three linked components: micro (ethnographic studies of multimorbidity and ageing); meso (organisational and system change, including embedding of technologies) and macro (policy analysis, public debate and industry engagement). As in our previous ATHENE study, we will use lived experiences of real participants, studied using ethnographic methods in and around the home,57 as a key element of the programme. These micro (individual) case studies will be an integral part of, and used to inform, a set of evolving meso (organisational-level and system-level) case studies drawing on the principles of action research58 and experience-based co-design.59 The case studies, and an accompanying cross-case analysis and synthesis, will, in turn, both inform and be informed by a wider case study of national policy, industry strategy and prevailing public opinion on key policy issues, such as ageing, multimorbidity (including integrated care) and assisted living technologies. This study design is illustrated schematically in figure 1.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Diagram of the SCALS (Studies in Co-creating Assisted Living Solutions) programme (first three case studies shown).

A crucial feature of the SCALS research programme is that it is technology agnostic (or more accurately, agnostic with respect to the value of particular technologies in specific circumstances). Aligning with others who have undertaken critical ethnography of technologies in the home,21 ,22 ,26 ,37 ,60–66 we deliberately seek to ‘de-centre’ technology and, instead, place at the centre of our analysis the person's lived experience of illness, the clinical and social care microsystems and the wider health and social care systems within which that experience is nested. Technologies may be a crucial component of that lived experience and those (micro)systems, but we analyse them as they emerge and are used as part of the system, not as freestanding objects of study with an ‘impact’ all of their own.

The SCALS case studies will explore further a key finding from the ATHENE study of the importance of pragmatic, needs-focused adaptation (‘bricolage’), which has since been replicated by others.67 Bricolage may involve acquiring new technologies—but it often consists mainly of ‘fiddling’ with legacy technologies (things already present in the home or passed on secondhand from friends or relatives) to make them work in new ways and/or for new purposes.68 Bricolage, as applied to living with multimorbidity, is an under-researched area. Our research focus is the process of bricolage and how it might fit with the personalisation of care packages (lay and professional), rather than any particular technology that gets produced.

Theoretical/conceptual framework

Our work is grounded in phenomenological philosophy; it seeks to understand illness and ageing from the perspective of the individual.21 ,49 ,69 Importantly, the introduction and use of assistive technologies by health and social care professionals looking after patients and clients is an ethical practice, bound by professional codes of conduct (confidentiality, respect for autonomy, etc) and strongly linked to professional identity.70 ,71 To link this with the organisational context of care and with wider society, we will apply Stones’ strong structuration theory, which conceptualises individuals’ internal structures (ie, their past experience, cultural background, knowledge, beliefs, values, perceptions and so on—akin to what Bourdieu called habitus—along with their ongoing, phenomenological experience of illness and ageing) as recursively linked to (ie, both shaping and shaped by) external structures of social norms, rules, laws, standards, policies and so on.72

As noted above, Greenhalgh and Stones have refined and applied strong structuration theory to study the relationship between individual agency and social structures in relation to self-management and the use of technologies in healthcare.47 ,69 ,73 What we have not done previously is focus this at the meso level to explore how social structures (cultural, regulatory, political) are inscribed in organisational forms and processes, and in the technologies that are supplied and supported by these organisations, to both create possibilities and limit what is possible for the human agents who work for, and/or are served by, those organisations.

The starting point in our meso-level case studies will be a ‘problem to be explored’ (referred to in strong structuration theory as an explanandum),72 identified through both ethnographic observation and staff interviews. This might well be identified as a set of current outcomes or patient/client experiences, which are deemed to be unsatisfactory. The notion of retroduction (ie, asking ‘what mechanisms and influences might explain…’) produced from within critical realism is useful here. For each explanandum, data will be gathered and discussions held to elucidate, and adjudicate, between possible mechanisms or processes that have brought this about. These mechanisms, and their interactions, will be empirically tracked by gathering further data as needed.

The purpose of this theorisation is to identify the generative causes of the unsatisfactory outcomes (or, in some cases, of satisfactory outcomes). One key data source will be the explanations and reflections of ‘proximate actors’—that is, of individual patients and clients, and (perhaps even more so) of front-line health and social care professionals, since the latter's perspective is informed by their close experiences with many patients/clients over the years—as they compare what would ‘ideally’ be in place with what is actually occurring on the ground. In the case of unsatisfactory outcomes, we would also be seeking possible strategies (eg, a change in policy) that would alter, replace, or counter some of the generative causes, so as to produce more satisfactory outcomes. It will be important here to explore the perceptions held by front-line professionals as to what are obdurate but potentially malleable obstacles to positive change, and to differentiate these from constraints that are considered to be entirely intractable.

The pressures and constraints within and among organisations that affect staff behaviour are complex and conflicting. Sometimes, a specific set of external, organisational or interorganisational pressures can be uppermost in people's consciousness, and strongly influence their judgments about how to behave. At other times, in relation to other tasks, it will be other parts of the contextual field that will loom largest. It is important to identify which pressures from the contextual field prevent proximate actors from carrying out their ideal practices with respect to which particular tasks.

The detail of this meso-level exploration will be developed further as the programme unfolds, and be the subject of a theoretical paper.

Setting and context

The SCALS programme is recruiting health and social care organisations across England, selected to provide maximum variety in demographic, geographic, sociocultural and organisational factors relevant to the adoption and use of assistive technologies (table 1). Each participating organisation seeks to develop or change a service for older people living with multiple health and social care needs, with the help of technology. Some organisations recruited to date have focused on a particular technology (eg, telehealth monitoring for heart failure, Global Positioning System (GPS) monitoring for people with cognitive impairment), and seek to implement this in a target population; others seek to achieve a particular service goal (eg, reducing re-admission rates in high-risk patients), and are agnostic about which technology(ies) they will use. One has a very general goal—to become ‘digital by default’ (ie, maximise the use of technology where appropriate by investing in appropriate technologies and training and supporting staff and clients to use these). All participating organisations have signed up to an action research design in which ethnographic and other (mostly qualitative) data collected by or with our research team will feed into reflection, planning and action phases. Work in three of the five initial case studies started before the SCALS programme was funded, and has been able to continue as a result of that funding.

Programme management and governance

The SCALS programme will be based at, and sponsored by, the University of Oxford. It will include academic partners at Queen Mary University of London and the University of Warwick; and National Health Service (NHS) and social care partners in participating sites across England.

The research will be organised as a series of organisational case studies as listed in table 1 (and with up to four additional case studies added later, taking account of the trade-off between depth and breadth). Each case study will be led by a locally based researcher. Meetings between research teams will occur at least 3-monthly by teleconference, and 6-monthly face-to-face to share emerging findings and develop the cross-case analysis.

The programme will be supported by an independently chaired, intersectoral steering group, with representation from the technology industry, health and social care services, policymakers (NHS England), people with assisted living needs, lay members and external academics. We anticipate that, as in our previous studies of technology development and adoption, this group will serve as a vibrant intersectoral discussion forum and a crucial, bidirectional conduit to national policy, the boards of NHS, local authority and industry partners and a link with front-line clinical and social services teams.

Meso-level case studies

For each case study, we will use action research (the cycle of asking a question, collecting data, analysing data, initiating change and collecting more data to assess progress) to work with clinical and social care teams to address organisational aspects of addressing the service challenge, setting and meeting project goals, embedding particular technologies into particular microsystems and service models, and evaluating progress. In each case study, we will help the participating organisation(s) to apply the ARCHIE principles with a view to achieving patient centred, adaptive and co-created sociotechnical solutions (see above). We will also draw on Bate and Robert's experience-based co-design methodology (which requires the careful study of individual patient/client experience, for which ethnographic study is ideal), adapting this as needed to incorporate the introduction, embedding and adaptation of systems for supporting technology-mediated interactions between patients or clients, and health or social care professionals.59

We will combine this pragmatic and adaptive action research design with a more theory-driven approach aimed at collecting and analysing a dataset to produce a series of multilevel (micro–meso–macro) case studies, along with a cross-case analysis (see below). We have previously used action research to (simultaneously) facilitate and study the change process, and how conflict plays out and is managed in acute trusts, community trusts, and clinical commissioning groups, and have (in several different studies) successfully achieved the twin goals of action research: (1) local learning and change and (2) contribution to the wider knowledge base.74–77

Micro-level ethnographies

The literature on the lived experience of ageing and assisted living includes numerous ethnographic studies by other research teams20–22 ,26 ,37 ,60–63 ,65–67 ,78 as well as our own case series of 40 detailed ethnographies for the ATHENE study,48 ,49 many of which have been published with participants’ consent (see http://www.atheneproject.org). In the SCALS programme, we will use in-depth ethnography of approximately 60–80 individuals (5–15 per case study) to inform the action research and meso-level theoretical analysis described above. These individuals will form the main (though not necessarily the only) group to inform the co-design process occurring at meso level.

Our sampling of individuals will be purposive so as to include people with complex needs, including (for example):

• People living with dementia (almost all of whom will have other physical, mental or emotional conditions), including the carer experience;

• People requiring personal and intimate care (eg, with washing, toileting);

• People with inexorably deteriorating and/or terminal conditions;

• People living with mental health problems as well as physical or cognitive impairments;

• People with potentially stigmatising conditions such as epilepsy, HIV, or alcohol or drug dependency along with other impairments;

• People living in profound poverty, unstable housing circumstances, adverse family circumstances (eg, no relatives or friends in touch) or uncertain citizenship status (eg, asylum seekers);

• People where there may be adult safeguarding concerns.

This sampling frame is deliberately constructed as a counterpoint to the ideal type ‘smart home’ resident (depicted as a confident, mentally capable owner–occupier with non-stigmatising illness, supportive family and no social problems1 ,79).

Subject to ongoing consent or assent of the participant and their carer(s), we will conduct a series of ethnographic visits and use cultural probes (especially cameras for people to take pictures of what matters to them), home tours (in which the individual explains the different rooms in their home and how they are used) and ‘go-along interviews’ (accompanying the person on a journey of their choice) to capture their experience of multimorbidity and ageing from the perspective of both the index individual and their family and carer(s).55 ,57 ,69 ,80 A particular focus will be the way in which participants and their carers attempt to use, adapt, combine and repurpose technologies in pragmatic ways within and outside the home to achieve what matters to them, and what kinds of constraints they encounter as they do so. Field notes, interview transcripts, photographic and other data will be combined using the narrative form to produce a rich descriptive case study of each participant. Interviews with general practitioners and/or social care staff looking after the person may be used to augment their personal narrative. These individual ethnographies will be theorised using phenomenology and structuration theory, paying particular attention to the material, symbolic and sociocultural aspects of technology use (and non-use), and fed into meso-level organisational learning and co-design.

Macro-level analysis of wider context and public engagement

With two goals—gathering rich data on the wider context and maximising dissemination and application of our findings—in mind, we seek to build and maintain relationships with policymakers, industry and other key stakeholders nationally, both through our intersectoral steering group, and more widely.

As with the meso-level element of the programme, this element will be informed by the principles of action research. Thus, there will be cycles of developing relationships with national stakeholders, identifying problems and issues, feeding emerging findings back and discussing these, and exploring possible ways forward, workarounds or pragmatic compromises for policy, industry and the professions. In this way, project activities, such as steering group meetings, documents and so on, will be captured as research data and be included in our analysis, along with other data on the wider context (eg, blogs, press articles, industry documentation). Discussions with stakeholders will include both formal, audiotaped semistructured interviews and also ‘informal interviews’ undertaken in the process of setting up and steering the programme.81

We hope to identify, from the perspective of these stakeholders and with their active input to our deliberations, how to overcome policy, economic, technical and other barriers to delivering a robust, user-centred, sociotechnical microsystems approach to developing, customising, introducing and sustaining assisted living technologies for the various target groups in our case studies (some of whom, we anticipate, will be heavy users of health and social services, but hampered to a greater or lesser extent by poverty, low health literacy and multimorbidity). We do not anticipate that these deliberations will generate winning formulae or universal solutions. Indeed, we position our work against the possibility of such solutions.

Rather, we hope that multistakeholder input, both local and national, will produce a kind of pragmatic ‘muddling through’ that generates effective local solutions along with transferable insights about how the co-design process for technology-supported, person-focused health and social care services might be planned and operationalised. Local solutions, like the ARCHIE principles, will always require additional and ongoing work to adapt and refine them to particular organisations, individuals and settings.

One element of this macro-level component will be working with industry (as we have already begun to do) to help shift technology development and business models from ‘cathedral’ to ‘bazaar’ by encouraging the adoption of open standards to provide the necessary technical underpinnings for pragmatic, user-centred bricolage of a range of adaptable and interoperable devices by the potential user and his or her carers. A key enabler will be for industry to put in place the capacity to track the evolving relationship between devices and users’ needs and feed this back into design and development processes.

A second element, chiefly via their representation on our steering group, will be linking with national policymakers to consider how far our emerging findings make sense to those charged with developing and implementing policy on assisted living technologies in the UK—and what additional perspectives these policymakers bring that have not been (fully) addressed by local actors.

A third element will be working with policymakers and others (eg, professional and regulatory bodies) who set strategy and develop and implement standards and guidance for the use of assistive technologies. These will include NHS England, the National Information Board, Information Governance Alliance, Royal Colleges and defence societies. We will build our sample iteratively depending on how issues and policies emerge over the life of the programme. Among other things, we will want to probe whether the standard forms of evaluation and monitoring employed by regulatory agencies would be able to accommodate the pragmatic, ‘muddling through’ solutions we advocate.

A final and central element of this programme will be public engagement. We are grateful to the Wellcome Trust for a Public Engagement Award linked to the SCALS programme. In engaging citizens, we seek to do more than ‘disseminate findings’ to passive audiences. Rather, our goal is to use civic engagement—with third-sector organisations, schools, the press and other public fora—to help us address complex and contested moral questions about the role of technology in an ageing society—especially as it relates to the key policy questions illustrated by our case examples (column 5 in table 1), namely ‘tagging’, ‘integrated care’, ‘ageing in place’, ‘care closer to home’ and so on. These and other policy themes will guide our macro-level data collection and our public engagement efforts.

Our engagement with industry, policymakers and the lay public will also be used as an important source of data for the macro-level analysis. We will, for example, record, transcribe and analyse public debates on themes relating to ageing and new technologies. Press articles (both ‘lay and ‘technical’), workshop outputs (eg, flip chart paper), blogs, social media threads, online responses to our publications and so on will represent a rich corpus of material for analysing contemporary discourse on the key policy themes of interest to us. Using strong structuration theory as an analytic lens, we will nest our organisational case studies in this wider analysis of public, political and industry discourse.

Integrating the dataset

This programme of work will generate a large amount of complex, multimodal data. It will be mostly qualitative (interviews, group discussions, observations, documents) but will also include aggregated quantitative data on the performance of local services (eg, access and uptake rates, biometric data, hospital admission rates), and technical and operational data (eg, technical design features, affordances, standard operating procedures). Furthermore, the nature of the data will probably change longitudinally over time as we learn what sources are most useful to our analysis. Managing and integrating these data will be a significant challenge but we anticipate that the diversity and richness of the data will ultimately allow us to produce a sophisticated higher order analysis and new theorisations.

As noted above, multimodal micro-level (ethnographic, interview, biometric, technical) data on individual participants will be drawn together into narrative summaries, and these summaries will be fed into the action research process in participating organisations. At the meso level, we will apply the principles of interpretive case study to produce narrative accounts of organisational development over time, focusing especially on how the organisation and its industry partners sought, and took account of, the user experience in evolving the design of the service and linked technologies, and how competing perspectives and conflicts of interest were surfaced and managed. These organisational case studies will be nested in the macro-level analysis of industry, policy, professional and citizen perspectives.

In developing the organisational case studies, we will initially draw on the work of Stake, who emphasised the importance of understanding the case for its own sake (‘what is going on here?’) rather than as an example of a particular theoretical phenomenon (What is this a case of?).82 ,83 To understand the case for its own sake, richness—that is, granular depiction of real examples—is needed. As Weick has emphasised, richness has a number of generative properties including thick description, reflexive theorising and ‘conceptual slack’—openness to the many new explanations that emerge when contextual detail is added to the account.84 And as Stake has emphasised, it is important to tell the story ‘warts and all’:We need to portray complexity. We need to convey holistic impression, the mood, even the mystery of the experience. The program staff or people in the community may be ‘uncertain’. The audiences should feel that uncertainty. More ambiguity rather than less may be needed in our reports. Oversimplification obfuscates. 85

This ‘warts and all’ perspective is important to tease out both the conflicts and the emerging synergies between different stakeholder organisations and different sectors.

The final stage in the analysis will be to synthesise findings across the sample of organisational case studies, and place these in the context of our macroanalysis of wider society. At this stage in the analysis, we will be looking for commonalities and contrasts across cases that will allow us to (1) consider the national ‘case’ of assisted living support in England and (2) at a more abstract level, make transferable statements about how organisations and networks can develop effective assisted living support. These statements will be such as to leave space for the variety of ‘bricolage’ solutions required to respond to the in situ specificities of particular organisations, individuals and settings. For the first of these tasks, we will still be asking ‘What is going on here?’ and producing a unique ‘n of 1’ case study account. But for the second, we will make more explicit use of strong structuration theory to consider how the macro context shapes, constrains and provides possibilities for the activity of organisations and networks (and, within those, individuals).

Ethics

The study of people with multimorbidity in their own homes, and the use of action research to support the introduction of technologies and services raises important and complex questions of research ethics, including consent, confidentiality and the nature of participation. Given that the study is funded through the Wellcome Trust's ‘Society and Ethics’ programme, all our research questions have an ethical dimension. Our theoretical perspective and analytic approach draws on the philosophy and theory of ethics (covering themes such as the ‘everyday ethics’ of what matters to people,86 the balance between autonomy and safety in caring for vulnerable older people,87 and the implications of a ‘panotpicon’ approach to digital surveillance).71

Dissemination and projected outputs

An important feature of action research is that outputs and impacts are coproduced throughout the study.58 Rather than seeing the ‘outputs’ as a series of papers describing findings and the ‘impacts’ as how these findings are subsequently put into practice, action research is itself that putting-into-practice—and any publications are likely to be retrospective accounts of what changed during the study and how impact unfolded. In coproduced research, impacts are as much about (networked) relationships, ‘productive interactions’ and contributions to the process as they are about end outputs.88 ,89 Two previous systematic reviews, one of community–campus partnerships and one of large-system change, identified partnership synergy and distributed leadership as key mechanisms through which positive change is generated in multistakeholder collaborations.90 ,91

Hence, an important feature of the SCALS programme (as in our previous ATHENE work) will be the relationships that are built within and between the different partner organisations, both in the local study sites and nationally on the steering group, and the level of collective engagement by the multistakeholder collaboration in the unfolding projects.

Just as the ATHENE project produced detailed case studies of the individual experience of assisted living technologies, SCALS will produce richly described case studies of the organisational experience of trying to incorporate such technologies into services. We will work with our partner organisations to address how and to what extent these case studies will need to be adapted (eg, certain aspects redacted or fictionalised) for public consumption.

In principle, however, we seek to place detailed, granular descriptions of organisations’ efforts to deliver assisted living in the public domain. As Flybjerg 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’.92 We believe the interdisciplinary field of assisted living will benefit greatly from our case studies, and furthermore, that our methodology may be taken up and applied by other intersectoral programmes to add more ‘thick descriptions’ to the currently sparse literature. Additional public outputs, supported by a Public Engagement Grant from the Wellcome Trust, will include promotion of debate, interaction with the media, and engagement with schools and other audiences.