The ARCHIE framework for effective telehealth and telecare services

15 Pages·2016·0.55 MB·English

by  Trisha Greenhalgh

Greenhalghetal.BMCMedicine
DOI10.1186/s12916-015-0279-6
RESEARCH ARTICLE Open Access
What is quality in assisted living technology? The
ARCHIE framework for effective telehealth and
telecare services
Trisha Greenhalgh1*, Rob Procter2, Joe Wherton3, Paul Sugarhood4, Sue Hinder2 and Mark Rouncefield5
Abstract
Background: We sought to define quality in telehealth and telecare withthe aim of improving the proportion of
patients who receive appropriate, acceptableand workable technologies and servicesto support them living with
illness or disability.
Methods: Thiswasathree-phasestudy:(1)interviewswithseventechnologysuppliersand14serviceproviders,
(2)ethnographiccasestudiesof40people,60to98yearsold,withmulti-morbidityandassistedlivingneedsand(3)
10co-designworkshops.Inphase1,weexploredbarrierstouptakeoftelehealthandtelecare.Inphase2,weused
ethnographicmethodstobuildadetailedpictureofparticipants’lives,illnessexperiencesandtechnologyuse.In
phase3,webroughtusersandtheircarerstogetherwithsuppliersandproviderstoderivequalityprinciplesforassistive
technologyproductsandservices.
Results:Interviewsidentifiedpractical,materialandorganisationalbarrierstosmoothintroductionandcontinued
supportofassistivetechnologies.Theexperienceofmulti-morbiditywascharacterisedbymultiple,mutuallyreinforcing
andinexorablyworseningimpairments,producingdiverseanduniquecarechallenges.Participantsandtheircarers
managedthesepragmatically,obtainingtechnologiesandadaptingthehome.Installedtechnologieswererarelyfitfor
purpose.Supportservicesfortechnologiesmadehigh(andsometimesoppressive)demandsonusers.Sixprinciples
emergedfromtheworkshops.Qualitytelehealthortelecareis1)ANCHOREDinasharedunderstandingofwhat
matterstotheuser;2)REALISTICaboutthenaturalhistoryofillness;3)CO-CREATIVE,evolvingandadaptingsolutions
withusers;4)HUMAN,supportedthroughinterpersonalrelationshipsandsocialnetworks;5)INTEGRATED,through
attentiontomutualawarenessandknowledgesharing;6)EVALUATEDtodrivesystemlearning.
Conclusions:Technologicaladvancesareimportant,butmustbeunderpinnedbyindustryandserviceproviders
followingauser-centredapproachtodesignanddelivery.FortheARCHIEprinciplestoberealised,thesectorrequires:
(1)ashiftinfocusfromproduct(‘assistivetechnologies’)toperformance(‘supportingtechnologies-in-use’);(2)ashiftin
thecommissioningmodelfromstandardisedtopersonalisedhomecarecontracts;and(3)ashiftinthedesignmodel
from‘walledgarden’,brandedproductstointer-operablecomponentsthatcanbecombinedandusedflexiblyacross
devicesandplatforms.
Pleaseseerelatedarticle:http://dx.doi.org/10.1186/s12916-015-0305-8.
Keywords:Telehealth,Telecare,Multi-morbidity,Quality,Co-design,Ethnography
*Correspondence:[email protected]
1DepartmentofPrimaryCareHealthSciences,UniversityofOxford,2ndfloor,
NewRadcliffeHouse,WaltonSt,OxfordOX26GG,UK
Fulllistofauthorinformationisavailableattheendofthearticle
©2015Greenhalghetal.;licenseeBioMedCentral.ThisisanOpenAccessarticledistributedunderthetermsoftheCreative
CommonsAttributionLicense(http://creativecommons.org/licenses/by/4.0),whichpermitsunrestricteduse,distribution,and
reproductioninanymedium,providedtheoriginalworkisproperlycredited.TheCreativeCommonsPublicDomain
Dedicationwaiver(http://creativecommons.org/publicdomain/zero/1.0/)appliestothedatamadeavailableinthisarticle,
unlessotherwisestated.
Greenhalghetal.BMCMedicine Page2of15
Background called the ‘critical’ literature on assistive technologies,
Assistedlivingtechnologies summarisedbriefly below.
Assistedliving(or‘assistive’)technologiesincludetelehealth
(remote monitoring for clinical biomarkers) and telecare Themythofthesmarthome
(for example, alarms, sensors, reminders), designed to de- In Designing a Digital Future, Dourish and Bell expose
liver health and social care services to the home [1]. The themyth(now25yearsold)ofthe‘smarthome’[11].They
arguments for developing them are well rehearsed [2-5], highlight designers’ preoccupation with the proximate
thoughnotunchallenged[3,6-8].Associetyages(sothear- future,animminenteraof ‘calm’–ever-present,invisible,
gument goes), more and more people have chronic condi- reliable – technology, when the mess and hassle (for ex-
tions; assistive technologies, alongside self-management by ample, glitches, bugs, interoperability, intellectual prop-
patientandcarer,willhelpmonitor,treatandevenprevent erty, information governance, set-up costs and so on)
such conditions, thereby improving quality and length of associated with technologies have been resolved, leading
life while also relieving pressure on increasingly stretched to ‘a future saturated with technology’ (page 22). This
healthandsocialcareservices. utopianfuture,theywarn,willnevermaterialiseandneeds
Drivenpartlybyconcernsaboutcostsoflong-termcare, tobepurgedfromourdreamsandplans:
investmentinassistivetechnologiesfromindustry,govern-
ment and research sponsors is high. In the UK, for ex- ‘Liftthecover,peerbehindthepanels,orlook
ample, the Department of Health’s £31 M Whole Systems underneaththefloor,andyou willfind amaze of
Demonstratortrial(2008to2011)was,asitsnameimplies, cables,connectors,andinfrastructuralcomponents….
designed to demonstrate the effectiveness and cost effect- Pushfurther,and youwillencounterregulatory
iveness of ‘whole system’ approaches to technology intro- authoritieswho authorizeinterventions andcertify
duction and use [9] – though arguably, it generated more qualifiedindividuals,committeesthat resolve
controversies than it resolved [10]. TheTechnology Strat- conflicting demandsintheprocessof setting
egy Board (TSB) allocated £25 M to its Assisted Living standards,governmentsthat setpolicy,bureaucrats
Innovation Platform (ALIP) in 2008 to 2011, with many whoimplementit,marketerswhoshapeour viewsof
projects attracting matched industry funding. ALIP’s main theroleoftheinfrastructureinourlives,andmore.
sequel, DALLAS (Delivering Assisted Living Lifestyles at Messisalwaysnearby’(page 4)[11].
Scale),a£23MpartnershipbetweenTSB,theNationalIn-
stitute for Health Research (NIHR) and government, runs In the modernist dreams of policymakers and designers,
from2011to2015.TheEuropeanFP7(2008to2013)and a new generation of more sophisticated information and
Horizon 2020 initiatives include large, inter-sectoral pro- communication technologies (ICTs) is what is needed to
grammes to produce assistive technologies at scale and rescue people from this ‘mess’ and to bring order to their
drivetheseintoproductionandwidespreaduse[3]. affairs [3]. Yet the reality often proves otherwise, especially
We have previously demonstrated, through discourse in applications where a close fit between technology and
analysis of academic, policy and lay texts, that such pro- user requirements is essential, but the heterogeneity of the
grammes are predicated on a modernist vision of the latter–and,insomecircles,aprivilegingofquantitativere-
‘smart’ home in which ubiquitous technologies, seam- search over qualitative – limits our understanding of them
lessly integrated with health and social care information [12]. Examining programmes through a socio-technical
systems,willenable dignifiedageingbypreventing,mini- lens,itislikelythatsuccessfultechnologydevelopment,in-
mising or compensating for the effects of degenerative stallation and use will be challenged by contestation about
disease [3]. The assumption behind this vision is that standards (clinical and technical), policies (national and
assistive technologies, if optimally developed and imple- local),practicalitiesofuse(theubiquitousmodel-realitygap
mented at scale by a thriving and innovation-driven tech- in all its forms), service support (what is the role of the
nology industry, will generate social change and thereby physician, specialist nurse, call centre or data processing
(at least partly) solve the uncomfortable problem of what hub in the ongoing support of an installed device?) and
we should do with the growing ‘burden’ of ageing and commercial interests (including manufacturers’ profit
dependentcitizens–whilealsosavingmoney. motive and associated barriers to interoperability). Rather
But this vision of a technology-supported ‘better soci- thanbeing‘plugandplay’,assistivetechnologieswillalways
ety’ remains elusive, for two reasons. First, there is per- need skilled human work, inter-sectoral negotiation and a
sistent over-confidence in the capacity of technological socialinfrastructuretoensurethatthey‘work’.
innovations to configure the future. Second, there are
material and ethical questions of how chronic illness Theuniquenessofassistedlivingneeds
and suffering affect people’s capacity to live in the world. Modernist research on assistive technologies, led by
These themes have been addressed in what might be computer scientists, is remarkably thin on clinical detail.
Greenhalghetal.BMCMedicine Page3of15
Because of this, it tends to generate superficially plaus- the person’s home to map the complex healthcare, social
ible solutions that may prove unusable in practice be- care and socio-cultural needs of older people and their
cause their design fails to take account of how multiple carers, encompassing a range of ethnic and social
medicalconditions affect aperson’sability to understand groups. Phase 3 took forward exemplar cases and used
and operate a technical device – or of variation in how participatory design methods to explore how older
people may want to use the device and, indeed, what people and their families might work directly with in-
they may want to use it for [3,12-15]. Yet clinical experi- dustry designers and service providers to produce fit-
ence readily demonstrates that people’s illnesses and im- for-purpose technologies (either new or adapted) along
pairments are unique, and every individual will have with appropriate service support, that fit in with people’s
different goals and a different view of how technologies care needs and lifestyles. Previous papers from the
willbesthelp them. ATHENE project have reported the methodology [16],
Everyday ethicsisakey themeincriticalassistivetech- findings from interviews with suppliers and service pro-
nology research. As Heidegger showed, we use technolo- viders theorised using diffusion of innovations theory
gies (when we can, and to the extent that they ‘work’) to [17],andanalysesofthe ethnographic data from a socio-
do things and make things – and at a more abstract logical perspective [13] and from a computer-supported
level, to achieve what matters to us [13]. When a tech- cooperative workperspective[18].
nology interferes with what matters to people (for ex-
ample, when it makes the bedroom look and feel like a Sampleandsetting
hospitalward),they quicklyrejectit. Thestudywasundertakenacrosstwosites,inLondonand
Underpinnedbythiscriticalperspective,andwiththe Manchester, both characterised by ethnic and socio-
aim of improving the proportion of patients who re- economic diversity with a predominance of poverty and
ceive appropriate, acceptable and workable technolo- deprivation.Thesampleofseventechnologysupplierswas
gies and services to support them living with illness or drawn from a range of medium and large companies that
disability, we sought to answer the following research made and supported assistive technologies. They were re-
question: what is quality in the design, implementation cruitedvianetworkingeventsonassistedlivingand/orin-
and use of telehealth and telecare, and how might we dustry representatives on our steering group; we set no
achievesuchquality? restrictions on the particular technologies they made. We
have deliberately not given detailed information about
Methods these organisationstopreserve anonymity. The14service
TheATHENEstudy providerrepresentatives weredrawn from 10 telecareand
The ATHENE study (Assistive Technologies for Healthy telehealthsupportservices,comprisingsixlocalauthorities,
Living in Elders: Needs Assessment by Ethnography), one private, two NHS trusts and one voluntary sector.
whichranfrom2010to2013,wasfundedbytheTechnol- Demographic characteristics of the participants living with
ogy Strategy Board Assisted Living Innovation Platform multi-morbidity are summarised in Table 1 and medical
programme. It sought to produce a rich understanding of conditions(objectiveandsubjective)inTable2.
the lived experiences and needs of older people with
multi-morbidityandtoexplorehowthoseinvolvedinpro- Theoreticalposition
viding and supporting the technology – technology sup- Our critical (in the sense of ‘critique’, not ‘criticism’) per-
pliers, health and social care providers – can work with spectiveonassistivetechnologiesrejects thetechnological
carerecipientsandcarersto‘co-produce’technologiesand determinism and naïve utopianism of many studies of
service solutions. The ATHENE study had an external telehealth and telecare [3]. It is grounded in phenomeno-
steering group with an independent lay chair and repre- logical philosophy, especially Merleau-Ponty’s work on
sentatives from academia, policy, health and social care perception and Heidegger’s concept of how technology,
providers, technology suppliers and service users (includ- when‘ready-to-hand’(thatis,smoothlyalignedwithaper-
ing technology users). Ethical approval was gained from son’s bodily and mental functions), extends both sensory
Queen Mary University of London Research Ethics Com- perception (the capacity to feel, see, hear and so on) and
mittee (QMREC2011/38 1 June 2011), Harrow NHS Re- motorintentionality(thecapacitytoactpurposefullyusing
search Ethics Committee (11/LO/0737, 8 July 2011) and the body) [19,20]. Within this framing, we align with
subsequentamendments. others’ research on the sociology of the body (particularly
The study consisted of three phases. Phase 1 involved Pickard and Rogers on the lived experience of illness and
initial interviews with 21 key stakeholders from technol- ageing[21,22]);the‘moralturn’inthesocialsciences,par-
ogy suppliers (n=7) and service provider organisations ticularly Sayer’s notion of ‘what matters to people’ [23]
(n=14). Phase 2 consisted of detailed ethnographic and Mort et al.’s work on the social and ethical implica-
studies of 40 individual cases, conducted in and around tions of telecare technologies, which, in order to ‘work’,
Greenhalghetal.BMCMedicine Page4of15
Table1Summaryofparticipantsinphase2 Table2Summaryofmedicalconditionsandsubjective
Age(median,range)81(60to98)years Number impairmentsinphase2participants
Gender Objectivemedicalconditions Number
Neurologicalconditions(stroke,Parkinson’s,othertremor, 20
Male 13
severemigraine,pastpolio,notformallydiagnosed)
Female 27
Arthritis 14
Ethnicity
Highbloodpressureand/orhighcholesterol 14
WhiteBritish 24
Chronicrespiratorydisease(asthma,chronicobstructive 13
OtherEuropean 1 pulmonarydisease)
SouthAsian 4 Diabetes 11
Chinese 3 Maculardegeneration,glaucomaorcataract 11
Caribbean 5 Coronaryheartdisease 10
African 2 Depression,anxietyorpsychologicalstress 7
Housingstatus Dementia,cognitiveormemoryproblems 7
Ownhouseorflat 19 Sideeffectsfrommedication 7
Privatelyrented 1 Trauma(forexample,recentorpersistingeffectof 6
pastfracture)
Housingassociation 7
Swollenfeetwithoutformaldiagnosis 3
Localauthority 10
Cancer 2
Shelteredhousing(thatis,withresidentwarden) 3
Other(e.g.urogenital,kidneyfailure,anaemia,tendencyto 16
Livingarrangements
infections,hormonedeficiency,pepticulcer,sleep
Alone 18 apnoea,deafness)
Withpartneronly 13 Subjectiveimpairmentsaffectingbasicdaytodaytasks
Withpartnerand/orothercarer 9 Generalisedtiredness/lowenergy 23
Significantandpersistentpain 18
Stiffnessorweaknessinjointsand/ormuscles 18
must be nested in networks of accountable human rela- Shortnessofbreath 13
tionsandresponsibilities[12]. Poorornovision 11
Our work also aligns with other research in the crit- Unsteadiness,dizzinessorbalanceproblems 9
ical ethnography tradition, including what Star has
Poorcognitivecapacity,concentrationorconfidence 11
called ‘the ethnography of [technological] infrastruc-
Oneormorelimbsparalysed 7
ture’ [24,25]; ‘health and place’ geography, in which
Bulkydeviceaffectingmobility(e.g.oxygencylinder,catheter) 7
healthcare technologies and their use are considered in
the context of the physical, material and symbolic Incontinence 6
spaces of the home and community, and the networks Difficultywithfinefingermovementsand/orwriting 5
of family and social relations linked to these [26,27]; Blackouts,lossofconsciousnessorperceivedriskofthese 5
and critical nursing studies, in which the old-fashioned
Physicalbulk(obesity,severelyswollenlegs) 4
dualism ‘high-tech’ versus ‘high-touch’ is replaced with
Wandering 2
a more contemporary theorisation of how technology
can support intimate nursing care of the body [15]. Much
of this critical literature comes from the Netherlands,
Sweden and Norway, where the study of technology-in- pointfromstudiesemphasisinghowassistivetechnologies
use(performative,practice-focusedandusingethnography could be used for supporting the biomedical agenda (for
in real-world settings) has high credibility. But such ap- example, monitoringofdisease).Inthepast,designersas-
proaches currently have less of a foothold in the UK and sumed that computers would be used in homes for the
North America, where research funders have tended to same tasks as they were used for in offices (for example,
privilege development of advanced technology solutions filing, calculating). Early computers aimed at the home
andrandomisedtrialsto‘demonstrate’these[9]. market emphasised how important these tasks were (or
Aphenomenologicallensbeginswiththeintendedtech- were likely to become) in the modern home. It was only
nology user’s perceptions and desires and asks how tech- whenpeoplebegantoenvisagehomecomputersinarad-
nology could augment the former and help achieve the icallydifferentway, to‘digitallyenable’ thehome activities
latter. This approach thus has a very different starting- thatmatteredtopeoplelikeplayinggamesandlisteningto
Greenhalghetal.BMCMedicine Page5of15
music, that home computing took off at scale [11]. In the in their home and life; key material properties (of the
study reported here, we sought to apply this general technologies) and key capabilities (of the individual to
principletoassistivetechnologies.Startingfromtheprem- operate and interpret these technologies); and specific
isethatattemptstoturnthehomeintoamini-hospitalare incidentsofusing(orchoosingnot touse) technologies.
doomedtofailure,wedevelopedandrefinedethnographic
techniques to build a rich picture of how people actually Phase3:Co-designworkshops
livewithmulti-morbidity.Wefocusedinparticularonthe Tenworkshopswere conductedwith61participants.Four
experienceofillnessandsufferingandhowpeopleuse(or were held with a total of 30 end-users (case participants,
whytheychoosenottouse)particulartechnologies. their carers and third-sector advocates); three were held
Phenomenology underpins the science of experience- withatotalof18serviceproviderrepresentatives(occupa-
based design, which takes the patient’s ‘ordinary experi- tionaltherapists,nurses,monitoringoperators,technicians,
ence’ as the starting point for clinical microsystem and servicemanagers,commissioners);andtwowereheldwith
wider health system redesign [28]. As noted above, few 13 technology industry representatives (designers, engi-
technologies designed for the so-called smart home are neers, business development, marketing). The final work-
‘plug and play’; there is an emerging literature on how shopbroughttogether11representativesfromacrossthese
individuals adapt and customise them to fit with per- differentuserandstakeholdergroups.
sonal needs and capabilities and with the material con- Co-design workshops are an established participatory
straintsoftheirlocalsetting [11]. design approach to help users and stakeholders articu-
late existing practices, identify challenges and develop
Phase1:Interviews new ideas [30]. They have much affinity with Robert and
Sixteensemi-structuredinterviewswereconducted witha colleagues’ work on experience-based co-design of clin-
purposeful sample of 21 participants (7 technology de- ical services [31], but include a more explicit focus on
signers and 14 service providers) involved in the develop- the design, adaptation and use of technologies. During
ment and provision of telecare in the UK; full details are the workshops, we presented vignettes from the ethno-
given in a separate paper [17]. Questions focused on per- graphic work to communicate the lived experience of
ceived challenges to the uptake and use of telecare; the older users, promote the sharing of personal stories and
technology design process; the installation and support of elicit ideas about how the technologies and services cur-
telecare technologies; and views about future develop- rently onoffer couldbeimproved.
ments. The interview protocol was adapted as the study In the four technology user workshops (users, carers
progressedtoexploreemergingthemesinmoredetail.In- and advocates), vignettes were presented using a ‘story-
terviewswererecordedwithconsentandtranscribed. board’format, which depicted, in cartoon-strip format, a
narrative in a series of frames (see example in Figure 1).
Phase2:Ethnographicstudies The stories were fictional but based on real accounts
In the ethnographic studies (described in detail else- from the ethnography of problems encountered with as-
where [13]), we visited 40 participants at home, each on sistive technologies. Workshop participants considered
several occasions one to two weeks apart, and encour- the material features of technologies, facilitated by a
aged them to help us build a rich picture of their lives, sorting exercise of cards depicting specific devices’ de-
including their daily activities and what mattered to sign features. They also considered aspects of telehealth
them. Our techniques include cultural probes (in which and telecare service provision, facilitated using a flow
participants become co-ethnographers, using cameras, diagram of the assistive technology provision process –
diaries and scrapbooks to collect data about their lives); ‘assessment’,‘decision for telecare/telehealth’,‘installation
home tours (in which the participant takes the re- andtraining’,and‘review’.
searcher round their home, describing each room’s sig- The four service provider and two technology industry
nificance and activities that occur in it) and narrative workshops began with a presentation of an anonymised
interviews (in which a conversational format is used to case study from phase 2 along with additional data ex-
explore stories about the person’s life raised by them) tracts(stories,quotes,probematerialsandphotographs).
[16,29]. We analysed the multi-modal dataset thematic- Participants were sent the example case summary prior
ally and applied narrative as a summarising and synthe- to each workshop (with the index participant’s consent)
sizing device to produce rich individual case studies and asked to reflect on three questions: (1) Bearing in
(four to ten pages long) of each participant, presented in mind what matters to this person, how could their life
a semi-structured format that covered: the social and be improved through a technology or service?; (2) What
cultural and historical context; the participant’s experi- would be the issues/challenges in implementing one of
ence of ageing and ill health; key people (lay and profes- these solutions and how might these be overcome?; and
sional) in their life; what mattered to them; technologies (3) How might the technology or service be sustained
Greenhalghetal.BMCMedicine Page6of15
Figure1Exampleof‘cartoonstrip’approachtogeneratingdiscussionaboutcasescenarios.Inthisexample,Frame1introducesthe
characters(Senthilandhisson,Ashok).Frame2describesAshok’shealthproblemsandfrequentvisitstotheclinic.Frame3describesthe
installationofatelehealthdevicetomonitorSanthil’sbloodpressureandoxygensaturation.InFrame4,Senthilisconfusedandconcernedabout
abeepingsoundfromthedevice.InFrame5,Ashoklaterrealisesthatthedeviceisnotpluggedintothepowersocket,andthatthebeeping
indicateslowbattery.InFrame6,AshokplugsthedevicebackintothepowersocketandremindsSenthilnottoremoveit.
and adapted over time? Following discussions on the vi- technicalandsocial systems.Promptcardswereusedfor
gnettes,participantsconsideredtheimplicationsforservice participants to brainstorm about whom they would like
design, facilitated byaflow-diagram oftheservicedelivery to communicate with across the care network, and what
process, such as ‘assessment’ and ‘review’. The industry type of information would help them support the tech-
workshopcentredonimplicationsfortechnologydevelop- nology user more effectively. Again, this workshop was
ment, articulated in design terminology, such as ‘require- recordedandtranscribed.
mentsgathering’,‘prototyping’,‘fieldtrials’,‘userfeedback’.
All workshops were audio recorded and professionally Dataanalysis
transcribed. The end-user, industry and service provider We used the 21 transcribed stakeholder interviews from
workshops were analysed separately using a constant phase 1, 40 individual case studies from phase 2, and the
comparative approach [32] to summarise participants’ written summaries of the 10 co-design workshops from
perspectives on the ethnographic data and priorities for phase 3 as an intermediate dataset. JW and TG re-
technology and service improvements. Common themes analysed these texts thematically using the question ‘what
across the first nine workshops were synthesised to is quality in telehealth/telecare provision?’ as a guiding
sharpen the focus for the final cross-sector workshop, question. Each researcher independently looked through
which brought together the different user and stake- the texts and identified characteristics of ‘quality’ tele-
holder groups. An anonymised case summary was used healthor telecareandalsoexamples of(real orperceived)
toprovokediscussion onhow collaboration betweenfor- quality failures. These were shared among the wider re-
mal and informal carers could be supported through searchteamandrefinedbydiscussion.
Greenhalghetal.BMCMedicine Page7of15
Results obstructivepulmonarydisease(COPD)asaresult,he
Stakeholderinterviews thinks,of hislifelong smoking.He saysheusedtohave
Participants identified multiple interacting influences on homeoxygen lastyear fortheCOPDbutthenurse
the adoption, assimilation, implementation and sustain- told himhedidn’tneed itbecausehisoxygen levels
ability of telecare. This included attributes of the tech- wereOK.Walterdisagreeswith thenurse’sassessment.
nology (for example, relative advantage over existing He said hefeelsheneedsoxygen sometimes.Hehas
arrangements, low complexity, risk involved in adop- high bloodpressure,and alsoprostateproblemsthat
tion), characteristics of the intended user (especially have ledtourinary incontinence,forwhichhewears
their physical and cognitive capabilities); the extent and pads.Walter sayshecango[tothepub]foracouple
nature of social influence (for example, limited awareness of pints,gotothetoilet,andthen stillfindhimself
of these technologies among many health professionals, ‘leaking’.Because of hisurinaryproblems and his
who, therefore, did not mention them to their patients); breathing, Walter’ssleepisverydisturbed.
low levels of organisational innovativeness (due partly but
not entirely to squeezed budgets); low levels of organisa- WhileWaltersays,‘Idon’thaveanyproblems,memory
tional readiness for telecare technologies (due partly to a wise’,Christineexplainsthathedoeshavememory
perception that these innovations would not be cost- problems(forexample,henapsforanhourthenwakes
effective); weaknesses in the assimilation process (espe- upandthinksitisthenextmorning),andhealsohas
ciallyinadequateassessmentandtailoringtotheindividual vacantperiodsforreasonsthatareunclear.Heis
and care network); weak embedding of telecare in the waitingforanappointmentforabrainscan.Atnight,
business-as-usual of the various organisations who might Waltersometimeswandersaboutthehouseandtriesto
contributetothesupportnetwork;andpoorlinksbetween cook.Thisisnotsafe(hehasburnedtoastinthepast),
usersanddevelopersatthedesignstage[17]. soChristinehasputalockonthekitchendoor.
Findings from this phase highlighted that solutions for
assistedlivingarecomplexinnovationsrequiringinputfrom, WalterspendsmostofhistimeindoorswatchingTV,
and coordination between, people and organisations. To goingoutsideperiodicallyforacigarette.Thehouseis
promote adoption and use, the different contextual factors smallandveryclutteredwithatinyoutsideyard,and
must be specified, understood and addressed. A number of with7peopleandnumerousanimals,thereisnotmuch
important questions were raised that sat largely outside the sparespace.ThenurseshavetalkedaboutWalterhaving
domain of technology design, particularly with regard to awheelchairbutChristineisnotkeen.Thereisnoroom
how we optimise the process of assessment and personal inthecrampedhousetostoreit.Walterdoesnotfeelhe
‘tailoring’ofanoff-the-shelfdevice:howtoovercomeorgan- needsawheelchair.Whattheywouldreallylikeisforthe
isational inertia and lack of resource when introducing as- counciltoputalargegateonthebackgardenfenceso
sistive technology services; how to make the services more theycangetthecarin,makingiteasiertogetWalter
cost-effective(and,hence,moreattractivetocommissioners intothecar,especiallyoncoldandrainydays.
and purchasers); and how to optimise the long-term sup-
port for the technology user so that it is sustained as a Walterisonvariousmedications,tabletsandinhalers
‘working’ technology in the long term. Full details of the fortheCOPDandmoretabletsfortheprostateproblem.
findingsfromthisphasearepresentedelsewhere[17]. Hehasregularvisitsfromhealthprofessionalsandis
somewhatconfusedaboutthese.Hehashadtelehealth
Ethnographiccasestudies equipmentinstalledforaboutayearbuthedoesnotuse
Our ethnographic research revealed a huge diversity of in- itnowandnobodyhasbeentocollectit[…].Thedevices
dividual and family circumstances, medical conditions (see includeanoximeter,bloodpressuremonitor,
Table2),personalpriorities,physicalandcognitivecapabil- thermometerandweighingscales.Hestillhasa
ities,installedtechnologies,homeenvironments,andextent nebuliser,whichheusesoccasionally.
of support from family and friends. The following anon-
ymised excerpt from our case summaries illustrates the Walter saysthat someone talked about himhavinga
uniqueness of assisted living needs in people with multi- pendant alarm butitdidn’t arrive. Hehadfallen 3or
morbidity while also highlighting a number of system-level 4timesinhisbedroom andhedidn’tknowwhathad
problemsthatwerecommonacrossmanyassistivetechnol- causedthefalls. Hewouldverymuchliketohave a
ogyusers(andnon-users)inbothstudysites: pendant alarm.
Walter,awhite Britishmanaged72,issingle and AmajorpracticalissueforWalterand hisadoptive
liveswith Christine[friend]along with herpartner family ishisincontinence.Christine saysthat the
Phil, andtheir fourchildren. […]Hehaschronic incontinencepadscost£13perpack andWalter
Greenhalghetal.BMCMedicine Page8of15
sometimesgoesthrough2packsperweek.Thisisa It is ironic that Walter has been equipped with a tele-
significant drain onthefamilyfinances.Theytried health package costing several hundred pounds (which
unsuccessfullytogetthemonprescription, and are he cannot use because the routine for sending the read-
nowtrying again.Christineexplains: ings does not align with the wider routine of his host
family), but has been classified as ineligible for the sim-
‘Lasttime,thenursesaid,“Writedownwhathedrinks, pler and cheaper solution of incontinence pads on pre-
howoftenhedrinksit,howmanytimeshegoesforawee, scription. The formalised assessment process, involving
seeifyoucanmeasurehiswee”,andI’matit‘You’re ‘objective’ measurement of the amount of urine passed
jokingaren’tyou,I’vegotfourkidsinthehouse,Ican’tbe with a view to categorising Walter as either ‘needing’ or
runninginandoutofthetoiletwhenhe’speeing’.Then ‘notneeding’incontinencepads,contrastswithChristine’s
whenIwroteitalldownshesaid“no,notgoodenough, account from a carer’s perspective: she knows Walter
he’snotentitledtothem.”Itgottothepointwherehewas intimately, and can describe from a subjective, lived-
leakingontothecouchandthey’retellingmehe’snot body perspective how the incontinence affects him and
goodenough,youknowwhatImean?’ the rest of the family. Similarly, the privileging of ‘ob-
jective’ measures of oxygen ‘need’ led to Walter’s oxy-
Waltergave upusing thetelehealth equipment gen service being withdrawn even though it gave him
becausethetimesoftaking themeasurementswerenot subjective benefit. The cases study thus raises (but does
convenient forhisfamily.Walter needed Christineto not answer) the question of whose perspective should
helphim take themeasurements,butbecauseof his ‘count’ in the provision of telehealth and other tech-
sleepdisturbance,hedidnotgetupuntillate nologiesfromalimitedbudget.
morning– bywhichtimeChristinewasupand out Walter’s case illustrates a much wider finding of our
of thehouse.Theyonlyused theequipment about ethnographicwork–thatoff-the-shelftechnologieswere
sixoreight timesintotal.Theresearcheraskswhether rarely useful or usable by people with complex medical
thetelehealth equipmentwasusefulwhilehewas needs. Rather, successful solutions, where they occurred,
using it. Herepliestherewerelotsof health had been produced for the participant through ‘brico-
professionals comingandgoing. Hethoughtthepoint lage’ – pragmatic, needs-focused customisation of the
of thetelehealth equipmentwastosavehealth technology by a person who knew them well [18]. But
professionals timebut,he says,itdidn’t seemto. his case also illustrates the limitations of bricolage, given
the current technological and service climate: Christine
Likemanyothercasesinoursample,Walterhasmultiple and Phil have made numerous material adaptations to
medicalproblemsthatinteracttoproducethecombination the house (for example, the kitchen door lock) to make
of low energy, low motivation and limited physical and it safe and accessible for Walter, but because the supply
cognitive capacity. Some of the installed technologies (for of telemonitoring service and domiciliary oxygen are
example, the weighing scales and thermometer) do not driven by system-level protocols, criteria and standards,
seem to match his medical conditions. Importantly, the Walter’s carers are powerless to customise these to meet
veryconditionsforwhichhewasdeemedtoneedassistive hisneeds.
technologies make him incapable of using them, mainly We have described Walter’s case in depth to highlight
because of their non-specific effects on his energy and key themes that were evident across many cases in our
motivation. His medical conditions are neither stable nor sample. Much more rarely, technologies were helpful and
fully diagnosed (his vacant periods, for example, may or valued – and this occurred when they extended existing
may not have a neurological origin). What matters to support from either family or professional carers. For ex-
him isgettingoutofthehouse–eithertothebackyard ample, participant Bonnie (aged 81) also suffered from
foracigaretteortothepubforhispint. chronicobstructivepulmonarydisease;herdaughterCarol
Walter’s case illustrates that telehealth relies on the liked the telehealth equipment, mainly because the oxim-
person’s own ability to use the equipment and/or on eter readings often allowed her to convince her mother
theabilityandwillingnessoftheirfamilyandfriendsto that there was no need to panic. However, using the
help them do so. Far from being ‘plug and play’, allow- equipment to obtain these reassuring readings involved
ing remote monitoring of the individual whatever their Carol making face-to-face visits and doing considerable
capability and motivation, the technology makes high additional work. In this respect, the technology was nei-
demands for cooperation and conformity – for ex- therlaboursavingnortimesaving.
ample, in this telehealth service, a phone call must be In sum, this ethnographic work revealed how people’s
made at 10 am. If the family routine does not mesh capacities and capabilities, shaped by both socio-cultural
with that of the telehealth service, the technology frames and the physical and cognitive effects of illness
quicklyfallsintodisuse. and ageing, align to a greater or lesser extent with the
Greenhalghetal.BMCMedicine Page9of15
material and symbolic properties of technologies in is,bye’.But actually,it’ssomething youhave gotto
particular settings. Most crucially, these rich case studies workwith. You’vegottogobackandgobackagain
have begun to characterise, in close clinical detail, how and make sure you reviewtheneed overall…Sothis
multi-morbidity affects people’s ability to use technolo- alwaysbringsithome tome,you know,whatyou do
gies and the (often limited) extent to which technologies inyourcase studies,whatreallymatters…’
can prevent or attenuate the suffering of multi-morbidity. (Occupational Therapist)
Furtherexamplesfromour40ethnographiccasestudies
are given in other academic publications [13,18]. In Experiences with installing assistive technologies
addition, 23 participants consented for the full text of brought participants to the view that practical reasoning
their case summary to be published on the open-access is required, focusing on individual contexts, material
ATHENEwebsite[33]. constraints and the ends that are to be achieved. How-
ever, they also felt that the importance of such reasoning
Co-designworkshops had notbeen fully acknowledged across the services,and
Workshops provided a lively and creative forum for that it was difficult to achieve because patients tended
people with assisted livingneeds,informalcarers,service tobepassedthroughdistinctcareteams,eachwithspe-
providers and technology suppliers to discuss the ethno- cific responsibilities and tasks along a so-called care
graphic data, share their experiences, and elicit technol- pathway with connotations of an inflexible,‘production
ogy and service design ideas to address issues raised. line’approach.
Key themes relating to the design of quality telehealth
and telecare solutions included customisation and ‘If you’reworking ina service,you wanttoknow,I’m
adaptation; information sharing and coordination; and doing this,this,thisand ifIhave donethat,thenit’s
ongoing social interaction and support. These are pre- attheend ofmydutyandIcansortof passthis
sented inturnbelow. person on…Butinreality,the situations thatyou face
inhealthand socialcare,they’re socomplex and
Customisationandadaptation confused thatif you reallywantedtoaddress
Users, technology designers and service providers re- somebody’sneeds,it’slike aminiproject.Wheredo
peatedly emphasised the need to provide tailored solu- you start and wheredoyouend?’(Occupational
tions and gave numerous examples of barriers to Therapist)
effective customisation. The initial assessment visit was
considered particularly critical to getting to know the One service provider described a one-off situation in
patient/client, especially the specifics of how they live which shewas(unusually) able totailor telecaretechnol-
and their experience of their health condition. An im- ogy to an individual patient’s needs, because she did not
portant component of this visit must be to spend time havetofollow setproceduresandprotocols:
talking with the end user and those close to them in
order to find out what matters to them and ensure ‘RecentlyIdid anOTassessment foraladywhowas
that any technology solutions are fully personalised. not eligiblefor socialcare.And soIwentinto –almost
Workshop participants commented that this counsel of likeinanadvisorcapacity,assessed herand
perfection was difficult to achieve in current clinical and everything, butitturned outwhat shereallywanted,
social care practice, and that few technology suppliers are whatwasreallyof value toher,wascompletelyoutof
sufficientlyskilledtoundertakethiswork(whichwascon- thebox,youknow.AndIkind ofmadeloadsofphone
sideredtorequireclinicalorclinically-relatedtraining). calls,Iwentonline,tocontact variouspeople and look
atwebsites,asweweredoing this…Andinsteadof
‘However muchtraining youdoandhowevergood kind ofdoing the standard,whichIwould have
people are atdelivering telecare,unlesstheytake into normallydone,becauseitwasoutsideofthe statutory
accounttheperson’ssituation andhow theylivein circuitIcoulddothis. AndIsortoffelt,youknow,this
their home,it’sgoing toberubbish.Imean,ranging isreallyquite good,thisismuchmore like arolethat
fromnot noticing they’vegot adog,alarge dog, which Ibelievewouldhelppeople. …Soit’snot all aboutthe
canmuckupthebed sensor somethingrotten, or,for technology itself,it’salsoaboutthe approach.’
instance,thattheyuse awoktocookwith,whichis (Occupational Therapist)
not verygoodif you’vegot ahigh temperature alarm
inthekitchen…Butit’sreallyabout talking tothe Beyond the challenge of understanding user require-
person,spending timewith them,not justonce. ments, personalisation appeared to be further hindered
Becausethe[current]ideaisit’slike aprescription, by the limited range of technologies available locally.
isn’tit?You getthisTelecareprescriptionand‘thereit Service providers were often ‘locked in’ with particular
Greenhalghetal.BMCMedicine Page10of15
devices and brands, provided as a standard package, and enable continued development and customisationofas-
there were contractual limits on what could be provided sistive technology solutions. Participants’ suggestions
by and to whom. Commissioners made the purchasing aligned closely with findings from the computer-
decisions, and clinicians then had to make the best of supported cooperative work literature that social and
what was available. The purchasing model preferred by technical subsystems should be organised to support
our participants was characterised by greater engage- collaboration through mutual awareness (the sense of
ment between commissioning and service staff; greater whattheothercollaboratorsaredoinginordertoprovide
control and flexibility to explore and trial different tech- acontextforyourownactivityonacommonproject)[34]
nology options; and a change in relationship between in- and facilitate sharing of both ‘formal’ knowledge (docu-
dustry and commissioners (which currently assumes mented and accessible by people within an organisation)
purchase oftechnologiesinbulk). and‘informal’knowledge(gainedovertimethroughevery-
day practice, and not generally documented) [35]. Partici-
‘Ithink someoftheproblemswe’vegotisthe pants suggested, for example, that an assistive technology
equipment in[nameofcity] wasboughtby should be reviewed in conjunction with a routine clinical
commissioners,non-clinicians,and hadtherebeen carevisit,whichcouldhelpsustainengagementwithusers
more engagementwith theclinicians whoweregoing to inamorecost-effectiveway.
useit,whohad anunderstanding ofthepatientswho
weregoing touseit,itmightbe slightlydifferentthan ‘Itmightbethatbyknowing that,say,adistrictnurse
whatwe’vegot.Theyjumpedinfeetfirst.…But isplanningareview,wecould slotinacouple of
ultimately,thedecision wasmadeonthebasisof simplequestions and savetheneed foranother
howmuchmoney,howmanyunits,anditwasa resource,orthen catchupwith thatdistrictnurse
commissioning decision,not acliniciandecision.’ afterwards…Orcome along, and sharethat
(TelecareLead) information.Ijustdon’tthinkwedoenough ofthat
and maybe there are somequickwinsabout when
Informationsharingandco-ordination thingsareplanned in.’(Telehealth ServiceManager)
Participants identified a need to support knowledge
sharing and co-ordination within and between care ser- Such collaborative models could potentially overcome
vices, as well as across formal and informal care net- many of the difficulties of data integration and patient
works. Currently, service chains – with several people consent to share personal and health information. But as
involved in supporting an individual patient or client – our workshop discussions highlighted, they presuppose
are complex and lack effective integration and informa- an altruistic and collegial rather than commercial or
tion sharing. To the extent that aspects of the telehealth contractual relationship between different professional
or telecare service (for example, installation, the moni- staffandtheirrespectiveorganisations.Inreality, achiev-
toring centre) are outsourced to subcontractors, this can ing high motivation across the multiple actors to engage
addanotherlevelofseparation. and contribute to the collective task of supporting an in-
dividual over time will be hard to achieve. Prioritising
‘Soif youarereferring toanother servicefor some the subjective lived experience of the patient over the
intervention,youcould haveclosedthecasehoping the application of standardised criteria and checklists as the
referralyou’vemadeisthen going todothatbitof shared quality outcome could go some way to strength-
work.Youmighthavetheluxuryofactuallyphoning ening thiscollective effort.
orcontacting that service and discussing something
together,butthat canbe aluxurytobe abletodo ‘Iprefertheidea–it’sidealisticmaybe,butthat
that. Orevenif youdothat,that service are saying, everything islooked atbecauseit’sworking forthe
we’vegot atwomonthwaiting list,we can’t seethis person,it’stheholisticword again,buttrying toget
person forages.Soagain,thebarrier,youknow,is that ticking over witheveryone understandingwhat
therephysically toactually worktogether.Andthen the aims are sowe’re allworking towardsthe same
thingsbecome abit sortof strungout,itbecomes abit thing.’(AssistiveTechnologyLeadfor AdultCare
likeChinesewhisper,asthingsgo throughdifferent Services,Local Authority)
services.’(OccupationalTherapist)
Altruism and ‘pro-sociality’ (going beyond formal job
The workshops identified that improved intra- and requirementsand procedures) inhealthcare isoften seen
inter-agency coordination and information sharing is as an integral part of the job to help the patient. This is
needed to track users’changing circumstances and needs, because the precise combination and sequence of skills
identifyanyactionsthatneedtobetakeninresponse,and to be used in particular circumstances cannot always be