Social Science & Medicine 132 (2015) 62e69

Contents lists available at ScienceDirect

Social Science & Medicine journal homepage: www.elsevier.com/locate/socscimed

Survival of the project: A case study of ICT innovation in health care Hege K. Andreassen a, *, Lars Erik Kjekshus b, Aksel Tjora c a

Norwegian Centre for Integrated Care and Telemedicine, University Hospital of North Norway, P.O. Box 35, 9038 Tromsø, Norway University of Oslo, Department of Health Management and Health Economics, Oslo, Norway c Norwegian University of Science and Technology, Sociology Department, Trondheim, Norway b

a r t i c l e i n f o

a b s t r a c t

Article history: Available online 7 March 2015

From twenty years of information and communication technology (ICT) projects in the health sector, we have learned one thing: most projects remain projects. The problem of pilotism in e-health and telemedicine is a growing concern, both in medical literature and among policy makers, who now ask for large-scale implementation of ICT in routine health service delivery. In this article, we turn the question of failing projects upside down. Instead of investigating the obstacles to implementing ICT and realising permanent changes in health care routines, we ask what makes the temporary ICT project survive, despite an apparent lack of success. Our empirical material is based on Norwegian telemedicine. Through a case study, we take an in-depth look into the history of one particular telemedical initiative and highlight how ICT projects matter on a managerial level. Our analysis reveals how management tasks were delegated to the ICT project, which thus contributed to four processes of organisational control: allocating resources, generating and managing enthusiasm, system correction and aligning local practice and national policies. We argue that the innovation project in itself can be considered an innovation that has become normalised in health care, not in clinical, but in management work. In everyday management, the ICT project appears to be a convenient tool suited to ease the tensions between state regulatory practices and claims of professional autonomy that arise in the wake of new public management reforms. Separating project management and funding from routine practice handles the conceptualised heterogeneity between innovation and routine within contemporary health care delivery. Whilst this separation eases the execution of both normal routines and innovative projects, it also delays expected diffusion of technology. © 2015 Elsevier Ltd. All rights reserved.

Keywords: Norway e-Health Health technology Health care management Innovation Telemedicine Case-study Qualitative

1. Introduction For about 20 years, we have witnessed steady growth in the number of ICT innovation projects in the health care sector. However, very few of these are brought into continued routine service. In fact, it has been suggested that the application is characterised by a “plague of pilots” where projects are established to be run as nonpermanent test-projects, rather than developed into normal practice (Wyatt and Sullivan, 2005). This critique echoes through later policies, as well as the telemedicine and e-health research literature (Broens et al., 2007; Helse og omsorgsdepartementet, 2013; Sosial-og Helsedepartementet, 2001; Zanaboni and Wootton, 2012). The problem formulation puts emphasis on the large number of local small-scale pilots and projects in health ICT, each of

* Corresponding author. E-mail address: [email protected] (H.K. Andreassen). http://dx.doi.org/10.1016/j.socscimed.2015.03.016 0277-9536/© 2015 Elsevier Ltd. All rights reserved.

which seems to meet the criteria for technological success, “yet fail [s] to become part of every-day clinical routines” (De Bont and Bal, 2008). Our empirical case is from Norwegian telemedicine. In Norway, as elsewhere, many of the promises of increased quality and efficiency of telemedicine and e-health have yet to be realised. As early as 1999, a Norwegian government report on telemedicine stated: “Through the financing of equipment and regional cooperation initiatives the Ministry [of Health] has granted funds for telemedicine activity in all health regions [of Norway]. It is now time to take a step forward, from single projects to the systematic use of telemedicine in routine [services] in areas where telemedicine does have a documented positive effect” (Sosial-og Helsedepartementet, 1999: 9, authors translation). This report established an important background for the innovation project that we studied: The Display Window (later referred

H.K. Andreassen et al. / Social Science & Medicine 132 (2015) 62e69

to as TDW). The quote above suggests that the implementation of health ICT innovations in the public sector is to be understood as a stepwise process. It is argued that we are ready to move from step one, single pilot projects, to step two, systematic application of telemedicine in selected routine clinical practices. The report thus states, from the ministry's point of view, that the “plague of pilots” problem (Sosial-og Helsedepartementet, 1999) has been identified and will actively be dealt with at a national policy level, as of the late 1990s. We observe that the ICT innovation project still survives in health care. In this article, we ask why this is so, when official health policy has long been to implement large-scale ICT applications in routine services. The official aims of implementing ICT in health care, such as increased efficiency and accessibility to health care, financial gain and patient empowerment, are clearly formulated and advocated in policies on all levels, from those of single institutions to overall national policies. Success criteria, as well as barriers to success and problems facing ICT implementation in health care, have already been thoroughly dealt with in the literature (May, 2013; Murray et al., 2011; Obstfelder et al., 2007). Overall, sociology dealing with telemedicine and e-health has been dominated by micro-level studies exploring the detailed relations between technologies and humans in practical health care work (Halford et al., 2010; Halford and Obstfelder, 2010; Langstrup Nielsen, 2003; Mort et al., 2009; Oudshoorn, 2008). Exceptions of note are some reviews looking for systematic patterns (Ekeland et al., 2010; Obstfelder et al., 2007), and requests for a stronger awareness of structure, as well as for policy and professional levels (Greenhalgh and Stones, 2010; Tjora and Scambler, 2009). Nevertheless, there is a lack of critical sociological studies investigating how ICT reforms are met and dealt with in everyday management work in public health care institutions. Indeed, a systematic review of the literature on the implementation of e-health found that methodological quality in this area was poor, and that very little information was provided on the ways in which managers and other users make sense of ehealth systems and appraise whether an e-health intervention is worthwhile or not (Mair et al., 2012). In this article, we aim to address this gap in knowledge: we investigate whether there are benefits of ICT innovation seen from a managerial point of view, and, if so; whether these can explain the persistence of ICT innovation projects in the sector. Thus, instead of looking into the technical outcomes of a health ICT project, i.e., whether it led to changes in clinical practice or to the adoption of new technologies, we explore the detailed contribution that a Norwegian telemedicine project “in the making” made to everyday health care management. The case illuminates how management responsibilities could be delegated to the innovation project, and emphasises that the project contributed to shape the processes of control in the organisations of which it was a part. Through the empirical analysis we develop the argument that innovation projects have been normalised in health care. Further, we draw on these findings to discuss how the growing number of ICT innovation projects relates to other contemporary reforms in public sector health care, and thus intervene in processes of governance. 2. Empirical case: The Display Window (TDW) The Norwegian Centre for Telemedicine (NST) initiated TDW in 1999 as a direct response to the previously mentioned ministry report where it was stated that: “The […] region should have the potential to appear as a display window for telemedicine solutions … A large-scale buildout of

63

telemedicine services [in this region] will show to other actors how the method can be applied, and provide a laboratory for testing new solutions” (Sosial-og Helsedepartementet, 1999: 39, author's translation). The Norwegian Directorate for Health and Social Affairs funded the project, which lasted through 2002. Through this project, ICTlabelled as telemedicine-was introduced to ease the communication and collaboration between general practitioners (GPs) on the one hand and specialist hospitals on the other. Two forms of technologically mediated interaction were enabled: video conferencing and electronic messaging. The project funded four full-time telemedicine advisors, who systematically contacted all general practices in the selected region (about 180 at the time). The practices were offered the necessary technological equipment, as well as support in installing and running the services. At the hospital side the project was directed toward three medical specialities: dermatology, otolaryngology (ear, nose and throat) and cardiology. In total, 90 local GPs installed the technical systems necessary for telemedical specialist support within dermatology; seven of these GPs also had the equipment to offer tele-otolaryngology, and 41 offered tele-cardiology. In 2003, three spin-off projects continuing the work from TDW were conducted. For simplicity, we use “TDW” for both the original project and the spin-offs. At the time of our study in 2012, the previously involved actors at the Norwegian Centre for Integrated Care and Telemedicine (NST) explicitly deemed the project a failure. No routine telecommunication was active between the institutions that were once involved in the project. In fact, there had been no activity at all for a long time. The equipment had not been maintained or updated, and our informants had trouble remembering when they last had been using it and whether it was still working. 3. Theory In this study, we look at how innovation projects get involved in management work and can be a form of delegated organisational control. Organisational control is an important part of management work. Stiles and Taylor (2001) conceptualise organisational control as a combination of financial and strategic control: the broad mechanisms that shape mission and vision, regulate the capacity for innovation and entrepreneurship, and facilitate necessary change. In the day-to-day routine of a regular health care organisation, achieving organisational control requires hard work and major effort. Latour's (1992) concept of delegation aptly illustrates how the innovation project TDW could be a response to this challenge. Latour argues that major efforts are transformed into minor ones through delegation of work to humans and nonhumans, and claims that “every time you want to know what a nonhuman does, simply imagine what other humans or nonhumans would have to do were this character not present” (Latour, 1992: 155). This perspective proved relevant for our study. The TDW case illuminates how parts of the managerial work to perform organisational control can be delegated to a heterogeneous network of human and nonhuman actors: the innovation project. Through delegating certain tasks to the project, health care management and administration transform some of their major efforts into minor ones. In social science studies of technical innovations in health care, it has been stressed that research needs to look beyond “the thingin-itself” when studying the phenomenon: In Normalisation Process Theory (NPT), Carl May underlines that it is crucial to understand the processes of normalising an innovation in daily routine (May, 2013). Findings from our study are consistent with NPT's emphasis of “the dynamic collective work” and the “relationships

64

H.K. Andreassen et al. / Social Science & Medicine 132 (2015) 62e69

involved in the implementation and social shaping of practices” (May and Finch, 2009: 549). We consider the innovation project itself an organisational innovation, and study the normalisation of it through this lens. Research in health sector management has described how current management practice is influenced by new public management (NPM) reforms, and how this creates tensions in the relationships between the state and the medical profession (Hunter, 1996; Kjekshus and Veggeland, 2011). We have included these perspectives in the discussion of our findings, and draw on the governmentality concept (Foucault, 1979) to introduce a discussion on how health care innovation projects not only matter for day-to-day institution management, but also contribute to public sector governance. Whilst contemporary health and medical sociology deals thoroughly with both technology and management, the two topics are rarely coupled. Our work contributes to combining knowledge from these hitherto separate fields, thus revealing new empirical insights and inviting new discussions of the topic. Resting the analysis on a combination of theoretical concepts, i.e., delegation, normalisation, and governmentality, has allowed for the consideration of several € ldberg's meanings of the material, drawing on Alvesson and Sko (2000) proposition of a reflexive interpretation within qualitative analysis. We discuss the significance of the innovation project on different levels of abstraction. The result is an empirically driven analysis of health sector management practice that informs a discussion on public innovation, health ICT reforms and governance. 4. Methods and material 4.1. Scope and sampling The “plague of pilots” hypothesis (Wyatt and Sullivan, 2005) in telemedicine was already acknowledged when we started our study, and was a significant inspiration for our first step: an initial mapping of Norwegian health ICT services from the last 20 years. Following a theoretical sampling procedure (Mason, 1996), our aim was to identify applications, projects and initiatives that could be of special relevance to our research interest in the slow diffusion of ICT in health care. The mapping followed a qualitative logic and was based on document analysis, ten semi-structured employee interviews and three open plenary group discussions at NST. We wanted to learn about the most interesting stories of ICT innovation from the viewpoint of actors within the field. Through this initial mapping we learned of TDW, a finalised, unsuccessful project that had been intended to ensure large-scale implementation of three different ICT innovations within tele-dermatology (skin), teleotolaryngology (ear, nose and throat) and tele-cardiology (in this case, electronic transmission of the heart sound of newborn babies). As a second step, we selected TDW as our case for an in-depth interview study. For the interview study, we used purposive sampling of participants from the TDW project whom we met during our initial mapping of Norwegian health ICT services. This resulted in four interviews, which were then followed by snowball sampling (Salganik and Heckathorn, 2004) resulting in a total sample of 21 individual participants. The position of the participants was the basis for recruiting them to the study, and we ensured that all levels of project management were included in the sample. Of our 21 participants, seven had been directly involved in the TDW project work group, three were local health institution managers, three represented regional health care administration, two were from state health bureaucracy, four were professionals who had managerial positions in day-to-day clinical practice (and who had been working in institutions where TDW had been implemented) and

two represented the ICT vendors. 4.2. Interview methodology Our interview data is retrospective: the interviews concentrate on actions that have taken place in the past. Although we were not able to observe the actions of the project, the retrospective accounts of project stories retold after being “coloured” by time are of value as such. Through the inter-subjective setting of the in-depth interview, the participants shared how they saw the innovation project's impact on their organisation and management as a process unfolding over time. The retrospective interviews produced project participants' own reflections on project attributes, and thus allowed for a meta-reflection related to our theoretical framework on delegation (Latour, 1992) and normalisation of innovations (May and Finch, 2009). The participants' accounts, not only of “what happened”, but also of “why it happened”, are conducive to an understanding of the normalisation process (Murray et al., 2011). The interviews took place between February and September 2012. All interviews were carried out in the Norwegian language, and were audio-recorded and transcribed. As part of the interviews, we asked the informants for important documentation suited to illuminate the case, and as a result 29 documents were included in the analysis. Ethics approval was administered through an agreement with, and informed consent from, participants in the study, who were all experienced professionals. Because no patients were included in the study, nor patient information, the study was not relevant for an ethical committee approval. The case project TDW was organised as a health service quality assurance project, and complied with the ethical requirements for such. 4.3. Analytic approach It is commonly agreed that the coupling of purposive sampling and analytic induction strengthens the generalisability of qualitative research (Silverman, 2001). In this work, accordingly, interview data from a purposive sample were analysed through an inductive process. We identified detailed accounts of project initiation, maintenance and context to conceptualise, on a general level, how the case project unfolded over time. In our interviews with core personnel from the TDW project and the institutions hosting it, we learnt how the project had interacted with its organisational surroundings. Once it was rolling, the ICT project had contributed to more than technology diffusion; it had also played a role in management practice. In the informants' stories, we saw examples of how certain managerial tasks had been delegated to the project. We grouped the empirical examples according to their contribution, which enabled us to identify four distinct processes where responsibility had been partly delegated to the project: Allocating resources; generating and steering enthusiasm; reflexive system correction/challenging existing practice and aligning local practice and national policies. The four processes structure the analysis that follows. 5. Delegation of managerial tasks to an innovation project 5.1. Allocating resources The first task that we found delegated to the project was the task of allocating resources to innovation work. Naturally, such delegation can be described as an aim of all innovation projects. Nevertheless, our material reveals that this practice deserves empirical reflection. One of our interviewees reflected on the fate of TDW:

H.K. Andreassen et al. / Social Science & Medicine 132 (2015) 62e69

“I am not exactly sure of what happened after [the project]. We were involved in implementation and organising, and then afterwards, it needed to be operated, and that … well it is my impression, at least I have not heard of it, that the funding for operating this [project] in daily work was not in place. It was project-funded, so that as long as there is a project, then you have funding for the different actors and so forth, but then afterwards when you need to stand alone on your own feet then you are totally dependent on the money supply that follows from doing what is ordered: the job has been done and the bill is sent.” Project funding has fostered much of the growth in Norwegian health ICT during the last 20 years, and this is still the situation; this has also been observed in other countries (Lepori et al., 2007). A practical consequence for health care managers is that ICT innovation projects add external funding to the running budgets and become tools to rearrangedand, in some cases, challengedthe established workflow and allocation of human and material resources. Our case, TDW, is an example of an initiative that attempted to meet the government demand for systematic use of telemedicine in routine services. Still, it was organised as a project; it had its own funding and its own staff operating alongside, although in cooperation with, the established organisation. A hospital manager who had been involved in the project explained: “The health care sector delivers what is requested, and then, in addition, we have some enthusiastic projects around things that have not been asked for, but that we think are fun; and those last as long as they remain fun.” This quote illustrates how routine work (what is requested), and innovative work (what we think is fun), are conceptualised as coexisting but heterogeneous tasks. The funding facilitated reallocation of both human and material resources in order to create an arena for “things that are fun” and at the same time meet the project goal of ICT innovation and implementation. From a managerial perspective, the project was a tool that eased flexible handling of resources in an otherwise strictly regulated sector, characterised by rather static hierarchies and routinised practices. This challenge would have had to be met by other means had the project not been present. Further, we have observed that project organising in the ICT innovation area holds advantages beyond the obvious aspect of funding and resources. TDW might have been intended to kick-off more enduring solutions, but once the project's organisation was in place, other tasks and responsibilities could be delegated to the project too. We describe this in detail below.

65

the] individual, but for the institution; [how to] enable enthusiastic participation in the development.” The quote underlines that local commitment and enthusiasm are seen as important aspects of professional development and clinical improvements in health care, and further that supporting the enthusiast is a task considered important by health care administration. We interviewed members of the project staff of the TDW initiative who had been involved in the actual hands-on implementation of the technological innovations. They explained that the approach had been stepwise and intended to foster enthusiasm. They had approached all local GPs in the three selected counties of Norway by letter, followed up with a telephone call. If the GPs were interested, they physically went to visit them and place the equipment at their offices, as well as to give them the necessary training. We asked if there had been any active opposition toward the new technological services: “Yes, yes, there are many who are very clear that they do not believe in it, like ‘just forget it; I don't want that.’ So, then, in a way, as a theory for implementation, you need an enthusiast in the end there, someone who ‘bootstraps’ the organization. So that, well, the enthusiasm: that had to be present.” This illuminate how TDW, in the making, contributed to both generating and steering professional enthusiasm: an appreciated element in everyday health care management and overall sector development. The interest in generating enthusiasm was shared by all levels of health care management and administration, from day-to-day institution managers to national administrative bodies. An interviewee from the Norwegian national health bureaucracy explained about TDW and the host organisation, NST: “Seen from the perspective of a technocrat, the governance was weak [regarding] the continuous follow-up of the funding and those aspects of it. At the same time, the dynamic and enthusiastic entrepreneur atmosphere was assessed as a positive quality of NST.” TDW became an arena for engagement and enthusiasm, understood to require less governance and more personal engagement than other health care practices. Management through projects is management through local enthusiasm. When the project generates and steers enthusiasm in certain directions, it has an active role in the performance of overall organisational control. Parts of the managerial responsibility of ensuring professional enthusiasm and engagement, as a means to achieve good organisational performance, were delegated to the project.

5.2. Generating enthusiasm It has been shown that health care organisations with more engaged clinical staff appear to have better performance measurements (Spurgeon et al., 2008). Generating engagement and professional enthusiasm would thus be an important task for health management. A representative from regional health administration reflected on the role of enthusiasm in general health service development: “There is nothing heroic to it; it's just the way things are. It is what gives job satisfaction; it is a core driver. We are currently discussing this, how to support the enthusiasts in a health care system with strong demands with regard to productivity. Not for the [sake of

5.3. Reflexive assessment of existing practices Management responsibilities also encompass continuous assessment and improvement of existing practices. Ensuring system correction through stimulating self-reflection was a third responsibility that was, in part, delegated to TDW. In one of our interviews, a representative from the national bureaucracy explained the background for establishing the NST in the first place, and the role of such a centre in the health care system: All in all, they [NST] challenge the system in an all-right way. I think that this is the most important resource they offer. When they can

66

H.K. Andreassen et al. / Social Science & Medicine 132 (2015) 62e69

dig a bit into things, such as, ‘This procedure, why continue to do it this way?’” Our informant expressed a need to constantly assess and reshape the established procedures in health care, and, commented further, that pushing technological innovations could be one way of doing this. This aspect of innovation projects is rarely communicated in official policies and statements. Officially, ICT innovation projects are presented as solutions to politically identified problems in health care, such as keeping the system efficient and of high quality despite growing demands and changing national demographics. The arguments for investing in health ICT development that we found in the document material included in our study were technology deterministic in that they stressed how the planned ICT innovation would provide certain predictable effects and bring us “forward”. As illustrated by the quote above, there was an alternative representation in our in-depth interviews with representatives of public sector health care administration. The process of testing technological innovations was conceptualised not only as a means to reach other goals, but also as an end in itself, a necessary tweaking of the system. This account represents an interesting contrast to the dominant technology-deterministic presentation in official policy documents. Evidently, this is an aim that can be reached without large-scale implementation of new innovations. The possibility to delegate the responsibility to “dig a bit into things”, and have state-run projects imply questions like “this procedure, why continue to do it this way?”, ensures important input for strategic management work performed on regional and national levels. During one of the interviews, a participant from the project group explained how he had once been involved in another ICT innovation project where the aim was to adjust the organisation of health care services in his region. However, this other project was also a failure, in the sense that it did not result in implementation of ICT in routine clinical work. Our informant reflected on why the implementation of the technology had faltered: “We did a pilot … and it resulted in a pretty comprehensive report that described, in detail and on a routine level, how services and occurrences should be handled, from the specialist hospital and in the single municipalities. It was […] when you wanted the final financial resources to implement it that […] things got complicated.” The quote above illustrates how a “failed” local innovation project can contribute to system correction through suggesting in detail and on a routine level how services and events can be handled in new ways. Our informant further talked about how the suggestions from this project report actually aligned with the goals of the latest national health reform in Norway, the Coordination Reform, launched some years later, in 2008. The responsibility to ensure reflexive system correction that lies with regional and national health care administration can be delegated to local projects. The willingness to test new solutions and the ability to think new thoughts that comes with innovation projects affect the organisation in which the project is anchored in fundamental ways. Piloting innovations is a way of opposing the established structures of health care from within. The corrective has a value of its own. 5.4. Aligning local health care practice and national policies An important aspect of public health care administration is

ensuring that local institutional practice and national health policies are well coordinated. This requires continuous attention and major effort from local health institution management and national policy makers alike. Local health managers have the responsibility to ensure their institutions are run in accordance with current national and regional health policies and administrative arrangements. The authorities on their side need to develop policies and arrangements in accordance with local needs in order to obtain legitimacy. This requires interaction and dialogue with local actors. In our case, some of this alignment work was delegated to TDW. First of all, TDW contributed to anchoring the problem representations of overall e-policies (e-government, e-transformation, etc.) in its local contexts. When Norwegian health authorities, like those of other Western countries, claim there is a need for more digitally mediated interaction, more efficient use of ICT and more large-scale e-health solutions in routine clinical practice (Sosial-og Helsedepartementet, 1999; Sosial-og Helsedepartementet, 2001; Helsedirektoratet, 2014), they present the lack of ICT in public health care as a problem in itself. Following the logic that presenting a solution also represents a problem or concept that must be solved (Bacchi, 1999), all ICT project plans and proposals in health care may be interpreted as confirmations of such problems. This also applies to the project we have studied. Further, it was not only the e-policies of the “informational paradigm” (Castells, 1996) and the problem representations belonging to these that were confirmed and reinforced through TDW. The project documents also connected the project to other policy and political processes on local and regional levels. Of interest for one of the regional administrations was a suggestion that increased access to the regional health services through video conferencing and electronic messages could reduce an evolving pattern in the patient population of patients turning to a neighbouring health region with a larger hospital. This “flight of patients” caused financial challenges on the regional level. Further, financial and quality estimates reported from the pre-phase of the project were relevant for local health management. The estimates indicated reduced travel costs for patients (a cost covered through local budgets), as well as a strengthening of local specialist competence through collaborative work, and, of course, better access to health services for their local community population. Our interpretation is that TDW confirmed four problem presentations from its contemporary health policies: (1) the lack of ICT in the public sector in general and in health care in particular; (2) the problem of high maintenance costs in Norwegian health care; (3) the quality problem of small medical specialist environments, and; (4) the problem of poor access to health services for rural districts. Interview participants explained to us how TDW was also relevant for an on-going political battle in one of the regions where it had been implemented. The regional health authorities there were challenged by an opposition toward recent health care reforms. The region had started a process of dividing different medical speciality functions between hospitals. In the local communities and the media alike, this process was met with anger and negativity. One of the interview participants explained that his impression was that the initiative to implement telemedicine was connected to an opportunity to integrate TDW with the already ongoing reforms to give a sort of compensation to the communities where the population was discontented. Through the telemedicine project, people would be given “something new”: a modern and high-tech service ensuring efficient access to specialist health care. “The idea was that they would be able to show how this [the reform] could contribute to something good.” The quote illustrates that in addition to confirming some of its contemporary policy problem representations, the project also

H.K. Andreassen et al. / Social Science & Medicine 132 (2015) 62e69

opposed another problem representation that has held a strong position in Norwegian political debate: namely, that a centralised hospital structure is an obstacle to equal access to health services. TDW played an active part in aligning local health care practice with national and regional policy processes. Through hosting the project, local health care institutions anchored their practice in contemporary ICT and health policies, and responded to some of the challenges formulated there. At the same time the project became an arena to manage and define local health policy problem representations. 6. Discussion: delegation, normalisation and governmentality The expected move in the field of telemedicine and e-health from step one, projects, to step two, full-scale implementation, has not happened. Most activities in this area are still undertaken through projects. Our analysis points toward an explanation of this situation that does not rest on the imperfection or immaturity of the individual projects, the specific technologies, the human users, nor in the individual organisations in which the innovations are expected to diffuse. Rather, the reason they do not move from step one to step two is that step onedproject organisingdholds the best view. 6.1. Delegation Our study has looked at how TDW, an ICT innovation project in health care, contributed to health care management. Through an analysis drawing on Latour's question of “what other humans and non-humans would have to do were this character (the project) not present” we have exposed how several local, regional and national management responsibilities, or tasks related to these, were delegated to TDW. The project was involved in the allocation of resources and in generating enthusiasm, it acted as system correction through challenging un-reflected segmentation of procedures, and, further, it turned out to be an arena where contemporary policy problem representations could be (re-) defined, confirmed or challenged locally. Hence, the project contributed to processes of organisational control. Based on these findings, we argue that the role of ICT innovation in the health care sector also needs to be studied from management and governance perspectives. Formal assessments of how ICT contributes to transforming service delivery and changing patient roles and clinical work routines are valuable, but they do not explain the totality of how ICT innovation projects contribute to shape contemporary public health care. 6.2. Normalisation As described, TDW was involved in complex networks of relations, and embedded in several management practices. Human engagement was “defined and organised with a material practice through long interaction chains” (May and Finch, 2009: 548), e.g., in the process of generating enthusiasm, where project staff worked to establish and foster enthusiasm for ICT, which in turn was received well and supported by managers in other levels of the organisation. The project was profoundly embedded in the dynamic collective work where social shaping of health management practice took place. The social processes that played out can be described as a normalisation process, in that they involved dynamic collective work where multiple relationships were involved in the implementation and shaping of social practice (May and Finch, 2009). The practices associated with the project were partly planned for, while others developed in the course of action. Using the project to allocate resources for innovation was indeed intended.

67

However, the sum of individual engagement and local enthusiasm coupled with the system correction and the policy alignment that the project made possible was only realised as these processes developed along the way, in a highly situated local context. May maintains that “the business of implementing and embedding an ensemble of practices associated with some innovation reflects the interaction between stochastic and purposive social processes” (May, 2013: 27). What we have identified as the integration of the ICT innovation project in health care management thus corresponds to what May et al. (2007) describe as normalisation, that is: the embedding of new complex interventions as routine elements of organisational work. Innovation projects are complex interventions that have normalised in health care. Recent statistics paint a similar picture. A report from 2012 found that 67% of Norwegian hospitals had run more than two projects “to improve patient treatment” that year, and only 4% claimed they had not hosted any such projects. Comparable figures from 2007 showed that the number of hospitals without projects was significantly higher 5 years earlier; 17% (Kjekshus and Bernstrøm, 2013). Although each project is temporary, the phenomenon of integrating innovation projects in everyday health sector management has been embedded as routine. The innovation project can in itself be considered an innovation that has been normalised, not in clinical, but in management work. It has become a part of the everyday workings of the public health care system. 6.3. Governmentality This case study suggests that the large number of short-term small-scale pilots and projects in health ICT is not merely an expression of failed implementation strategies. There is a need to discuss the growing number of projects in a wider context and relate them to on-going changes in public governance. Studies of public sector reforms in Norway have emphasised that recent health care reforms have changed the role of the medical profession. Up until the 1980s, Norwegian medical doctors, supported by a strong union, could be described as a largely self-governed professional group, but during recent decades, public sector reforms have increasingly made them an instrument of the state (Kjekshus and Veggeland, 2011). The observed tension between the state and medical professional interests, caused by neo-liberal politics (Hunter, 1996), is an important part of the context in terms of understanding the growth in ICT innovation projects in the public sector. As shown, some of our informants with managerial and administrative responsibilities talked about the need to ensure personal commitment and enthusiasm among their staffs. This is a challenge in a state-regulated system focused on productivity and measurable performance indicators. Thus, our empirical material reflects some of the tensions emerging in the wake of NPM reforms, notably the changes in the role and power of the medical profession. Numerous individual enthusiasts, driven by technologyoptimist visions, are committing to ICT projects. This could, in part, explain the continuous emergence of new ICT projects, but does not stand in the way of our interpretation of the projects' affordances on a managerial level. Each of the ICT innovation projects that have been introduced in health care during the last 20 years could very well have been initiated for other reasons than to contribute to organisational control, but still, the flexibility and ability to serve managerial interests can in part explain what keeps the practice alive and growing. Project funding and organising adds flexibility and room for local adoption in an increasingly regulated sector. We have found that ICT innovation in health care is handled through mobilising enthusiasm, thus motivating staff to create

68

H.K. Andreassen et al. / Social Science & Medicine 132 (2015) 62e69

change through project organisation. Governing through projects can be described as governing through individual and local engagement and initiative. In sum, the many single projects result in a relatively stable flow of enthusiasm and system correction in public sector organisations. “The centralisation of the project is part of a process of decentralisation”, argues Danish philosopher Jensen (2012: 31). He has launched the theory that our historical era can be described as a period in transition from traditional disciplinary society, as was described by Foucault, to a new age that Jensen labels “project society”. One of the differences between these two fundamentally different ways of organising the social is the handling of problems. In a project society, overall, societal challenges are handled in local contexts through smaller projects; in contrast, within disciplinary society, such problems are solved by central governing bodies. The aspects underlined in the critical accounts of Foucault and Jensen are relevant to the practical discussions arising from our empirical analysis. Applying a governmentality perspective (Foucault, 1979), we may ask if ensuring enthusiasm, learning and system correction through project-based innovation is a “new wave” of governing public health care in Norway. As the governmentality concept illustrates so well, governing through people's mentalities instead of imposing external rules is extremely effective. Our findings suggest the relationship between the organising of innovation as projects and changes in public sector power structures and governance should be the subject of further research. Who gains power through the growing number of innovation projects in the public sector? Where is power accumulated in contemporary health care structures? The answers are not simple; there is not one single actor or institution controlling the process. As we have seen in our empirical analysis, the growth of ICT innovation projects in the health care sector has evolved in a complex social environment, where the innovation project itself is best understood as one of the actors. We might be witnessing a dispersion of power and responsibility, leaving us with a system where no single actor has the competence, nor the means, to get the complete overview. Our study suggest that the critique as well as the continued embracing of pilots and smaller projects in telemedicine and e-health can be understood in light of contemporary tensions in the relation between state regulations, medical professional autonomy and ICT innovation. The growth of ICT innovation projects in the health sector is linked to other reforms, and affects power structures in public sector health care. On the one hand, the normalisation of the innovation project might result in local commitment and influence, in line with the ideals of recent reforms oriented toward NPM (Christensen and Lægreid, 2001). On the other hand, the increasing number of ICT innovation projects in health care may support the arguments of the critics of NPM reforms. There is growth in local responsibility to audit budgets and day-to-day activity (Power, 1999), but no delegation of actual power to influence goals or political direction, resulting in alienation and less actual power to shape health care in line with professional ideals. 6.4. Concluding remarks We conclude that from a managerial perspective there are important benefits from organising health ICT innovation through projects. This is because efforts to implement organisational control in the health care sector can partly be delegated to innovation projects. For local managers, delegating responsibilities for resource allocation and for generating enthusiasm in an innovation project, eases the day-to-day administration of their health care institutions. For regional and state administrative bodies, health ICT innovation projects ease and ensure governance of the health care sector, when they provide an arena for local engagement and

commitment as well as alignment of policy and local practice, in addition to contributing to system correction on an aggregated level. We argue that the ICT innovation project's affordances in management and governance processes can explain how and why the concept of ICT innovation projects has normalised in contemporary health care. Separating ICT activities from routine practice has become a way to handle the conceptualised heterogeneity between innovation and routine within health care delivery. While this separation eases the execution of both normal routines and innovation projects, it also delays expected diffusion of technology. Consequently, more ICT projects are then initiated. And this is how the project survives; it fits perfectly into the social conditions that it contributes to create. References € ldberg, K., 2000. Reflexive Methodology e New Vistas for QualiAlvesson, M., Sko tative Research. Sage, London. Bacchi, C.L., 1999. Women, Policy and Politics: the Construction of Policy Problems. Sage Publications Ltd. Broens, T.H., Vollenbroek-Hutten, M.M., Hermens, H.J., van Halteren, A.T., Nieuwenhuis, L.J., 2007. Determinants of successful telemedicine implementations: a literature study. J. Telemed. Telecare 13 (6), 303e309. Castells, M., 1996. The Rise of the Network Society. Blackwell, Oxford. Christensen, T., Lægreid, P., 2001. New Public Management: the Transformation of Ideas and Practice. Ashgate Pub Limited. De Bont, A., Bal, R., 2008. Telemedicine in interdisciplinary work practices: on an IT system that met the criteria for success set out by its sponsors, yet failed to become part of every-day clinical routines. BMC Med. Inf. Decis. Mak. 8 (1), 47. http://dx.doi.org/10.1186/1472-6947-8-47. Ekeland, A.G., Bowes, A., Flottorp, S., 2010. Effectiveness of telemedicine: a systematic review of reviews. Int. J. Med. Inform. 79 (11), 736e771. Foucault, M., 1979. Governmentality. m/f 6(3). Greenhalgh, T., Stones, R., 2010. Theorising big IT programmes in healthcare: strong structuration theory meets actor-network theory. Soc. Sci. Med. 70 (9), 1285e1294. Halford, S., Lotherington, A.T., Dyb, K., Obstfelder, A., 2010. Un/doing gender with ICT? Nord. J. Fem. Gend. Res. 18 (1), 20e37. http://dx.doi.org/10.1080/ 08038741003626791. Halford, S., Obstfelder, A., 2010. Changing the record: the inter-professional, subjective and embodied effects of electronic patient records. In: Halford (Ed.), 2010-New Technology, Work and Employment. Wiley Online Library. New Technology. Helse og omsorgsdepartementet, 2013. Morgendagens omsorg (St. meld. nr. 29 (2012e2013). Helsedepartementet S.-o., 1999. Telemedisin og veien videre. Rapport fra arbeidsgruppe. Sosial og helsedepartementet (Ministry of health), Oslo. Helsedepartementet S.-o., 2001. “Si @!” Statlig tiltaksplan 2001e2003 elektronisk samhandling i helse- og sosialsektoren. n journal”. HelsedirektorHelsedirektoratet, 2014. Utredning av “en innbygger e e atet, Oslo. Hunter, D.J., 1996. The changing roles of health care personnel in health and health care management. Soc. Sci. Med. 43 (5), 799e808. Jensen, A.F., 2012. The Project Society. Aarhus Universitetsforlag. Kjekshus, L.E., Bernstrøm, V., 2013. Helseforetakenes interne organisering og ledelse: INTORG 2012. University of Oslo, Oslo. Kjekshus, L.E., Veggeland, F., 2011. State regulatory capacity: experiences from public sector reforms in Norway. Public Adm. 89 (4), 1568e1584. Langstrup Nielsen, H., 2003. The multiplicities of a modest innovation: the introduction of a monitoring technology for asthma treatment in primary care. In: The EGOS 2003, Track 34: “Materialities of Organizing”. Latour, B., 1992. Where are the missing masses? The sociology of a few Mundane Artifacts. In: Bijker, W.E., Law, J. (Eds.), Shaping Technology/Building Society: Studies in Sociotechnical Change. MIT Press, Cambridge, Mass. Lepori, B., Van den Besselaar, P., Dinges, M., Van der Meulen, B., Potì, B., Reale, E., Theves, J., et al., 2007. Indicators for comparative analysis of public project funding: concepts, implementation and evaluation. Res. Eval. 16 (4), 243e255. Mair, F.S., May, C., O'Donnell, C., Finch, T., Sullivan, F., Murray, E., 2012. Factors that promote or inhibit the implementation of e-health systems: an explanatory systematic review. Bull. World Health Organ. 90 (5), 357e364. Mason, J., 1996. Qualitative Researching. SAGE, London. May, C., 2013. Agency and implementation: understanding the embedding of healthcare innovations in practice. Soc. Sci. Med. 78 (C), 26e33. http:// dx.doi.org/10.1016/j.socscimed.2012.11.021. May, C., Finch, T., 2009. Implementing, embedding, and integrating practices: an outline of normalization process theory. Sociology 43 (3), 535e554. May, C., Finch, T., Mair, F., Ballini, L., Dowrick, C., Eccles, M., Rapley, T., et al., 2007. Understanding the implementation of complex interventions in health care: the normalization process model. BMC Health Serv. Res. 7 (1), 148.

H.K. Andreassen et al. / Social Science & Medicine 132 (2015) 62e69 Mort, M., Finch, T., May, C., 2009. Making and unmaking telepatients: identity and governance in new health technologies. Sci. Technol. Hum. Values 34 (1), 9e33. Murray, E., Burns, J., May, C., Finch, T., O'Donnell, C., Wallace, P., Mair, F., 2011. Why is it difficult to implement e-health initiatives? A qualitative study. Implement. Sci. 6 (6) http://dx.doi.org/10.1186/1748-5908-6-6. Obstfelder, A., Engeseth, K.H., Wynn, R., 2007. Characteristics of successfully implemented telemedical applications. Implement. Sci. 2 (1), 25. Oudshoorn, N., 2008. Diagnosis at a distance: the invisible work of patients and healthcare professionals in cardiac telemonitoring technology. Sociol. Health Illn. 30 (2), 272e288. Power, M., 1999. The Audit Society: Rituals of Verification. OUP Catalogue. Salganik, M.J., Heckathorn, D.D., 2004. Sampling and estimation in hidden populations using respondent driven sampling. Sociol. Methodol. 34 (1), 193e240.

69

Silverman, D., 2001. Interpreting Qualitative Data, Methods for Analysing Talk, Text and Interaction, vol. second ed. Sage, London-Thousand Oaks-New Delhi. Spurgeon, P., Barwell, F., Mazelan, P., 2008. Developing a medical engagement scale (MES). Int. J. Clin. Leadersh. 16 (4), 213e223. Stiles, P., Taylor, B., 2001. Boards at Work: How Directors View Their Roles and Responsibilities. Oxford University Press. Tjora, A.H., Scambler, G., 2009. Square pegs in round holes: information systems, hospitals and the significance of contextual awareness. Soc. Sci. Med. 68 (3), 519e525. Wyatt, J.C., Sullivan, F., 2005. ABC of health informatics: eHealth and the future: promise or peril? Br. Med. J. 331 (7529), 1391. Zanaboni, P., Wootton, R., 2012. Adoption of telemedicine: from pilot stage to routine delivery. BMC Med. Inf. Decis. Mak. 12 (1), 1.