Seite - 201 - in Applied Interdisciplinary Theory in Health Informatics - Knowledge Base for Practitioners

protocol and recruited few patients, citing previous poor experiences with telehealth, concern that a remote monitoring service would threaten their jobs, or a belief that patients ‘deserved better’. The implied role change for the specialist heart failure nurse was potentially far-reaching. Instead of spending her time seeing patients in clinic or visiting them at home, nurses would now be spending a proportion of their time sitting in a data processing centre looking at on-screen data and trends. Furthermore, one driver for the introduction of the telehealth programme was a rapidly rising incidence of heart failure (and, because of improved care, patients were surviving many years after diagnosis). One cardiologist spoke of a health economic model in which the case load for each nurse would increase from 35 to 200 patients. Whilst some nurses embraced this vision enthusiastically, others strongly resisted it on the grounds that a dramatic reduction in direct patient-facing activity meant that they were no longer being heart failure nurses. Patients expressed a wide range of views about remote biomarker monitoring in the SUPPORT-HF study; some took an active interest in their readings, engaged enthusiastically with the feedback they received, and found this monitoring reassuring. Others found the experience confusing and burdensome; they did not know (and did not wish to learn) what the numbers meant. In some cases, a research nurse who knew the patients well provided (unofficial) telephone support to maintain engagement. The organisations: Participating sites in the SUPPORT-HF study were generally semi-autonomous cardiology units based in large district general or teaching hospitals. With few exceptions, leadership and managerial relations were good and (because of research support funding for the trial) there was sufficient financial slack to support introduction of the technology. As a research initiative, the SUPPORT-HF technology was not integrated into mainstream services, but we tentatively predict that because of the major knock-on implications for work routines (especially in relation to community heart failure nurses), this technology will be experienced as ‘disruptive’ and hence prove difficult to mainstream after the ‘proof of concept’ phase ends. One further external factor is the complexity of heart failure services, which typically span general practice, community clinics and hospital services – each of which has a different funding stream and different patient caseload. A telehealth-supported service in one of these sectors may need to interface with other sectors in the same locality that do not support (and perhaps do not trust) telehealth. The wider system: The SUPPORT-HF study unfolded at a time when there was a strong policy push for telehealth initiatives in general and for initiatives to reduce outpatient attendance in particular. But whilst the policy environment was positive, our data showed that in some sites up to half the eligible patients could not be randomised because of the variability of broadband speed outside the main cities. The extent of inter- organisational networking among participating departments in the SUPPORT-HF study was limited as this was not an explicit component of the trial intervention; we suggest that if this technology is introduced as a business-as-usual intervention post-trial, networking and knowledge-sharing among organisations should be supported (either via a virtual link or occasional face to face meetings). Evolution and adaptation over time: The tablet technology used for SUPPORT- HF included some limited scope for adaptation and customisation, but our qualitative data suggested that both staff and patients wished to adapt it further (either to accommodate individual needs and preferences or to adjust to external factors such as a changing technical infrastructure in the participating service). We are somewhat T.GreenhalghandS.Abimbola /TheNASSSFramework–ASynthesisofMultipleTheories 201