Telemedicine services: How to make them last over time

Health Policy and Technology (]]]]) ], ]]]–]]]

Available online at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/hlpt

Telemedicine services: How to make them last over time Federica Segaton, Cristina Masella Department of Management, Economics and Industrial Engineering, Politecnico di Milano, Italy

KEYWORDS Healthcare; Telemedicine; Implementation; Sustainability; Qualitative research; COPD; Italy

Abstract Objectives: Telemedicine is extensively used in healthcare settings, although we still lack knowledge on how to make telemedicine services last over time. This study aims to: investigate how the factors supporting the implementation of telemedicine services affect their duration over time; explore if further factors need to be considered, to foster the services duration. Methods: We conducted a six-year in-depth study on three Italian cases of telemedicine services lasting more than 10 years. Dimensions explaining the duration of services over time are explored and discussed against existing literature. Results: The three cases show that, to support the duration over time, financial and organizational stability should be set before the “champion” leaves the service. Financial stability was reached through different strategies. About organizational stability, we found that providing opportunities to enrich the competences and getting more responsibilities over the patients enhanced the professionals’ acceptance, which, in turn, supports the organizational stability of the service over time. About patients, to meet their crucial needs for their health and to receive the nurses’ support on the use of technologies contains the abandon and increases the chances for the service to last over time. Last, the three services observed pursued a strategy of focalization on a specific need. Conclusions: The findings provide insights for policy makers and hospital managers on how to set effective services and avoid service abandon, thus reducing waste of resources, and on how to motivate the professionals and patients, by increasing the chances of duration of the services over time. & 2017 Fellowship of Postgraduate Medicine. Published by Elsevier Ltd. All rights reserved.

n Corresponding author. Postal address: Via Lambruschini 4/b, 20156 Milano, Italy. E-mail address: [email protected] (F. Segato).

http://dx.doi.org/10.1016/j.hlpt.2017.07.003 2211-8837/& 2017 Fellowship of Postgraduate Medicine. Published by Elsevier Ltd. All rights reserved.

Please cite this article as: Segato F, Masella C. Telemedicine services: How to make them last over time. Health Policy and Technology (2017), http://dx.doi.org/10.1016/j.hlpt.2017.07.003

2

F. Segato, C. Masella

Introduction In the last decades, the services enabled by Information Technologies have supported the paradigm shifts in service delivery running over healthcare settings worldwide [1,2]. This is evident if we consider applications such as telemedicine (i.e. the use of Information Technologies to support the delivery of healthcare from a distance [34]). On the one hand, policy makers of developed countries are currently enacting a shift from acute/hospital centered health to primary care centered solutions, which can better meet the needs of chronic patients [26]; therefore, telemedicine provides a chance to keep the patients monitored, even at home or in primary care organizations [27]. On the other hand, in developing countries, telemedicine can increase the access to specialized care, by limiting geographic dispersion and health disparities, thus enhancing the citizens’ quality of life [3–5]. Telemedicine services are implemented extensively in these contexts, even in a cross-country mode [6–8]. However, despite their undeniable potential, examples of telemedicine services, that were implemented but failed to last over time, are countless in practice and well documented in literature [1,8,10–13,27]. In this regard, we argue that a lot has been said on how to implement these services, that is how to put them into practice within healthcare organizations [14,27,31,33– 35]. In contrast, few studies have been conducted on how to enhance the possibility of telemedicine services to last over time, once they are implemented [15,36]. According to recent reviews, to support the implementation of telemedicine services within healthcare organizations, policy makers and healthcare managers should provide appropriate legislative framework and financing, and leverage on the stakeholders’ acceptance, on technology reliability, and on organizational factors, such as process, structure, culture, management [11,14]. However, limited research has been conducted on them until now, and in particular on "if" and "how" these factors affect the long-term duration of the services, once they are implemented. We argue that furthering knowledge on how to support the duration over time of telemedicine services (and ITenabled services more broadly) is crucial for policy makers and managers in the healthcare setting. From such knowledge, they can get useful insights to set effective services and avoid service abandon thus reducing waste of resources, and to motivate the actors, by increasing the chances of success of the services over time [16,36]. Therefore, to further inform the discussion on this issue, we conducted an in-depth exploratory study, guided by the following research questions: 1) how do the factors, which support the implementation of telemedicine services, affect their duration over time?; 2) are there further factors to be considered, to improve the chances of long-term duration of telemedicine services? The paper is structured as follows: in the methods, we report the rationale for the definition of the theoretical framework and for the case selection, and we provide details on the data collection and analysis. Then, we report the findings about these two research questions, and we

discuss them, considering also the limitations of our study. Last, we push forward the implications of the study for policy makers and healthcare managers, together with new considerations for future research.

Methods The theoretical framework The primary goals of our study were to understand how the factors that support the implementation of telemedicine services within healthcare organizations affect their duration over time and, in this respect, to investigate if further factors should be considered. With this purpose, first, we conducted a literature review, to identify contributions about the factors supporting the implementation of telemedicine services within organizations, and we collected them in a research framework. Subsequently, we proceeded with the empirical part of the study, by investigating how and if these factors affected not only the implementation, but also the duration of the services over time. The literature review was conducted in two steps. First, we performed a search in PubMed (keywords: “telemedicine” AND “implementation” AND “success”; “telemedicine” AND “implementation” AND “sustainability”; “telemedicine” AND “implementation” AND “duration”; “telemedicine” AND “implementation” AND “organization”). Second, we performed a search in those journals whose denomination specifically refers to “telemedicine” (i.e. Journal of Telemedicine and Telecare; Telemedicine and e-Health), with the same keywords. Most of the contributions we retrieved from this second search had already emerged at the first step of the review. The contributions in scope are listed in the last column of Table 1, which represents the theoretical framework of our study. The dimensions of the framework (technology, acceptance, organization, financing, and policy and legislation) were adapted from contributions by Broens et al. [11] and Brebner et al., [12], while the sub-dimensions of the factor “organization” (i.e. process, structure, culture, and management) were adapted from contribution by Rassmussen et al. [14]. We grounded the framework on these three contributions because they provide rigorous reviews of factors emerging from both theory and practice; moreover, they are published on relevant Journals with up-to-date debates on telemedicine applications. The remaining part of the contributions in scope, emerging from the literature review, were used to better define the factors and to inform the data analysis.

Case selection The goal of this study is to explore how to make telemedicine services last over time. Consistent with the need to perform and indepth investigation of processes and organizational dynamics over time, a longitudinal case study was selected as the appropriate methodology to carry out the research [17,18]. Starting from the research questions and from the theoretical framework, we selected the cases through a theoretical sampling strategy [39]. First, we screened the telemedicine services implemented in Italy between 2008 and 2010. This screening was done by accessing a database created by the Italian Ministry of Health. Specifically, the database contained information about services activated for patient affected by Chronic Obstructive Pulmonary Diseases (COPD), which is a disease with worldwide critical levels of incidence and prevalence [42]. The aim of the database was to collect information on services for patients affected by this specific disease, to identify an effective service model and eventually to transfer it to services for patients affected by different chronic diseases.

Please cite this article as: Segato F, Masella C. Telemedicine services: How to make them last over time. Health Policy and Technology (2017), http://dx.doi.org/10.1016/j.hlpt.2017.07.003

Telemedicine services lasting over time

3

Table 1 Theoretical framework: factors fostering the implementation of telemedicine services. Source: factors adapted from Broens et al., Brebner et al., Rasmussen et al. [11, 12, 14] Factor

Description

Further references

Technology

     

[12,31,34,28,32,21,30]

Acceptance

Organization



   Financing

 

Policy and legislation



Support for installing; Ongoing maintenance; Training on use for patients and professionals; Usability and reliability. Commitment of patients and healthcare professionals; Sustained by: perceived practical benefits, evidence-based effectiveness of the service. Process: new work routines; task and responsibility shifting from physicians to nurses; new guidelines; training on new clinical practices (for nurses especially) and on the use of technologies; enhanced collaboration between acute care and primary care staff; Structure: new workflow, new communication platforms; Culture: getting used to a new way of communicating; Management: setting incentives for cooperation among hospital and primary care structures; keeping each level informed. Availability of funds for purchasing devices, for their maintenance and for sustaining the operational costs of the organizations who implements the service; Definition of an exit-strategy, i.e. how to sustain the costs of the service once the initial funding expires. Regulatory framework for the service: ○ Clear reimbursement for the professionals involved; ○ Adequate incentives to promote implementation and sustain implementation over time; ○ Patients’ safety and privacy.

The initial screening was performed in 2010. We found 9 telemedicine services in the database and we carried out a phone survey with the contact points indicated for each service, to collect more information. We discovered that, out of 9 services, 5 were implemented by single professionals, within medical offices and nursing homes. Different technology providers contacted these professionals to offer telemedicine devices for a free test of technology. Therefore, these services were temporary projects, which were neither included in the organizational processes and practices, nor financed or sustained at the management or policy level. Moreover, in these 5 settings, we could not explore any implication about the choice of technologies (which was not chosen, but imposed by the provider) or about strategies for financing, since the costs for technologies were covered by the providers. Therefore, these 5 cases were not suitable to investigate the dimensions included in the framework and we excluded them from the sample. In 2012, we surveyed the informants of the 4 remaining cases again, to investigate if the services were still in place. When the survey was repeated again in 2015, only 3 out of 4 were still ongoing, since one of the services ended after the director of the hospital unit retired in 2013. The three services still ongoing in 2015 represent the sample of this study.

Data collection and analysis Data collection relied primarily on face-to-face interviews with key informants [18]: Local Health Agency (i.e. Primary Care Structures in Italy) managers, hospital managers, physicians, nurses and home care assistants (8 to 10 informants for each case). We could not interview patients and informal caregivers, as their names could not be disclosed for privacy constraints. However, we attempted to minimize this potential bias by

[12,31,21,23]

[12,33,34,29,35,36,32,21,23,30,24]

[12,35,36,28,30]

[31,36,28,21,23,30,24]

considering secondary sources such as patient satisfaction survey reports and patient complaint reports, where available. To enhance the reliability of the results we defined an interview protocol [17,18], based on semi-structured questions aimed to gather knowledge about the specific service offered to the patients and to engage key informants in the discussion about five factors included in the theoretical framework (i.e. technology, acceptance, organization, financing, and policy and legislation). During the interviews, we collected information about the organization of the services; then, we explicitly invited the key informants to discuss “if” and “how” the five factors included in the framework (see Table 1) were perceived as relevant for the service duration over time. Eventually, through the interviews, we collected evidence on factors not included in the framework, which supported the long-term duration of the services. The same strategy was replicated in the three cases, to enhance external validity of the results [18]. After a first round of interviews in 2010, we interviewed the informants 2 years later (2012), and then 3 years later (2015). The interviews at the first stage lasted, on average, one hour and a half, while they were shorter (about 45 minutes each) in the following years. Each interview was recorded, transcribed verbatim and validated by the informants, to enhance the construct validity [18]. The data gathered through the interviews was complemented and triangulated with secondary sources. Specifically, we relied on the following sources of information: internal audit reports and patient and healthcare professional satisfaction survey reports (where available), documentation about best practices in telemedicine for pulmonary diseases, service protocols and guidelines. Finally, we gathered additional information through the direct observation of daily activities, by spending time in the hospital units and shadowing professionals in their daily activities, before and after performing the interviews.

Please cite this article as: Segato F, Masella C. Telemedicine services: How to make them last over time. Health Policy and Technology (2017), http://dx.doi.org/10.1016/j.hlpt.2017.07.003

4

F. Segato, C. Masella

The data analysis consisted of data transcription, validation and coding, by drafting schemes and tables that facilitated the examination of the cases [18]. The main purpose of the analysis was to find evidence if the factors included in the theoretical framework proved to play a role in the long-term duration of the telemedicine services, and to investigate how these factors (or other possible factors emerging from the interviews) shaped the long-term duration of the services implemented in the three cases. Therefore, the coding process was repeated; in a first round, codes were derived deductively from the literature, then inductively from the interviews’ transcriptions and field notes collected during the observations [19,20].

Description of the cases A description of activation and implementation processes of the three cases is provided below. Table 2 shows further details on the number of patients enrolled, and on how the patients were selected and managed in the three services. The telemedicine service implemented in case 1 has been activated in a pulmonary unit of a medium-size hospital (about 1.000 beds for hospitalization) in Cremona (Northern Italy). The service started in 2006, to answer to address the specific needs of patients with COPD at an advanced stage with disabling comorbidities (e.g. Amyotrophic Lateral Sclerosis). Before the service activation, the only way to manage these patients was to keep them hospitalized. However, these long-lasting hospitalizations greatly increased the costs of care, and they were a heavy social burden for the patients’ families. Telemedicine was a smart solution, as it enabled the creation of a protected environment at the patients’ home, so that they could be de-hospitalized, while physicians kept control over the clinical situation. The head of the pulmonary unit fully endorsed the implementation of the service. The physicians and nurses involved in the service were helped since the beginning to understand how the service could increase the effectiveness of the care provided to these patients; the open endorsement of the head of the hospital unit further spurred them toward the service acceptance. The service was initially offered to 20 patients living close to the hospital neighborhoods; the number of patients did not increase over the years, since the hospital staff was unable to bear the monitoring, surveillance and consultations for a higher number of patients. To finance the activation of the service, the head of the pulmonary unit used, for this purpose, the relation with an International Service Club, which collects and invests financial resources to support services of public utility. The mission of the Club for 2006 was to finance a service within the hospital, and the head of the pulmonary unit was consulted to see if there was any unmet need. Therefore, the Club financed the initial purchase of the devices and provided a small sum as partial reimbursement for the professionals involved. Some problems about the reimbursements and fees for professionals were already in place at activation (e.g. initially nurses were working for free), but were soon solved, since the service received further support from the most famous association of patients affected by amyotrophic lateral sclerosis and from the Local Health Authority (LHA). In 2008, to sustain the service after the financing from the Club expired, the hospital managers, in collaboration with the LHA, included the purchasing of telemedicine devices as a requirement for the providers participating to the call for tender for the hospital oxygen supply. The telemedicine service implemented in case 2 has been activated in a pulmonary unit of a small-size hospital (380 beds for hospitalization) in Verona (North-Eastern Italy). The service started in 1988, to respond to a specific need related to the lack of resources available for patients affected by chronic respiratory diseases. Before the service activation, since the hospitalization was extremely expensive, the patients were discharged quickly, but

they needed assistance at home. However, the dispersion of the patients living in the hospital neighborhood made it very hard for practitioners to perform a satisfying number of domiciliary visits to the patients in need. In the words of the head of the pulmonary unit: “we needed to put telemedicine into practice. Some practitioners working in other units of the hospital boasted about how good they were in taking care of the patients: they performed visits at their homes once every 3 months. Is this really “taking care of” a patient? What about the rest of time, when the patient is left alone? In my view, they were facing just the 0,2% of the problem in this way! To really take care of this population we needed telemedicine, and we needed it fast!”. The physicians and nurses, spurred by the head of unit, soon realized that telemedicine was the only way to effectively assist patients, as it allowed to keep control over a wide number of patients simultaneously, without moving from the hospital ward, unless in case of emergency. Nurses play a dominant role in this service. This aspect is crucial for its sustainability, given that in 2010 there were about 240 enrolled patients and the number grew further in the years: the physicians alone would not be able to cope with this wide number of patients. Moreover, the competences acquired through the service is a source of great satisfaction for all the professionals involved. The service started with unstable financing conditions. Initially, the Regional Government and the Italian National Center for Research gave a small initial funding. Despite the devices were very costly at that time and the financial amount was not sufficient to buy a consistent number of devices, the service was activated. In 1995, under the persistent requests of the head of the pulmonary unit and encouraged by the positive outcomes of the services, the Regional Government activated a specific Diagnosis Related Group for reimbursing the service as an intervention of Home Hospitalization. Within this specific frame, the services offered by the pulmonary unit received a stable financing. A consistent number of publications let the service have a resonance in the scientific community; this telemedicine service has been taken as a model for other services activated in Italy (e.g. the service reported in case 3 has started based on the encouraging results of this one). The telemedicine implemented in case 3 has been activated by the staff of a pulmonary unit of a small-size hospital (350 beds for hospitalization) in Rieti (Central Italy). The service started in 2006, when the head of the pulmonary faced the fact that the resources available at the hospital were not sufficient to face the increasing need for care expressed by the population in the area. In fact, before the service activation, the late diagnoses, the requests for hospitalizations, and the number of emergency interventions were increasing at a dramatic rate and the resources of the hospital unit were not sufficient to support prevention, diagnosis and care. Since some esteemed colleagues (i.e. the professionals involved in case 2) were using telemedicine devices with very good results, the head of the pulmonary unit strongly encouraged the staff to work as a “test site” for a provider of telemedicine devices. At the beginning, some concerns were raised about the fact that the telephone line is not always reliable in the area surrounding the hospital, but the devices have a feature, which, in case of a line black out, allows the patients to save the record and send it later, when the telephone line is restored. The professionals (nurses especially) expressed enthusiasm for the service, which makes it possible to keep the patients out of the hospital, but simultaneously safeguarding a personal relationship with them. Moreover, the service let the nurses learn new skills. The main criticality emerged in 2015 was that the need for remote monitoring expressed by the population could not be meet entirely. The number of patients enrolled in the service was expected to increase (60 patients was the target set in 2010), but the number of patients was then kept constant in time (31 patient enrolled in 2015) due to lack of resources (especially nurses). Regarding financing, the service was activated when a private

Please cite this article as: Segato F, Masella C. Telemedicine services: How to make them last over time. Health Policy and Technology (2017), http://dx.doi.org/10.1016/j.hlpt.2017.07.003

Cases Overview. Activation

Actors involved in service delivery

Overview of the service

Case 1

About 20 patients affected by COPD at a very advanced stage with severe co-morbidities (e.g. Amyotrophic Lateral Sclerosis) and living close to the hospital (30 km max).

2006

Physicians, nurses, domiciliary assistance staff, emergency unit.

Case 2

About 240 patients affected by COPD at medium to severe stage requiring oxygen therapy and living in the hospital neighborhoods.

1988

Physicians, nurses, general practitioners (GPs), psychologists, emergency unit.

Case 3

About 30 patients affected by COPD at medium to severe stage requiring oxygen therapy and living within the hospital neighborhoods.

2006

Physicians, nurses, domiciliary assistance staff, emergency unit.

Type of service(s)n: tele-monitoring; tele-consultation; home tele-nursing; telesurveillance.Description: Five physicians manage the service, with the help of two nurses of the hospital unit, and the domiciliary assistance staff (which is hired by the local health authority). The staff of the hospital unit selects the eligible patients. According to a protocol shared by the unit with the hospital managers, the patients are enrolled and the caregivers undergo intensive training sessions.Each patient is assigned to a specific physician, who is in charge of monitoring his/her clinical conditions. Once the service is activated, devices record some clinical parameters of the patients enrolled (e.g. blood pressure, pulse and saturation) during the night or at specific time during the day. The record outline is automatically sent to a specific server and collected. At predefined deadlines, the hospital physician reads the record and sends a medical report back to the patient through e-mail.The nurses of the Domiciliary Assistance performs routine interventions at the patients’ home and, by having frequent access to the patients’ homes, they understand the care needs and activate eventual additional remote surveillance or home visits from different hospital specialists if needed.A direct contact with the hospital emergency unit is active. Type of service(s)n: tele monitoring; tele-consultation; tele-surveillance.Description: The patients selected for enrollment, receive some devices at home, which allow them to record the main vital parameters (blood pressure, pulse, saturation) at some predefined deadlines.At the hospital unit, three computers are always active and receive the records from the patients. Daily, the nurses read the records and evaluate the patients’ health conditions. If they detect any parameter out of normal, the nurses either call the patients at the domicile or ask the clinicians to do it, in case they suspect that something severe is happening. During the phone call, they evaluate the best intervention to be undertaken (emergency included).In some cases, a visit at home is enough to solve the problem; if so, the specialist contacts the GP who performs the intervention. If something severe occurs, an alarm is set on the program: in case some specific parameters (different for each patient) exceed predefined thresholds, the system activates a loud ringing noise. In this case, a nurse or a practitioner logs in on the available computer, check the situation, call the patient and fix the problem or call the emergency service. Out of the activity hours of the hospital unit, the alarm is not active, but the caregivers are accurately trained and, in case of specific exacerbations, call the emergency unit. In 2012, two psychologists were hired in the pulmonary unit to support the relationship with the patients and the caregivers. Type of service(s)n: tele monitoring; tele-consultation; tele-education.Description: According to a shared protocol, patients can be enrolled in the program on a GP's or hospital specialist's request. At the enrollment, nurses train the patients and the caregivers about the use of technologies, they plan the frequency and type of data recording. Supported by a software that receives the data from the devices at the

5

Patients (N. of pts enrolled in 2015 & type)

Telemedicine services lasting over time

Please cite this article as: Segato F, Masella C. Telemedicine services: How to make them last over time. Health Policy and Technology (2017), http://dx.doi.org/10.1016/j.hlpt.2017.07.003

Table 2

Tele-monitoring = timely transmission of data from home to the hospital unit; home tele-nursing = nurses go the patient's house to monitor the clinical status and send data to the hospital unit; tele-consultation = physicians or nurses provide feedbacks to patients seeking for help; tele-surveillance = the system detects records exceeding predefined thresholds and connects to the emergency service in case of need; tele-education = the staff provides education through devices at distance [29,37]

n

Actors involved in service delivery Activation Patients (N. of pts enrolled in 2015 & type)

Table 2 (continued )

patients’ home, the nurses control the patients’ record; once the nurse receives the record, she contacts the patient/caregiver at the domicile to give feedbacks. If something wrong emerges from the record, the nurse considers if a specific intervention is needed and she calls the specialist, the GP or even the emergency service accordingly. If the patients’ conditions are under control, the nurse calls the patients to give positive feedbacks.From 2012 on, while talking to the patients at the phone after analyzing the record, the nurses provide also basics of health education, following a formal predefined checklist. Last, since most of the patients enrolled in the service presented several co-morbidities, mostly related to poor cardio-vascular conditions, for some patients the service was extended to include also the cardiology unit, with a nurse monitoring the patients’ cardio-vascular conditions in remote.

F. Segato, C. Masella

Overview of the service

6

Company won the call for tender for oxygen supply in 2010. Together with oxygen supply, this company offered some telemedicine devices. In 2015, the service still relied on the same source of financing.

Results The results emerging from the cases are organized according to the two research questions. First, we provide evidence on how the factors included in the framework reported in Table 1 affected the duration of the services over time (research question 1). Second, we report further factors that emerged from the cases as relevant to improve the chances of long-term duration of telemedicine services (research question 2).

Framework analysis (research question 1) We organized the results emerging from the cases according to the elements of the framework in Table 1. Technology With respect to technology, the interviewees reported that most providers purchase devices with similar functions, a good level of usability, and high reliability, although some issues still exist. For example, in case 3, at the first provision of services some issues emerged due to the poor quality of the telephone line available in the mountain area. However, the provider of the technology soon solved the issue by supplying devices that allowed to record the patients’ data and send it later if the first data transmission failed due to a poor connection. Last, in the three cases, the standard contracts with providers of technologies agree on further provision of periodical maintenance and training. Concerning the patients, the interviewees reported that the acceptance of technology-mediated interactions is a critical issue when a new patient is enrolled in the service, as the patients (and their caregivers), initially, do not have trust in the devices. In case 3, a nurse reported that a relative had sabotaged the patients’ record on purpose, just to check if the nurse noticed the malfunctioning and intervened accordingly. However, as time passes by, the patients tend to rely more on technologies. Furthermore, given that technological devices are now pervasive in several ordinary activities of daily life, the patients tend to be less reluctant to receive technology-mediated care. Moreover, the professionals (especially nurses) are keen to use technological devices more than in the past and their confidence, together with the participation to structured training programs, boosts the patients and caregivers’ confidence in technologies. Acceptance The contributions in the literature agree upon the fact that acceptance from both patients and professionals tends to increase when telemedicine services prove to meet their needs [7,22,23]. This statement might seem trivial, but it is not in practice: too often, it is not clear to what extent telemedicine services are intended to meet a specific care need, rather than being “solutions in search of a problem” [8,12,21]. For professionals involved in these cases, acceptance was spurred by the presence of a person (the head of the pulmonary unit of the three cases), who strongly endorsed the service implementation. In the extant literature, some authors had referred to these figures as the “service champion” [15,24], i.e. the person who “believes in the cause”, as he/she perceives the value and benefits that might be gained from the application of telemedicine within routine practice. In addition, the fact that the service champions in the three cases were the head of the pulmonary units conferred legitimacy to the introduction of the

Please cite this article as: Segato F, Masella C. Telemedicine services: How to make them last over time. Health Policy and Technology (2017), http://dx.doi.org/10.1016/j.hlpt.2017.07.003

Telemedicine services lasting over time telemedicine service and made it easier to gather initial commitment from the professionals. A nurse in case 3 said: “she [the head of the pulmonary unit] was convinced that the service was exactly what we needed. I wasn’t aware at that time, but I trusted my chief…and…well, now I can tell that she was right!”. Yet, the head of the pulmonary unit in case 2 said: “People are the key. To get their commitment I took those nurses and specialists that work better in my unit, those that I see having a deep sense of their practice, and told them: ‘you see you are doing a lot of useless, entropic activities. With telemedicine, we are going to limit all of that… and they started to understand the value of this service”. In the three cases, the champion insisted on a specific need of the professionals that telemedicine might target (e.g. saving time on useless activities by keeping out of the unit those patients that do not actually require an hospitalization): this proved to be crucial for the initial implementation and the following normalization of the services into the clinical practice. Moreover, the cases showed that telemedicine is accepted and endorsed if the staff perceives its implementation as an opportunity to get new abilities and skills. In many cases, the professionals (especially nurses) argued that being part of the service was source of great satisfaction and this works as a durable motivating factor. About the patients, the cases showed that their acceptance is enhanced if the services target a specific need that they perceive as crucial for their personal wellbeing. In this regard, a nurse in case 2 reported a negative episode. The unit had a device available for use: “we did not have pending requests for the devices, so we decided to give it to a patient affected by a COPD at a moderate stage of the disease who, by the way, did not have a driving license and, as such, might have problems in reaching the hospital. We asked him to use the device once a week, although maybe he was not feeling that bad. As a result, the patient went home, put the device in a drawer and never used it! […] the point was that he did not perceive the urgent need to monitor his clinical conditions”.

Organization Regarding the framework dimensions of process, structure and culture, the three services had activated new work routines, formal guidelines (shared between primary and secondary care organizations), formal training on technologies, and new practices and protocols. Since all these factors were present in the cases, as well as in the theoretical framework, we cannot draw significant implication on if and how they shape the long-term sustainability of the services; rather, we argue that they appear to be necessary conditions for the service to last over time. About management, we gathered insights on this factor both at hospital unit level and hospital level. At the hospital unit, we found that a significant factor, emerged from the cases, is how to manage the service activation, which appeared to increase the chances of survival of the service over time: the selective activation of the staff. This element is common to the three services, although it was made according to different criteria. In fact, in case 2 and 3, the selection was done according to the trust that the champion has in the professionals he activates (“I took those nurses and specialists that work better in my unit, those that I see having a deep sense of their practice”, case 2).In case 1, the champion selected some nurses whose contract left them free from afternoon and night shifts and, for this reason, they had time to spend on the service. Regardless of the criteria guiding the selection, the selective activation of the staff creates the necessary consensus which allows, on the one hand, to introduce telemedicine within the routine clinical practice of the units more easily and, on the other hand, to sustain the service when the champion retires. We found that hospital managers were positive about the services, but they did not play a crucial role in supporting them and in fostering cooperation between the hospital unit and the primary care structures. In fact, in these three cases, the hospital

7 managers were basically interested in the fact that the services were effective from a clinical point of view and – especially - selfsustainable from a financial point of view. In this respect, the Hospital Manager, in case 3, argued that: “I believe that the telemedicine service activated in the pulmonary unit respects three crucial requirements: first, it is in line with the needs of the population we serve; second, it fosters a more efficient use of the resources and therefore it releases critical factors that we can use differently; third, it allows to treat the patient at home, which is a positive aspect for the patients, their families and the hospital itself”. Financing About financing, we found that telemedicine services initially rely on spot financing, from private (e.g. case 1) or public (e.g. case 2, with the National Center for Research) sponsors [4]. However, in order to be normalized in the organization, the telemedicine service requires an institutionalization from the hospital management and/or the regional or national authorities. In case 1, the financing was institutionalized because the service became “too important to be stopped”. The head of the pulmonary unit stated that: “In a very short time, we made this service so crucial for the patients and their families that it actually became irrevocable. I challenge the one who has the courage to stop it!”. Therefore no specific exit strategy was planned, but the service became so essential, that, by now, it is not possible to stop it without causing difficulties to the patients, their families and the hospital, who should re-hospitalize the patients who currently receive the service at home. This fact has forced the hospital managers to find ways to support the service over time. In case 2, after 5 years of implementation, the staff of the pulmonary unit had published studies showing that the patients enrolled in the service saved 2.7 hospitalizations per year on average, if compared with similar patients followed in usual care. Part of the savings allowed by the telemedicine service were invested to finance a service of home hospitalization. Within this framework, the service found stable financing and could be institutionalized. Financing, in case 3, remained connected to the call for tenders of oxygen supply for the hospital. The fact of being dependent of a private supplier, potentially exposed the service to stop, in case the provider that wins the call for tender does not include telemedicine devices in the offer. In 2010, the head of the pulmonary unit express worries about this fact. Actually, in 2015, this was not source of concern any more, since providing telemedicine devices has been included as a prerequisite in the call for tender for oxygen supply for the hospital, and devices for remote monitoring offered by the main Companies are comparable in terms of functions and usability. Therefore, disruptions in the service are not likely to occur. Policy and legislation Concerning policy and legislation, the three telemedicine services in the cases required a top-down action of the middle management (i.e. the head of the hospital unit) to support the operationalization within the organization: in the three cases, the champion initially “forced” the staff to use the devices. Simultaneously, the services were proposed to the top management/policy level in a bottom-up thrust: the head of the hospital units promoted the service with the hospital board or the LHA managers. Interviewees reported that the other way around (i.e. professionals undertaking the service for a policy makers’ top-down mandate) is unlikely to result in a longlasting service, as the professionals would lose the motivation to proceed. In this regard, despite the professionals’ claim that financial incentives are not enough to motivate the professionals, an institutional frame is needed when the champion leaves the service. In case 2, the head of the pulmonary unit retired in 2013

Please cite this article as: Segato F, Masella C. Telemedicine services: How to make them last over time. Health Policy and Technology (2017), http://dx.doi.org/10.1016/j.hlpt.2017.07.003

8

F. Segato, C. Masella

and a physician stated that: “having an institutional framework [i.e. the activation of the service of home hospitalization with a specific reimbursement] justifying the efforts made for the service with specific rewards, helped the service to survive”.

Further factors emerging from the cases (research question 2) Three elements emerged from the cases, which were not included in the theoretical framework. First, many studies stress the importance for telemedicine service to scale-up, by going from a localized project, to a wider and diffused application [38]. Actually, our cases show that the duration over time for a telemedicine service is not related necessarily to a growth in the dimensions and spread of the service [11]; rather, it is about the focus of the service on a specific need, by targeting it with the available resources (i.e. staff and funding) [23]. We do not argue that going for a “small service” is better that enrolling a wider number of patients; rather, we argue that the service size depends on the goal that the service intends to reach and on the resources available. In case 1, for example, being focused on a small number of patients (about 20) increased the possibility to make the service last over time: if the number of patients enrolled would grow, the resources (e.g. physicians, nurses) for training the caregivers and monitoring patients with such complex needs at distance would not be sufficient. Therefore, the goal of scaling-up the size of telemedicine services could be in contrast with the need to keep a high intensity of monitoring over the patients enrolled, with the available resources. Second, the right attention should be put on considering that not all medical needs can be adequately targeted by telemedicine; a careful consideration of the need that a telemedicine service should target is crucial to enhance the long-term duration of the service itself. The three cases reported here aim to monitor patients with a specific chronic condition (i.e. COPD at a severe and very severe stage) and overall stable conditions (for compromised as they can be, as in case 1). Telemedicine might not be equally effective for more specialized/acute care, where treatments should be continuously customized and tailored through multidisciplinary actions as the disease progresses [25]. As stated by a nurse in case 3: “We cannot follow them all [all the patients affected by COPD at any stage]: we have selected a specific sample of conditions which require monitoring. Without this accurate selection, the service is not sustainable…and the patients are not going to use it [the service] if they don’t see that it's useful for them!”. Third, most of the contributions in the literature refer to telemedicine as a “service” or a system standing alone. In the three cases, telemedicine is the tool that enables the monitoring of the patients’ clinical conditions at distance, which is just one feature of the service, together with empowerment of patients and caregivers, health education, home assistance, psychological support. With words of the interviewees: “Telemedicine is just one petal of the flower” (Case 1); “without the service all around, telemedicine is an empty, useless box” (Case 2); “the people make the service: telemedicine is just one of the tools we use, just as a telephone, an oximeter…” (Case 3).

Discussion The theoretical framework reported in Table 1 shows the factors that, according to the extant research, support the implementation of telemedicine services within organizations [11,12,14,28–36]. The aim of this study was to investigate how these factors affect the duration of services over time and to report eventual further factors that

improve the chances of long-term duration of the services. Table 3 provides an overview of the findings against the existing body of literature. About technology, our results give further insights on the importance of the device usability and reliability. However, although the devices produced and purchased by most providers are smart and safe now, the cases show that there is still an initial need to enhance the users’ trust in technology. In this respect, the role of the nurses in educating and supporting the patients as well as their caregivers in the use of devices is crucial, together with timely responses by the providers in case of technology breakdown. These elements increase the patients’ perception of being safe and thus their engagement with the service over time, with a secondary effect: by being responsible for the training of patients to the use of devices and for their monitoring over time, nurses acquire new competences and expand their role. Overall, the nurses enriched their practices and skills, and this was a source of personal satisfaction, which increases their acceptance and support for the service over time. Physicians, in turn, are positive about giving leeway to other professionals on activities such as the periodical monitoring of stabilized patients, since this fact allows them to focus on those patients that really need medical assistance. In the three cases observed, the presence of a “champion”, who had leeway to activate the staff and to force the setting of specific processes within the organizations, facilitated the activation of the services [24].The champion also conferred legitimacy to the services, both inside the hospital unit, and at hospital / policy level. In the three cases, the bottom-up actions undertaken by the champions, to raise the managers’ awareness, were particularly relevant, also because for some diseases, such as COPD, the evidence about the benefits of telemedicine with respect to usual care is still limited [41]. However, with respect to the duration of the services over time, some might argue that much more evidence on the effectiveness of telemedicine devices is available now, compared to the time in which the three services were activated. As such, the presence of a champion might not be relevant for the duration of the service now as it was a few decades ago, when telemedicine was still at nascent stage. Nevertheless, the duration of three services over time relied on the fact that the conditions for the organizational and financial stability of the services were set at policy, hospital and unit level, before the champions left the units. Specifically, at the unit level, the organizational stability was assured through the enhancement of the role of the nurses, which was crucial to boost their acceptance (since they felt enriched and esteemed), to enhance the practitioners’ acceptance (since they could focus on more value adding medical activities), and to increase the patients’ acceptance (since nurses made them feel safer). At the hospital and policy levels, the services found support when they brought evidence of benefits for the patients and the hospital, by also respecting the necessary safety and privacy requirements. These conditions helped the services to get a specific reimbursement, which facilitated their financial stability and therefore their duration over time. In case 2, the savings induced by the service were used to finance a specific service (the

Please cite this article as: Segato F, Masella C. Telemedicine services: How to make them last over time. Health Policy and Technology (2017), http://dx.doi.org/10.1016/j.hlpt.2017.07.003

Telemedicine services lasting over time home hospitalization). In contrast, Case 3 was financed at the hospital level, through the definition of specific requirements for the call for tender for the hospital oxygen supply, which force the potential providers to include telemedicine services in their offer. The longitudinal investigation of the three cases also brought additional insights into which conditions foster the long-term duration of the services. First, the three services pursued the focus of the service on a specific need, rather than a strategy of scaling-up. This strategy might seem in contrast with the need to save costs, since small-scale services are not likely to provide remarkable savings for the healthcare system. However, the diffused scarcity of resources for the healthcare systems challenges the possibility of an extensive diffusion of telemedicine services; in this context, small, focalized and self-sustaining services could be a viable solution for telemedicine to take advantage of its potential. Second, the services target chronic patients with overall stable conditions; caution should be put on considering that not all clinical needs can be adequately met through telemedicine and that targeting ambitious needs that exceed its capacity is detrimental for the duration of the service over time [14]. Last, in the three cases telemedicine devices were used as tools part of a wider service, which encompassed training activities, empowerment, home assistance and psychological support.

Limitations We are aware that our study has some limitations. Among these, coherently with the aim of our study, we purposefully selected three services that lasted over time over a sample of telemedicine services for patients affected by COPD in Italy. We are confident that these cases respected the requirements for theoretical sampling in case study research [39], they assured replicability and the process of interest was transparently observable [17]. However, due to the purpose of our sampling strategy, we came out with cases that present limited internal variation, since the services lasting more than 10 years are all hospital-based and implemented in pulmonary units. This fact could compromise the generalizability of our findings. Table 3

9 Moreover, we focused on three services implemented in Italy. Although we found that this choice is common in studies about telemedicine services, since it allows to avoid the bias unintended variance due to context-specific features (as suggested by Ohinmaa, 2006 [40]), we are aware of the limitations imposed by this choice on the possibility to transfer the findings in different contexts.

Conclusions The primary goal of our study was to understand how the factors that support the implementation of telemedicine services within healthcare organizations, affect their duration over time, and to investigate if further factors should be considered in this respect. Due to a lack of research on this topic, we conducted an exploratory in-depth longitudinal study on three cases of telemedicine services lasting 10 years and more and we gave the interpretation of the results in the light of the theoretical framework, informed by the extant literature on factors fostering the implementation of telemedicine services within organizations. We found that to enhance the duration of the service over time, conditions for financial and organizational stability should be set before the champion leaves the service. Financial stability was reached through different strategies in the cases, with the involvement of the hospital managers (case 1 and 3) or the policy makers (case 2). About organizational stability, we found that providing opportunities to enrich the practice, or to focus on core medical practice by shifting less value-adding activities to other professionals, proved to enhance the nurses and physicians’ acceptance respectively; their acceptance, in turn, creates the conditions for the organizational stability of the service over time. Moreover, providing benefits perceived as crucial for their health, and receiving the nurses’ support on the use of technologies limits abandon and provides bottom up support from the patients. Furthermore, the duration over time of the three services was not assured by a scaling up in dimensions; rather, it was due to their focus on a specific need. Last, in the three services observed, the telemedicine devices were part of a wider service and not perceived as a stand-alone system.

Impact of the factors fostering the implementation on the duration over time.

Factor

The service lasts over time if…

Technology

     

Acceptance

   Financing Policy and legislation  Organization

…technologies are usable and reliable; …the nurses educate and support the patients. …the providers offer maintenance and assistance; …a champion supports the service activation; …the service addresses a urgent need for patients and professionals; …professionals and patients perceive the benefits they can get from the service (e.g. feeling safer; acquire new competences). …the champion enacts a selective activation of the staff. …conditions for organizational stability are set before the champion leaves the service. …financial stability is assured, when the champion leaves the service. …institutionalization is pushed through a bottom-up action (from professionals to hospital managers or policy makers).

Please cite this article as: Segato F, Masella C. Telemedicine services: How to make them last over time. Health Policy and Technology (2017), http://dx.doi.org/10.1016/j.hlpt.2017.07.003

10 Finally, the primary aim of our study was to interpret this phenomenon, of which we do not know much (i.e. how to make telemedicine services last over time), through an exploratory and qualitative research design [17,18]. When the choice fell on this specific research design, our goal was to provide insights for managers and a grounding for future research [43]. Hospital managers and policy makers who are interested in setting services that last over time can get initial insights and causes for reflection from our results. In addition, our findings could be transferred to different organizational or clinical settings, such as, for example telemedicine services delivered by inter-organization networks or services for multi-chronic patients.

Author Contribution. Federica Segato has designed and conducted the research, analyzed the data and wrote the parer drafts. Cristina Masella has supervised the reseach, discussed the results and revised the paper drafts.

Funding No funding was received for this study.

Conflict of interest No Conflict of interest to declare.

Acknowledgements The authors are thankful for the availability of the staff in the three case sites. They are especially thankful to Dr. Rita Le Donne (Rieti hospital), Dr. Giancarlo Bosio (Cremona Hospital), Dr. Roberto Walter Dal Negro.

References [1] Andreassen HK, Kjekshus LE, Tjora A. Survival of the project: a case study of ICT innovation in health care. Soc Sci Med 2015;132:62–9. [2] Alsulame K, Khalifa M, Househ M. EHealth status in Saudi Arabia: a review of literature. Health Policy and Technology 5 (2), (2016), 204-210. http://dx.doi.org/10.1016/j.hlpt.2016. 02.005. [3] Kamsu-Foguem B, Foguem C. Telemedicine and mobile health with integrative medicine in developing countries. Health Policy Technol 2014;3(4):264–71. [4] Carroll M, Horton MB. Telehealth and Indian healthcare: moving to Scale and sustainability. Telemed e-Health 2013;19 (5):377–9. [5] Nchise A, Boateng R, Mbarika V, Saiba E, Johnson O. The challenge of taking baby steps—preliminary insights into telemedicine adoption in Rwanda. Health Policy Technol 2012;1 (4):207–13. [6] Legido-Quigley H, Doering N, McKee M. Challenges facing teleradiology services across borders in the European Union: a qualitative study. Health Policy Technol 2014;3(3):160–6. [7] Doering N, Legido-Quigley H, Glinos IA, McKee M, Maarse H. A success-story in cross-border telemedicine in Europe: the use of intra-operative teleneuromonitoring during aorta surgery. Health Policy Technol 2013;2(1):4–9.

F. Segato, C. Masella [8] Saliba V, Legido-Quigley H, Hallik R, Aaviksoo A, Car J, McKee M. Telemedicine across borders: a systematic review of factors that hinder or support implementation. Int J Med Inform 2012;81(12):793–809. [10] Uscher-Pines L, Kahn JM. Barriers and facilitators to pediatric emergency telemedicine in the United States. Telemed eHealth 2014;20(11):990–6. [11] Broens TH, Vollenbroek-Hutten MM, Hermens HJ, van Halteren AT, Nieuwenhuis LJ. Determinants of successful telemedicine implementations: a literature study. J Telemed Telecare 2007;13 (6):303–9. [12] Brebner JA, Brebner EM, Ruddick-Bracken H. Experience-based guidelines for the implementation of telemedicine services. J Telemed Telecare 2005;11(suppl 1):3–5. [13] Hailey D, Jennett P. The need for economic evaluation of telemedicine to evolve: the experience in Alberta, Canada. Telemed J e-Health 2004;10(1):71–6. [14] Rasmussen BSB, Jensen LK, Froekjaer J, et al. A qualitative study of the key factors in implementing telemedical monitoring of diabetic foot ulcer. Int J Med Inform 2015;84:799–807. [15] MacFarlane A, Clerkin P, Murphy AW. Role flexibility among telemedicine service providers in the north-west and west of Ireland. J Telemed Telecare 2005;11(suppl 1):S62–4. [16] Hendy J, Barlow J. Adoption in practice: the relationship between managerial interpretations of evidence and the adoption of a healthcare innovation. Health Policy Technol 2013;2(4):216–21. [17] Eisenhardt KM. Building theory from case study research. Acad Manag Rev 1989;14(4):532–50. [18] Yin RK. Case study research: design and methods (Applied Social Research Methods). Fifth Edition Sage Ed; 2014. [19] Dowding DW, Turley M, Garrido T. Nurses' use of an integrated electronic health record: results of a case site analysis. Inform Health Soc Care 2015;40(4):345–61. [20] Miles MB, Huberman AM. An expanded sourcebook qualitative data analysis.Thousand Oaks. CA: Sage; 1994. [21] Jennett P, Yeo M, Pauls M, Graham J. Organizational readiness for telemedicine: implications for success and failure. J Telemed Telecare 2003;9(suppl 2):S27–30. [22] Doolittle GC, Cook DJ. Defining the needs of a telemedicine service. J Telemed Telecare 1999;12:276–84. [23] May C, Harrison R, MacFarlane A, Williams T, Mair F, Wallace P. Why do telemedicine systems fail to normalize as stable models of service delivery? J Telemed Telecare 2003;9(suppl 1):25–6. [24] Wade V, Eliott J. The role of the champion in telehealth service development: a qualitative analysis. J Telemed Telecare 2012;18(8):490–2. [25] Vitacca M, Mazzù M, Scalvini S. Socio-technical and organizational challenges to wider e-Health implementation. Chronic Respir Dis 2009;6(2):91–7. [26] Saltman R, Bankauskaite V, Vrangbaek K. Primary care in the driver's seat?: Organizational reform in European primary care, 51. UK: McGraw-Hill Education; 2005. [27] Zanaboni P, Wootton R. Adoption of telemedicine: from pilot stage to routine delivery. BMC Med Inform Decis Mak 2012;12 (1):1. [28] Adewale OS. An internet-based telemedicine system in Nigeria. Int J Inf Manag 2004;24(3):221–34. [29] Bartoli L, Zanaboni P, Masella C, Ursini N. Systematic review of telemedicine services for patients affected by chronic obstructive pulmonary disease (COPD). Telemed e-Health 2009;15 (9):877–83. [30] Doolittle GC, Spaulding RJ. Defining the needs of a telemedicine service. J Telemed Telecare 2006;12(6):276–84. [31] Gagnon MP, Duplantie J, Fortin JP, Landry R. Implementing telehealth to support medical practice in rural/remote regions: what are the conditions for success? Implement Sci 2006;1(1):1.

Please cite this article as: Segato F, Masella C. Telemedicine services: How to make them last over time. Health Policy and Technology (2017), http://dx.doi.org/10.1016/j.hlpt.2017.07.003

Telemedicine services lasting over time [32] MacFarlane A, Murphy AW, Clerkin P. Telemedicine services in the Republic of Ireland: an evolving policy context. Health Policy 2006;76(3):245–58. [33] Nicolini D. The work to make telemedicine work: a social and articulative view. Social Sci Med 2006;62(11):2754–67. [34] Obstfelder A, Engeseth KH, Wynn R. Characteristics of successfully implemented telemedical applications. Implement Sci 2007;2(1):1. [35] Whitten P, Holtz B, Nguyen L. Keys to a successful and sustainable telemedicine program. Int J Technol Assess Health Care 2010;26(02):211–6. [36] Zanaboni P, Lettieri E. Institutionalizing telemedicine applications: the challenge of legitimizing decision-making. J Med Internet Res 2011;13(3). [37] Jaana M, Paré G, Sicotte C. Home telemonitoring for respiratory conditions: a systematic review. Am J Manag Care 2009;15 (5):313–20.

11 [38] Zanaboni P, Wootton R. Adoption of telemedicine: from pilot stage to routine delivery. BMC Med Inform Decis Mak 2012;12 (1):1. [39] Glaser B, Strauss A. The discovery of grounded theory: strategies of qualitative research. London: Wiedenfeld and Nicholson; 1967. [40] Ohinmaa A. What lessons can be learned from telemedicine programmes in other countries? J Telemed Telecare 2006;12 (suppl 2):S40–4. [41] Bolton CE, Waters CS, Peirce S, Elwyn G. Insufficient evidence of benefit: a systematic review of home telemonitoring for COPD. J Eval Clin Pract 2011;17(6):1216–22. [42] Mannino DM, Buist AS. Global burden of COPD: risk factors, prevalence, and future trends. Lancet 2007;370(9589):765–73. [43] Blalock HM. Theory building and causal inferences. Methodol Soc Res 1968:155–98.

Please cite this article as: Segato F, Masella C. Telemedicine services: How to make them last over time. Health Policy and Technology (2017), http://dx.doi.org/10.1016/j.hlpt.2017.07.003