FACTORS INFLUENCING SUCCESSFUL ADOPTION OF eHEALTH
INNOVATIONS: THE STORY OF DIVERGING TWINS
MSc. BA Strategic Innovation Management Faculty of Economics and Business
University of Groningen Date: 18th of January, 2016
Supervisor: Dr. Pedro de Faria Co-assessor: Dr. Thijs Broekhuizen
DENNIS LINN Student number: 1889540
Johannes Vijghstraat 25 6524 BN Nijmegen, the Netherlands
E-mail: d.g.p.linn@gmail.com Word count: 12.884
Abstract: The potential of eHealth innovations to contribute to sustainable healthcare is
TABLE OF CONTENT
INTRODUCTION ... 3
THEORETICAL BACKGROUND ... 5
eHealth definition ... 5
Contributions and challenges of eHealth innovations ... 6
Factors influencing the adoption of eHealth innovations ... 7
Pre-implementation phase ... 7
Post-implementation phase ... 8
Summary of factors influencing the adoption of eHealth innovation ... 10
METHODOLOGY ... 12 Research design ... 12 Research quality ... 12 Research setting ... 14 Cases selection ... 14 Data collection ... 16 Data analysis ... 17 RESULTS ... 19
MijnZorgnet’s online community Vaatcentrum ... 19
MijnRadboud ... 22
Similarities and differences ... 26
DISCUSSION ... 30
Novel factors ... 30
Known factors ... 32
Adjusted table of factors influencing adoption of eHealth innovations ... 35
Managerial implications ... 36
Limitations and future research ... 36
CONCLUSION ... 37
REFERENCES ... 38
APPENDICIES ... 43
Appendix 1: Interview guide ... 43
INTRODUCTION
Research (e.g. in behavioural genetics) shows that identical twins are valuable for conducting nature-nurture based experiments (Bouchard & Propoping, 1993). Even though identical twins share the same genes, they diverge under the influence of environmental factors into different individuals. In other words, they share “nature” but differ in “nurture”. Under this analogy, the current study performs an in-depth case analysis of two (twin-like) electronic healthcare (eHealth) innovations, which share the same initiator and the same goal to create a Personal Health Record (shared “nature”). However, under the influence of environmental circumstances, these entities evidentially emerged to one successful, and one unsuccessful eHealth innovation (“diverged in nurture”). The current study aims to look at the (“environmental”) factors leading to the different success rate of these cases.
The general consensus about potential beneficial outcomes of eHealth innovations is overly positive, however empirical unsupported. Regardless of the lack of evidence, eHealth is widely seen by academia and policymakers as a ‘silver bullet’ for working towards a future of affordable healthcare (Mettler & Eurich, 2012). Unfortunately, eHealth innovations do not yet fulfil their potential (Black et al., 2011). The main reason for this is the relative low adoption rate ( Van Gemert-Pijnen, Nijland, et al., 2011).
Although, the current literature can already name a few factors influencing the adoption of eHealth innovations, it is still highly unclear what the whole scope of these factors is. Furthermore, the empirical evidence supporting and explaining how these factors relate to eHealth adoption is scarce (Black et al., 2011; Pagliari, 2007).
To address this gap in the literature, an in-depth case study approach is adopted in which two contrasting eHealth innovations, successful (MijnRadboud) versus unsuccessful (Vaatcentrum), are compared. Qualitative data, in forms of interviews and additional documents (e.g. internal documentation and personal notes) regarding the adoption of both innovations were gathered and analysed.
The research results shows that a co-developed vision, substitute clinical workload, and ease of use are important factors that positively influence the adoption of eHealth innovations. In addition, empirical evidence is provided, supporting the most important factors identified in the literature: (1) Business models (Van Limburg, Van Gemert-pijnen, Nijland, & Ossebaard, 2011), (2) user involvement (Van Gemert-Pijnen et al., 2011), patient empowerment (Aujoulat, d’Hoore, & Deccache, 2007), quality of care (Asoh & Rivers, 2010), cost reduction (Chaudhry et al., 2006), and safety & privacy (Black et al., 2011).
THEORETICAL BACKGROUND
This chapter will first define eHealth and then elaborate on the problems regarding the adoption of eHealth innovations in healthcare.
eHealth definition
Shortly after the commercialization of the Internet (1990s), the field of eHealth emerged. The number of publications mentioning eHealth increased since then steadily (see figure 1). However, no conclusive definition of eHealth has so far been agreed upon (Boogerd, Arts, Engelen, & van de Belt, 2015; Oh, Rizo, Enkin, & Jadad, 2005; Pagliari et al., 2005).
Figure 1. Number of papers
mentioned eHealth per year in Pubmed until fall 2014 (adopted from Boogerd et al., 2015)
Most definitions include the use of digital networks, information and communication technology (e.g. the Internet) in the delivery of healthcare services to patients. Additionally, the optimism regarding the postulated benefits is reflected in the eHealth definitions (Oh et al., 2005). In literature, eHealth is often affiliated with terms like benefits, improvements, enhancing, efficiency and enabling. In contrast, none of the definitions incorporate disadvantages or negative effects of eHealth (Pagliari et al., 2005). Despite the disagreement on the definition of eHealth, the term eHealth has been generally understood and adopted in the healthcare sector (Oh et al., 2005).
In the on-going debate, two definitions stand out (Pagliari et al., 2005):
“e-health is an emerging field of medical informatics, referring to the organization and delivery of health services and information using the Internet and related technologies. In a broader sense, the term characterizes not only a technical development, but also a new way of working, an attitude, and a commitment for networked, global thinking, to improve health care locally, regionally, and worldwide by using information and communication technology” (Eysenbach, 2001, p.2-3)
Currently, new research is conducted to scout for the need for a revaluation of the definition (Boogerd et al., 2015). Since this research will not be published within the time span of this research, a careful choice has to be made between the two previous stated definitions. The current study chooses to adopt the definition of Eng (2004). This definition is more concise and understandable for those who are not familiar with the extensive debate in this area. This is an important factor for the conductance of the current study since this definition will be provided to interviewees with limited expertise.
Contributions and challenges of eHealth innovations
eHealth innovations are regarded as highly important for the healthcare sector. eHealth innovations have the potential to increase the quality of care, while reducing its costs (Lakovidis, 2012; Van Gemert-Pijnen et al., 2011). These assumptions are important regarding the current need for more affordable healthcare systems, which ought to provide high quality care to an aging population and to an increasing number of chronic patients (e.g. Aujoulat et al., 2008; Lakovidis, 2012; Schubart et al., 2011). Furthermore, eHealth innovations are said to foster patient empowerment (e.g. Alplay, Blanson Henkemans, Otten, Rövekamp, & Dumay, 2010; Aujoulat, d’Hoore, & Deccache, 2007), which addresses the
increasing demand for involvement of patients in their healthcare.
Despite the fast increase of eHealth technologies and research, the adoption rate of these innovations is relatively low and the number of successful eHealth innovation is scarce (Van Gemert-Pijnen et al., 2011). eHealth innovations are facing high attrition rates (Schubart et al., 2011). Meaning that the proportion of users that drop out before completing their application is high. This stands in conflict with the evidence-based medicine paradigm, which dictates that high dropout rates makes outcomes less reliable. Consequently, researchers often choose to ignore dropout numbers or not publish their work (Eysenbach, 2005).
evidence supporting the benefits communicated by developers. Unfortunately, successful eHealth innovations are rare. Therefore, evidence confirming their benefits are scarce (Black et al., 2011; Pagliari, 2007). This misfit between expected and evidence-based benefits creates an impasse, hindering further adoption of eHealth innovations (Van Gemert-Pijnen et al., 2011). Therefore, further research is needed for a comprehensive understanding of factors influencing eHealth adoption.
Factors influencing the adoption of eHealth innovations
The current study categorized the factors that have been mentioned in existing
literature into a implementation and a post-implementation phase (Szajna, 1996). The pre-implementation phase covers the contextual inquiry, value specification and development (Van Gemert-Pijnen et al., 2011). The most important factors identified here are the lack of user involvement (e.g. Van Gemert-Pijnen et al., 2011) and ineffective business models (e.g. Limburg, Gemert-pijnen, Nijland, & Ossebaard, 2011). The post-implementation phase covers the operationalization, adoption and summative evaluation (Van Gemert-Pijnen et al., 2011). Most important factors here have to do with the misfit between the postulated and realized benefits (Van Gemert-Pijnen et al., 2011): (1) Patient empowerment (e.g. Aujoulat et al., 2007), (2) Quality of care (e.g. Asoh & Rivers, 2010), (3) Cost reduction (e.g. Chaudhry et al., 2006), (4) Safety & privacy (e.g. Black et al., 2011).
Pre-implementation phase
Business model. Deficient business models are a major factor contributing to eHealth
innovations failure (Van Limburg et al., 2011). Successful eHealth innovations are not only able to articulate a clear value proposition to the patient users, but also have an adequate business model. A business model is defined as a description of: “.. the rationale of how an organization creates, delivers, and captures value” (Osterwalder & Pigneur, 2010, p.14). It incorporates the way in which the innovation will generate sustainable profits (Menko, Visser, Janssen, Hettinga, & Haaker, 2013; Mettler & Eurich, 2012; Van Gemert-Pijnen et al., 2011).
Furthermore, eHealth innovations are often technology-driven and loose focus on the needs of their user (Menko et al., 2013; Mettler & Eurich, 2012). This opposes the idea of a value proposition directed towards the user. Also, the involvement of HPs is essential for eHealth innovations (Valeri, Giesen, Jansen, & Klokgieters, 2010). Consequently, effective eHealth business models should have a value proposition that incorporates the added value for the users (patients and HPs), while generating a sustainable profit.
User involvement. As stated above, user involvement is important in the development
phase of eHealth innovations. User involvement is defined as “... the importance and
personal relevance of a system to a user” (Hartwick & Barki, 1994). The point where a user is actively showing behaviour that represents a contribution to the development process is referred to as user participation (Hartwick & Barki, 1994).
The involvement of end-users (patients and HPs) is regarded as a major factor that increases the probability of adoption of eHealth innovations (Van Gemert-Pijnen et al., 2011). In addition care providers, patients, citizens, organizations, managers, academics and policy makers are identified as stakeholders for eHealth innovations (Pagliari et al., 2005). One challenge lays in the diversity between these groups (Catwell & Sheikh, 2009).
There are two main approaches in literature that are concerned with how to change the development process. One approach argues for a user-centred design, i.e. HPs and patients (e.g. Christensen & Mackinnon, 2006; Dansky, Thompson, & Sanner, 2006). On the contrary, the second approach suggests a comprehensive overall design, with all stakeholders involved (e.g. Black et al., 2011; Eysenbach, 2005).
Thus, user involvement during the development phase (either the end-user and/or all the stakeholders) positively influences the adoption of eHealth innovations. In this case, the actual needs of patients and HPs can be identified, and matched with technological capabilities and the business model design. To focus on the most important users, the current study will adopt an user-centred design approach.
Post-implementation phase
Patient empowerment. The benefit that is most identified with eHealth innovation is
patient as a unique individual case before conducting a diagnosis (Holmström & Röing, 2010). Patient empowerment on the other hand, represents the paradigm in which there is power equality between patient and physician (Holmström & Röing, 2010). Additionally, empowered patients are more involved and concerned with healthy living (Asoh & Rivers, 2010; Aujoulat et al., 2007). Thereby, the focus shifts from healthcare for disease treatment, to disease prevention (Aujoulat et al., 2007).
eHealth innovations have the potential to increase patient empowerment in a four-fold manner: (1) Insight into one’s own health condition, (2) making informed choices, (3) engaging self-care activities and developing self-care habits, and (4) living independently (Alplay et al., 2010). By informing patients about their condition, patients have increased knowledge, skills and motivation (Alplay et al., 2010; Aujoulat et al., 2007).
Furthermore, there are some studies showing positive effects of self-management (i.e. patient empowerment) to cost reduction and clinical outcomes (Bodenheimer, 2002). However, more convincing empirical evidence is needed in terms of effect-sizes and quality of data (Alplay et al., 2010; Kuijpers, Groen, Aaronson, & van Harten, 2013; Samoocha, Bruinvels, Elbers, Anema, & van der Beek, 2010).
Quality of care. Although much attention is directed to the increase of the quality of
care, systematic reviews show at most modest effects (Black et al., 2011; Pagliari, 2007). This contrasts the wide assumption by academia, managers, and policy makers, that eHealth innovations will ‘without any doubt’ benefit the quality of care (e.g. Lakovidis, 2012; Mettler & Eurich, 2012; Tsiknakis & Kouroubali, 2009). Despite the lack of empirical evidence, the beneficial effect of eHealth innovations on the quality of care is often mentioned in introductions (e.g. Lakovidis, 2012; Tsiknakis & Kouroubali, 2009) or papers even start with it (e.g. Asoh & Rivers, 2010; Mettler & Eurich, 2012). The reason for this is two-folded. Firstly, literature tends to focus on the benefits rather than risks and costs, although there is strong evidence for failed innovations (Black et al., 2011). Secondly, eHealth literature is said to be characterized by low quality, low quantity and inconsistency of evidence (Black et al., 2011). Only a handful of studies provide empirical evidence supporting quality benefits generated by eHealth innovations, without being generalizable (Black et al., 2011).
benefits when discussing eHealth innovations (e.g. Sabnis & Charles, 2012). A reason for this is that many authors rely on non-generalizable case studies or unsupported testimonials (Gustafson & Wyatt, 2004; Schweitzer & Synowiec, 2012). However, providing clarification about the expected cost reduction is essential for attracting investors (Mettler & Eurich, 2012; Schweitzer & Synowiec, 2012; Van Limburg et al., 2011). Furthermore, an increasing demand for eHealth innovation would potentially reduce prices and create economies of scale (Schweitzer & Synowiec, 2012).
Privacy & safety. Another postulated benefit in eHealth literature is the increase of
privacy and safety which is demanded by patients regarding their personal health related information (Gunter & Terry, 2005). Many eHealth innovations have entered the market using some kind of healthcare information in combination with the Internet. Currently, there is a large number of providers of eHealth innovations that use medical patient data. This fragmented industry creates a security risk for patient privacy and safety (Asoh & Rivers, 2010). A security breach can damage the trust of patients in eHealth, and therefore negatively influence the adoption of eHealth innovations. Therefore policy makers should focus on system governance and standards to protect the privacy and safety of patients (Lakovidis, 2012). In sum, privacy and safety might be a prerequisite for eHeatlh innovations and essential for successful adoption.
Summary of factors influencing the adoption of eHealth innovation
To conclude, this chapter provides an overview of the factors influencing the adoption of eHealth innovations in the pre- and post-implementation phase in table 1.
Pre-implementation phase Post-implementation phase
+ Business models (e.g Limburg et al., 2011)
+ User involvement (e.g Van Gemert-Pijnen et al., 2011)
+ Patient empowerment (e.g. Aujoulat et al., 2007)
+ Quality of care (e.g. Asoh & Rivers, 2010)
+ Cost reduction (e.g. Chaudhry et al., 2006)
+ Privacy & safety (e.g. Black et al., 2011) Table 1: underlying factors in existing literature which support eHealth adoption
METHODOLOGY
This chapter will first elaborate on the research design, quality, and setting. Then, the case selection and data collection will be discussed.
Research design
To answer the research question, the current study focuses on how underlying processes affect the adoption of eHealth innovations.
Because of the explorative character of this study, an in-depth case study design with two contrasting cases (successful and unsuccessful, Eisenhardt, 1989) was chosen as appropriate. Hence, the study focuses on the underlying dynamics within a single setting. This type of qualitative research combines observations, interviews, and archives evidence on which a content analysis is conducted. The current study aims to inductively built theories and formulate propositions that can be used for future research purposes.
Research quality
In order to enhance the quality of this case study, several techniques were used to increase the controllability, reliability and validity (Van Aken, Berendsen, & Van der Bij, 2012; Yin, 2004).
Controllability. The controllability reveals how the study has been conducted. It
enables others to redo and control the study and is a prerequisite for the other quality criteria. By carefully documenting and storing research progression and data, this study attempts to maximize its controllability. For example, recordings and detailed transcriptions were made of all interviews.
Reliability. The reliability refers to the degree in which a study can be replicated by
other researchers. Standardized methods were used to analyse and collect data. In order to increase reliability, a variety of stakeholders were interviewed and their perceptions were taken into account when proposing assumptions. In addition, an interview protocol was used during the data collection. Ideally, multiple researchers would have analysed the raw data in order to compare findings. However, due to time constraints, this was not possible. There are three main biases regarding reliability:
• Instruments bias: Study results depend on the particular instrument used in your study • Respondents and circumstances bias: Study results dependent on circumstances or
respondents
See Table 2 for more details on which tactics were used to guarded the reliability of this study.
Validity. The validity refers to the justification of the study results and the way they
have been generated. The validity contributes to the acceptance of study results. There are three main validity dimensions:
• Construct validity: Measurement of constructs in a study - did you really measure what you intended to measure
• Internal validity: Relationship between phenomena - can you really prove that a is caused by b
• External validity: How generalizable are study results
See table 2 for more details on which tactics are used to guard the validity of this study.
Test Case study tactic When tactic occurs
Construct validity
• Use multiple sources of evidence (triangulate) • Establish chain of evidence
• Have key informants review draft case study report • Data collection • Data collection • Composition Internal validity • Do pattern matching • Do explanation building • Address rival explanations
• Data analysis • Data analysis • Data analysis
External validity
• Use theory in single-case studies • Specify case specific content
• Research design • Research design
Reliability • Use case study protocol
• Develop case study database
• Data collection • Data collection
Research setting
The current study was set in the Radboud University Medical Center (Radboudumc) in Nijmegen, the Netherlands. In order to reduce the number of confounding variables influencing both cases, one specific medical department was chosen (out of a total of approx. 50). Three steps characterized the selection process.
First, an initial selection of departments was made (six in total) based on unstructured interviews (Faems, Janssen, Madhok, & Van Looy, 2008) with knowledgeable people from the REshape Centre of Innovation at the Radboudumc. The contacted experts were most knowledgeable to provide guidance in the department and case selection, given that the REshape Centre of Innovation is one of the centralized departments for innovation endeavours in the Radboudumc and a highly viewed national- and international authority on healthcare innovation.
Secondly, these potential departments were ranked according to feasibility of conducting the current study (and hence answering the research question), available documented information, and innovativeness as perceived by the interviewees. In addition, a societal relevance of the department was added as a factor, meaning that the department of choice preferably represented an ageing population and chronic patients. Key persons in the six pre-selected departments were approached in succession.
Lastly, an unstructured interview was conducted with a departmental key person (30 minutes). The first interview with a key person resulted in a match. In this interview, the commitment to perform the research in this department was made and the possibilities for eHealth innovation cases were discussed. After consulting with the REshape Centre for Innovation, it was decided to use the medical department of General Internal Medicine (Algemene Interne Geneneeskunde; from now on referred to as AIG) as research setting. This department represents a variety of chronic illnesses, has a high number of elderly patients, and a wide variety of disease diagnoses. Hence, by choosing this department, the societal relevance of the current study was secured. After the department selection, relevant contrasting cases were selected.
Cases selection
selecting contrasting cases, it is important that both cases differed based on theoretical arguments. On the other hand, eHealth innovations are characterized by the wide diversity among each other. The key person is a member of the Executive Board of AIG and is well aware of innovation activities within the department. Potential cases were discussed and in a follow up e-mail the decision was made for a successfully (MijnRadboud) and unsuccessfully (MijnZorgnet’s online community “Vaatcentrum”) adopted eHealth innovation. The evaluation of successfulness of the adoption was based on the perception of the key person in combination with (at that time) available information about the individual cases. Furthermore, the provided definitions were taken into account. Aim was to identify two cases of eHealth innovations, which were extreme and clear in their contrast of success and failure. In addition, frequent consultation with knowledgeable experts of the Centre for Innovation ensured the right assumptions in this matter.
Furthermore, for the case selection it was important to choose cases based on their likelihood of replication and extending the emergent theory, i.e. theoretical sampling (Eisenhardt, 1989; Yin, 2004). Namely, both innovations have been applied organization-wide, and similar innovations are applied in hospitals around the world. Additionally, the same IT-architecture used in both cases is being applied worldwide (i.e. EPIC and digitalepoli.nl). Therefore the selected cases are well suited for the current study.
Aim of this exploratory study is to conduct theory building, rather than generalizing. Yet, case studies do allow room for analytical generalization (rather than statistical generalization), and therefore help producing topics and issues of interest and proposing conclusions based on logic (Yin, 2004).
What is interesting about these two cases is that they have the same origin but their success differs. Therefore a certain level of control regarding confound variables in these early phases can be assumed.
Both innovations initially aimed to create a Personal Health Record. MijnZorgnet was a commercial entity providing online communities to healthcare institutions (one of which was the online community Vaatcentrum, which will be subject to this case study). MijnRadboud was an internal innovation of the Radboudumc (The use of MijnRadboud within AIG will be subject to this case study).
Data collection
The primary sources of data are gathered through semi-structured interviews. The construction of these interviews was done in an iterative process. Meaning that the interview guide changed along with the progression of insights gathered out of data collection and analyses.
Additionally, secondary data was gathered concerning cases, department and organization (e.g. product initiation documentation, presentation slides, personal notes of workshops, data from the intranet). The triangulation of this data increased the internal validity of the study (Eisenhardt, 1989; Faems et al., 2008).
For the theory building a number of stages can be distinguished in this study. As a start, unstructured interviews have been conducted with innovation-oriented professionals at the REshape Centre for Innovation in the Radboudumc. In addition, own observations, background documentation and a literature review spurred and enriched this search. The cases, the department, and its key person were selected. With the help of this key person, interviewees were selected and later approached. They were asked to participate in a semi-structured interview. In communication with old interviewees, new interviewees were identified and approached, leading to a snowballing effect. In three weeks time, a total of 23 interviews have been held.
Interviews took 30 to 90 minutes, with a maximum of three interviews a day, and were conducted by one and the same interviewer. Afterwards, interviews were transcribed and analysed. Every interviewee was granted anonymity. Furthermore the interviewees were asked about their personal details, affinity with ICT, and involvement and role in MijnRadboud and Vaatcentrum. At the end of every interview a recap was given to ensure that the interviewee had nothing to add. Sometimes interviewees were asked for follow up information, documentation, or possible new interviewee contacts. Furthermore, during the interviews, field notes were taken, describing notable feelings, events or situations perceived by the interviewer.
Interviews were held with patients, HPs (e.g. internists, nurses, physician assistants), healthcare managers (HMs, e.g. executive board department, operational manager), and other professionals outside AIG that were involved in the cases on an organizational level.
every interviewee ensured to reduce respondent bias. Nevertheless, based on advice, the email sent to patients was adjusted to make it more understandable. However, the core content was not changed. Notably, the interview with the key person was potentially biased by the 30 minute unstructured interview that had taken place beforehand (see above). Nevertheless, by being on the Executive Board and knowledgeable about both cases, this person was too valuable to exclude from the current study. Due to privacy restrictions, the interviewee selection among patients could only take place with the help of medical HPs.
Interviewees from outside the department (or hospital) have been selected mainly because of their role in the development and/or implementation of one (or both) eHealth innovations. For example the project managers of both eHealth innovations, the director of the REshape innovation center, and a senior researcher at IQ healthcare have been interviewed.
Data analysis
Data analysis started after the first ten interviews were held. In the following interviews, the interviewer did refer and relate back to the statements in previous interviews to evaluate and test gained insights and assumptions.
Data was analysed using an inductive approach. For this a content analysis of each case was completed. Content analysis was conducted by reviewing the transcript of every interview to discover patterns and formulate codes. As time was progressing, these codes were categorized and more solid concepts were formulated. Statements and observations were highlighted in interviews. This way cross-interview patterns emerged for both case, leading to supporting and contrasting insights. To come closer to the underlying reasons that influence the adoption of each eHealth innovation, the interviewer focused in later interviews more on how and why certain concepts influenced the adoption. Additionally, the insights gathered from interviews were compared with documentation and personal notes that were sent by interviewees and gathered from data sources.
Afterwards, a cross-case analysis was preformed containing the similarities and differences between both cases. In the discussion, these outlines will be linked to literature and explanations for occurring insights will be discussed and eventually propositions will be formulated.
RESULTS
This chapter will focus on performing a within-case analysis of the cases subject to this study (Eisenhardt, 1989). Primary data gathered from interviews and secondary data (e.g. project documentation and personal notes) guided this chapter. Afterwards, a cross-case analysis is performed to assess the similarities and differences between the two cases. Vaatcentrum was identified as unsuccessful; MijnRadboud was identified as successful.
MijnZorgnet’s online community Vaatcentrum
Beginning of 2012
In the beginning of 2012, the Board of Directors of the Radboudumc provided a subsidy for the implementation of 50 online communities. A key issue was the financial support. MijnZorgnet received €150.000 to facilitate software development for online communities. Furthermore, the internal department for process improvement and innovation (i.e. PVI) also received €150.000 to facilitate the successful implementation of these online communities. As a result, medical departments were supported in the implementation of a free online community. The free option and a committed attitude of the Board of Directors convinced medical departments to adopt the eHealth innovation. As HP-#22 underlines this: “When Vaatcentrum was introduced, it was like ok everybody has MijnZorgnet so we also need it.”
In order to enhance the implementation of these 50 online communities, an implementation plan was formulated (by PVI and MijnZorgnet). It was a centralized document aimed to advise and support the implementation of the online communities (on an organizational level). This implementation plan, together with a scenario guide provided a detail look into pre-implementation phase of this eHealth innovation. Account managers of MijnZorgnet and PVI advisors facilitated and supported the implementation.
required, and a community manager (who is closely involved with the patient group) should be appointed. Departmental documentation shows compliance of AIG to these requirements.
February 2012, Workshop 1 “Organize a community on MijnZorgnet”
As part of the implementation plan, a total of three workshops were organized for all participating departments. In addition to a general introduction to online communities in
healthcare, project groups were formed. According to personal notes, members of a project group were a community ambassador, a community owner, redactor, content expert, and a PVI advisor. So far, data (i.e. documents) implies a clear and structured deviation of functions and responsibilities. However, members of the project group perceived this differently. When asked how many HPs were involved in Vaatcentrum, HP-#21 states: “… I was the one who had the overview of which task was done by whom, and very occasionally someone else posted something. I was able to receive feedback from my colleagues about the content, but I was the one posting it.”
Furthermore the workshop consisted of a demo of a successful community, a trial design of community, training in the user surface of the platform. Lastly, the workshop was evaluated in the group and action points were formulated to complete before the next meeting.
March 2012, Workshop 2 “Start a community on MijnZorgnet”
In March, PVI organized a second workshop for all project groups (Radboudumc-wide). Here, the project groups had to present and pitch a demo of their online community and discuss their experiences. In personal notes a clear outline of a possible interface of a webpage is sketched, indicating that different design possibilities were discussed and advice was offered. The successful cases of IVF-poli and ParkinsonNet were shared, the supported resources were discussed and project groups formulated a communication plan. In other personal notes it was clearly stated (underlined, rimmed and in capital letters) “Prevent empty forum”, and “always respond, don’t wait, be proactive”. Later, the lack of content generation or dynamic content would be identified by interviewees as being an important factor for the lack of adoption.
April 2012, Workshop 3 “How to use information from MijnZorgnet”
illustration. The Vaatcentrum group then worked on an assignment in which they had to think about which steering information was valuable, what that information indicated, to whom and how they should report it. One of the most important findings from reviewing personal notes was “involve target group!” (written in capital letters and underscored). The importance of involving patients was discussed and outlined during this workshop. Additionally, personal notes indicated that information would be communicated twice a month to the Executive Board of AIG. An elaborated interface was sketched and content generation initiatives (e.g. blogs) were mentioned.
July 2012 – December 2012, Implementation Vaatcentrum
In July, Vaatcentrum started. HP-#21: “We started very enthusiastic”. Nonetheless, wide participation of patients was lacking. In January 2013 only 7 patients were enrolled. In December 2012, a presentation was facilitated to the community owner and the PVI advisor. In this presentation the vision and added value of the community to HPs and patients was clarified. The community at that time was still not successful.
December 2012, end of Vaatcentrum and MijnZorgnet
After the subsidy directed by the Board of Directors to MijnZorgnet was completed. Due to a number of reasons (e.g. diverging vision among initiators) MijnZorgnet failed to adopt a sustainable business model and therefore stopped operating after the subsidy was completed. Only a handful of online communities were transferred from MijnZorgnet to another platform (digitalepoli.nl, owned by the former ICT architect of MijnZorgnet). MijnZorgnet was bought by the Radboudumc and is now used for other research purposes (by the department IQ healthcare).
MijnRadboud
MijnRadbouddossier
January 2010, is generally considered as the start of (the predecessor of MijnRadboud) MijnRadbouddossier. HM-#07: “In 2010 [MijnRadbouddossier] started for the department of fertility. What they had was a application that enabled you to make a patient portal per medical department and then you also had control over the enrolment procedure.”
MijnRadbouddossier was an ICT product designed by the Information Management department of the Radboudumc. It consisted out of three functionalities: (1) Information website page, (2) community functionality and chat, and (3) specific aspects of a patients medical file that could be viewed by patients. The Radboudumc at that time had its own designed Electronic Patients Dossier (EPD). MijnRadbouddossier was designed to work on this software.
In 2011, the Radboudumc aimed to roll out the product to other departments.
HM-#07: “I was asked to go and approach those twenty departments that actually did want the software and try to sell and implement that, and then really make that also part of the work process.” At the same time, MijnZorgnet was entering the market with their new platform. As a consequence, unwanted competition between MijnRadbouddossier and MijnZorgnet arose. In a meeting at the beginning of 2012 it was decided, together with the Board of Directors, that MijnRadbouddossier should drop their community functionalities, and focus on their coupling to the hospitals EPD. In addition, MijnRadbouddossier would no longer focus on individual departments, but implement their product for organization-wide (all medical departments). The strategic idea here was outlined by HM-#02: “So you talk about … one eco-system instead of different ego-systems” and HP-#01 “all those walls should get out and all must not be ego-systems but become eco-systems”
MijnRadboud
In October 2013, EPIC was implemented and therefore MijnRadbouddossier officially ended. Accordingly, information and users from MijnRadbouddossier were now integrated into MijnRadboud. The implementation of EPIC had a large impact on the organization and some professionals felt like they experienced an ICT overload. The impact of the change in the work-process was considered severe. Therefore, the implementation of MijnRadboud was not a priority until the end of 2014. This is confirmed by internal documentation concerning MijnRadboud.
In December 2014, a project mandate for MijnRadboud was again assigned. Accordingly, in April 2015, the project mandate was developed and approved. As a result, a detailed project initiation document concerning the implementation and adoption of MijnRadboud was created. The following four phases can be defined from this document: Initiation, definition, realization, and evaluation. Additionally, an implementation plan, a communication plan and a business model were presented in this document.
April 1st 2015 – June 1st 2015, Phase 0: Initiation
After the project mandate was turned into a project proposition the first phase of initiation and constructing and Project Initiation Document (PID) started. Nine months of project management was conducted on an organizational level.
Beginning of May 2015, the project proposition was delivered and further developed. At AIG, a research was conducted (submitted: May 2015) to the effect of the departmental functionality of MijnRadboud to document blood pressure. This functionality enabled patients to measure their blood pressure at their homes - making a physical visit to the hospital redundant. This departmental research showed there were only minor factors hindering further adoption of the “blood measurement from home” functionality. Furthermore, this research states that MijnRadboud is a good tool for the implementation of this functionality. Still, the minor barriers for patients (e.g. boring activity, time consuming, interruption of daily activities) were managed appropriately.
June 1st 2015 – August 1st 2015, Phase 1: Definition
Integrating MijnRadboud into the clinical work process was mentioned as an important issue in the implementation plan. In terms of promotion, every new patient was asked at the registration desk if they wanted to enrol into MijnRadboud since June 2015. This led to an increase of 125% in the number of patient enrolments per week (from 200 p/w, to 450 p/w). At the end of 2015 approx. 15.000 patient users were enrolled.
Beginning of July 2015, a multidisciplinary organizational-wide session was organized including HPs and patients. Here, the proposed vision of MijnRadboud was defined and discussed in order to increase its success. Notes from this meeting indicated that over 50 participants shared their thoughts and perspectives in this meeting. Notes of these participants highlighted the tension field of transparency of medical patient information from a HP perspective: “The dilemma [whether patients have to get access to all data of their file, including notes of HPs] is concentrated on the suspicion of diagnoses that would make the patient ‘unnecessary worried’ to ‘the patient has the right to know.” On the other hand, patients underscored their legal right to have insight into personal information in this meeting. Additionally, the meeting report states one patient arguing that the individual patient should be given a choice to review medical information (e.g. ‘bad-news’ lab results). The concluding theme of meeting was ‘leave it to the patient’. The fear of misinterpretation of information by patients is also seen in interviews.
HP-#06: “There is absolutely nothing to hide, but too much information can be dangerous. It is about interpretation. If you formulate everything in a certain way because you know that the patient knows. That it is the question. If you can be that openly. For example, somebody has a spot on his lung photo, I briefly ask or call a pulmonologist how to continue from that and the pulmonologist states ‘yeah it can be cancer’. If the patient sees this kind of communication, you could say this is not good for him. Should the patient not know that he has a spot on his long and needs further treatment? Yes, but that information must be given to him in the right context with the right explanation.”
Concerning the implementation plan, news from the hospital’s internal communication (i.e. intranet) showed a warning that “in second half of 2015 many HPs would have to deal with MijnRadboud”. The importance of user involvement in developing a vision and implementing is highlighted in this message. Therefore the project leaders indicated that they intended to contact medical departments, patients, super users (HPs that are trained before the general population in order to facilitate extra support to their colleagues), and other internal stakeholders.
August 1st 2015 – November 1st 2015, Phase 2: Realisation
At the end of October 2015, more than 13.000 patients had access to MijnRadboud (450 new enrolments every week). The number of patients that used MijnRadboud in this period (August – October) over seeded 8.000. Most patient users (total of 6.500) were between 50 and 69 years old. Most used functionalities of MijnRadboud were: (1) Appointment overview (3.422 in October 2015), (2) Lab results (3.178 in October 2015), and (3) Reviewing letters (2.895 in October 2015)
In October, patients have asked 347 (non-urgent) questions to their healthcare team. Patients also graded MijnRadboud on average a 7.8 out of 10 (March: 7.6, April: 7.9, May: 7.3, June: 7.9, July: 7.6, August 8.1, September: 7.9).
Additionally, phase 2 evaluated the employment of the patient portal of MijnRadboud. In this phase the implementation of MijnRadboud was further expended from four pilot departments (including AIG) to all medical departments and the communication plan was realized.
November 1st 2015 – December 31st 2015, Phase 3: Evaluation
Similarities and differences
Co-developed vision
Even though given the chance for feedback, patients and HPs did not felt they had any impact. Patient-#16: “You are able to ask questions, but you don’t want to bother anybody. I once filled in a questionnaire after my operation. ...but I never heard of it again.” Furthermore HPs indicated that they resolve issues decentralized. HP-#05: “I’m not the central contact person, but colleagues ask me because I have good connections with a developer”. This opposes the view of the central contact person for MijnRadboud within AIG: “I observe most of the feedback not by myself because I work too little with the system, but when I hear that from employees or patient of employees I can bring it to the heads of the policlinic and agree on a common strategy…” and of HM-#7: “MijnRadboud has a wish-list of things that people want to have changed. You test this on impact, technical, legal and ethical limitations. Because functionalities on MijnRadboud are either ON or OFF…complains are heard, evaluated if we can do something with it, but you cannot always resolve these issues.”
Business model
Both, Vaatcentrum and MijnRadboud were funded by a central body in the Radboudumc. However, MijnZorgnet was not able to articulate a sustainable business model and eventually failed to continue its practice after funding stopped.
Furthermore, both innovations were offered for free, but Vaatcentrum had a higher dependency on user involvement. Without a requested charge, users were not committed to contribute to the online community. According to HM-#02: “…users have to buy themselves in”. This lack of commitment was also visible on a managerial level for the online community (e.g. project team). HP-#21: “…let’s just start and look how it goes”. Contrary, the service of MijnRadboud is independent on the number or participation of users.
User involvement
with the eHealth innovation. HP-#12: “I’m not familiar with the online community, I know that it works within MijnZorgnet, I know that there were communities and that there were also doubts about implementing them because, A it is a lot of work, and B what you heard from other people that there were not a lot of people active on this platforms.” The same becomes apparent for patient users. Patient-#11 from Advisory Board of AIG (PAR-AIG) indicates (when asked if MijnZorgnet has ever been discussed in meetings): “Yes they talk about it, but it is not like advise is being asked about it”.
MijnRadboud includes both user groups in the pre-implementation phase (e.g. steering committees and meetings) and in the post-implementation phase (e.g. evaluation sessions and online questionnaires). The patient member of the steering committee of MijnRadboud indicated that he provided benchmark information about a similar eHealth innovation from another Dutch medical centre. Furthermore, the Patient Advisory Council (PAR) actively advised the Board of Directors in this matter and patients can fill in an online evaluation questionnaire on MijnRadboud.
Clinical integration
Although Vaatcentrum strived to do so, it did not substitute clinical workload. Members of the project team did not receive any allocation of time to manage Vaatcentrum. In the end, because no clinical benefits were received from working with the online community, they could not see the added value in it.
MijnRadboud has so far been able to articulate its benefit to the clinical process better. It has done so via generating clinical benefits. Patient-#15: “A measurement (blood measurement from home functionality) preformed by a doctor is just one moment… if you fill in your blood values from home, you can do this twice a day and then a few days in a row, than you have a better average trend and that increases your understanding in the development of your blood values”. By substituting clinical processes that previously were done at AIG, no additional time is required from HPs. According to HM-#8 this is an important factor contributing to adoption, since it would be hard to incorporate innovations into routines of HPs.
Ease of use
to the ease of use of MijnRadboud in the upcoming year…for elderly patients it is not always simple”. In addition to the ease of use on a clinical outcome, MijnRadboud also excelled on an administrative level, HP-#19: “…the benefit is mainly a better organized patient administration.” This view is widely supported by patient users, patient-#16 (when asked to the advantages of MijnRadboud): “Everything is just there, you can review whenever you want.”
Patient empowerment
Data form interviews strongly suggests that Vaatcentrum led to no increase of patient empowerment, which is in line with an earlier conducted study (Tuil, Verhaak, Braat, de Vries Robbé, & Kremer, 2007). MijnRadboud however, is better perceived. Patient-#20 (when asked to degree in which he is better informed by eHealth innovation): “That is for MijnZorgnet totally nothing 0.0, and for MijnRadboud currently a 6.5. If you compare it then you did not have the information in MijnZorgnet but is now very clear in MijnRadboud, and the possibility to schedule appointments…I am able to view and interpret my own lab results now”.
Quality of care
Vaatcentrum did not add sufficient value to patients. HM-#13 underscored this by: “What influences the success is whether patients understand what the added value is for them”.
MijnRadboud added value to the quality of care, according to both user groups. HP-#22: “telephone calls, emails and patient questions are part of the quality of care… but they are not incorporated in the DBC (diagnose treatment combination) reimbursement system… The increase of quality, you don’t measure in research, but you see that because patients are more involved, which positively influences their treatment…”. Patient-#20: “For MijnRadboud the quality of care improved, MijnZorgnet did not cause any change…”
Cost reduction
Vaatcentrum failed to generate cost reduction. HM-#14 indicated that maintaining the discussion forum required a lot of time. Patient-#20: “Added value of MijnZorgnet was very limited, only news into new developments about my condition, but you can also find that on the Internet”.
MijnRadboud is currently integrating new functionalities (like pre-consult questionnaires) which safes time for HPs. The “blood measurement from home” functionality actually “decreases the costs because patients do not have to come to the poli anymore” (HP-#05).
Safety & Privacy
DISCUSSION
eHealth innovations can bring a valuable contributions to today’s challenges faced by healthcare systems worldwide. The current study aimed to gain insight into factors leading to the adoption of eHealth innovations. For this an in-depth case study was performed, using two contrasting cases (successful vs. unsuccessful adoption). The implications of the results are two-folded:
Firstly, results provide evidence for new factors that potentially underlie the adoption of eHealth innovations, namely: (1) co-developed vision, (2) clinical integration, and (3) ease of use.
Secondly, supportive evidence is provided for factors from existing literature, i.e.: (1) Business models, (2) User involvement, (3) Patient empowerment, (4) quality of care, (5) cost reduction, and (6) Privacy & safety. The similarities between these findings and the existing literature indicates that the results are likely to be generalizable.
Hence, the current study provides valuable insights for stakeholders (e.g. managers) that use, develop or manage eHealth innovations.
Novel factors
Co-developed vision
eHealth innovations which are co-developed are more likely to succeed, based on the current data. Data suggests two contrasting insights. On the one hand, eHealth innovations should be very tailored to the specific needs of individual patients. On the other hand, missing a generic character and shared organizational vision was a major factor influencing the low success rate of Vaatcentrum in contrast to MijnRadboud. By incorporating a co-developed vision, MijnRadboud was able to provide a generic eHealth innovation that addresses individual patient needs.
Existing literature addresses the lack of multidisciplinary communication and cooperation by working towards an interdisciplinary eHealth field (Pagliari, 2007).
In order to provide an answer to these contrasting insights, the current study proposes the following proposition:
This entitles that developers of eHealth innovations should incorporate users in the pre-implementation phase and work together with the Board of directors to co-develop a vision. Taken together, eHealth innovations should be owned and created by all users.
Clinical workload
eHealth innovations have to be integrated into the clinical (i.e. medical) work process of the HP and substitute the current workload. The data shows that the reason for this is as follows:
Benefits of eHealth adoptions are not often directed to the users, who have to invest (work) time and effort for its successful adoption. HPs were particular important stakeholders for achieving successful adoption. They had to manage the online community, generate dynamic content and attract patient users. Oddly enough, they had to perform these activities in addition to their current activities and without financial compensation. Nevertheless, the added value of the eHealth innovation was directed to patients. Hence, the eHealth innovation generated an added value to one user group (i.e. patients), but is dependable for it success on the input of another user group (i.e. HPs). Although the data shows that HPs are primarily concerned with the benefits of the patients, to have no other incentive at all to participate seems to have a negative influence on the successful adoption. When reviewing MijnRadboud regarding this issue it becomes apparent that HPs did not experience an increase of workload although providing input. Furthermore, the benefits were allocated towards patients.
This difference is due to the integration into the clinical process. MijnRadboud has a coupling to the EPD of the Radboudumc. Therefore input of HPs is automatically generated and furthermore made available for patient users. These insights lead to the following proposition:
Proposition 2: Substituting the clinical workload has a positive influence on the adoption of eHealth innovations
This way, an eHealth innovation adds value to one user group, while keeping the workload for the other user group (i.e. HPs) unchanged.
innovation will be provided as long as clinical outcomes are lacking. Hence, a department could loose money by lowering the workload. Therefore, one solution is to substitute current workload till innovations are reimbursed.
Ease of use
Strikingly, ease of use was among the most mentioned benefits an eHealth innovation should generate, for all groups of interviewees. So far, only a few studies have incorporated ease of use into eHealth settings (e.g. Dünnebeil, Sunyaev, Blohm, Leimeister, & Krcmar, 2012; Spil, Schuring, Stegwee, & Michel-Verkerke, 2005), despite the fact that it is one of the core variables in adoption theories, i.e. the Technology Acceptance Model (TAM, Davis, 1989; Venkatesh & Davis, 2000). Currently, there is one framework coupling TAM to eHealth (Tsiknakis & Kouroubali, 2009). This framework focuses on the significance of the optimal fit between individual user, technology and tasks. Nevertheless, TAM related frameworks have only experienced peripheral attention by the existing eHealth literature and furthermore have been completely neglected by the high-impact papers of the field (e.g. Black et al., 2011; Van Gemert-Pijnen et al., 2011). The results indicate that generic technology acceptance theories should be implemented into the field and therefore expand its inter-disciplinarity (Pagliari, 2007).
Therefore the following proposition is formulated and future research is encouraged: Proposition 3: The ease of use of an eHealth innovation positively influences the adoption of eHealth innovations
Known factors
Business models
Results support the importance of business models for the adoption of eHealth innovations, in particular for eHealth innovations that require a high degree of user participation. Existing literature indicates that eHealth innovations often end when project funding stops (Menko et al., 2013). Main reason is the inability to generate a business model with sustainable profit (Mettler & Eurich, 2012; Valeri et al., 2010; Van Limburg et al., 2011). The same pattern could be found for MijnZorgnet.
(Van Limburg et al., 2011). Even though, MijnRadboud is also offered for free, its use is independent of the number of users (as oppose to Vaatcentrum). Therefore future research should gather insight about the relationship between the level of dependency on user participation and a sustainable business model.
User involvement
User involvement is of major importance for eHealth innovation adoption (e.g. Hamid & Sarmad, 2008; Hesse & Shneiderman, 2007; Van Gemert-Pijnen et al., 2011).
In Vaatcentrum the involvement of HPs was limited (only project group) and patients were, as far as the data indicates, not involved in the development process. Contrasting, MijnRadboud involved HPs and users in the development and the evaluation.
Vaatcentrum and MijnRadboud both emphasize the importance of involving the patient user in project documentation. The HPs are gatekeepers in both eHealth innovations and therefore their involvement is important. For MijnRadboud, the majority of HPs indicated to actively point patients to MijnRadboud. Contrasting, Vaatcentrum was merely promoted by its project team.
Thus, user involvement has a positive influence on the adoption of eHealth innovations. Therefore, future research should elaborate about the use of HPs as gatekeepers to increase the adoption of eHealth innovations.
Patient empowerment
Patient empowerment showed the biggest difference between the two eHealth innovations. Mastery of control and self-efficacy are mentioned in the literature as tools to measure patient empowerment (Aujoulat et al., 2007). In contrast to Vaatcentrum, patient indicated that they feel more informed and involved by the use of MijnRadboud. Especially the clinical functionality was mentioned by users as involving patients in their own healthcare. Additionally, the data indicates that patients felt more involved and as a consequence, were more satisfied. Hence, results confirm the relationship between patient empowerment and (successful) eHealth innovations.
Quality of care
increased and that this would be related to quality of care. However, the negative results should be taken with caution since the added value of MijnRadboud to the clinical outcome of patient’s healthcare is currently limited. According to project documentation and interviewees, there is an intention to extent the number of functionalities that support clinical functionalities. Taken together, results only support a more broad definition of quality of care that incorporates the satisfaction of the patient. Hence, more research is needed to understand which specific dimensions of the quality of care benefit from eHealth innovations.
Cost reduction
Similar to quality of care, cost reduction has only been mentioned as benefit of eHealth innovations, when specifically asked for it. HPs and HMs indicated the potential that cost reduction would have come from work processes. This is in line with the literature, which emphasizes the acknowledged potential, but empirical evidence is needed (Black et al., 2011). A main factor contributing to this is the finance structure of the Dutch healthcare system; using fixed tariffs by diagnose treatment combination. These financial arrangements are agreed upon an organizational-wide level; therefore it might be beneficial to allocate cost reduction to this level, rather than to a departmental level.
In sum, the data only partially support the assumption that the cost reduction and the adoption of eHealth innovations positively influence each other.
Safety & Privacy
Safety & privacy are thought to be a prerequisite of eHealth innovations (Nazi, 2003). Both patients and HPs indicated that they ‘trusted’ the hospitals ICT system more than other ICT solutions. The results indicate that the login procedure was a barrier for adoption. However, trust of users in the eHealth innovation appeared to be a prerequisite. Arguably, this is similar to information systems in other industries, but arguable even more important in the healthcare sector. Especially when functionalities containing medical information or functionalities are coupled to it.
Adjusted table of factors influencing adoption of eHealth innovations
Taking all these new factors into consideration, and in an effort to answer the research question, an extended table is proposed containing the new propositions that contribute to successful eHealth innovation.
Pre-implementation phase Post-implementation phase
+ Business models (e.g Limburg et al., 2011)
+ User involvement (e.g Van Gemert-Pijnen et al., 2011)
P1: Co-development vision (e.g. Pagliari, 2007)
+ Patient empowerment (e.g. Aujoulat et al., 2007)
+ Quality of care (e.g. Asoh & Rivers, 2010)
+ Cost reduction (e.g. Chaudhry et al., 2006)
+ Safety & privacy (e.g. Black et al., 2011)
+ P2: Integration clinical work process (e.g. Rodrigues, 2008)
Managerial implications
The current research contributes to the effective management of eHealth innovations by explaining the importance of a co-developed vision, clinical integration, and ease of use for achieving successful adoption. Furthermore, the study provides empirical evidence for post-implemented benefits of eHealth innovations (i.e. patient empowerment, quality of care, cost reduction, and privacy & security). This could help managers to convince potential users to adopt their eHealth innovations. Lastly, the current study provides insights into factors that should be well managed in order to achieve successful eHealth innovation adoption in the pre-implementation phase (i.e. business models and user involvement).
Importantly, the current study gives time-dependent importance to the factors, by differentiating between factors of the pre- and post-implementation phase.
Hence, results have the potential to help managers in any other settings, working with similar products to enhance the adoption of their projects.
Limitations and future research
Inherent to research, this study has a number of limitations. First, the cases selected were operational at different moments in time. Vaatcentrum had already stopped, while MijnRadboud is still operational. Secondly, a single investigator has conducted the research and analysis, therefore increasing researcher bias. Thirdly, interviewees were mainly highly educated. Future research should focus on intergroup differences. Since the current study is explorative in nature, quantitative research is needed to confirm the results.
CONCLUSION
By comparing a successful and an unsuccessful eHealth innovation, the current study has provided empirical evidence concerning the factors that underlie the adoption of eHealth innovations. By performing an in-depth case study the current research identifies and formulates propositions about three novel factors that underlie the adoption of eHealth innovations: (1) Co-developed vision, (2) clinical integration, and (3) ease of use. Furthermore, the current study provides empirical evidence supporting factors in the pre-implementation phase (business models and user involvement) and in the post-implementation phase (patient empowerment, quality of care, cost reduction, and privacy & safety).
REFERENCES
Alplay, L. L., Blanson Henkemans, O., Otten, W., Rövekamp, A. J. M., & Dumay, A. C. M. (2010). E-health Applications and Services for Best Practices in The Netherlands. Telemedicine and E-Health, 16(7), 787–791. doi:10.1089/tmj.2009.0156
Asoh, D. a, & Rivers, P. a. (2010). The empowerment and quality health value propositions of e-health. Health Services Management Research, 23(4), 181–184.
doi:10.1258/hsmr.2010.010007
Aujoulat, I., d’Hoore, W., & Deccache, A. (2007). Patient empowerment in theory and practice: Polysemy or cacophony? Patient Education and Counseling, 66(1), 13–20. doi:10.1016/j.pec.2006.09.008
Aujoulat, I., Marcolongo, R., Bonadiman, L., & Deccache, A. (2008). Reconsidering patient empowerment in chronic illness: A critique of models of self-efficacy and bodily control. Social Science & Medicine, 66(5), 1228–1239. doi:10.1016/j.socscimed.2007.11.034 Black, A. D., Car, J., Pagliari, C., Anandan, C., Cresswell, K., Bokun, T., … Sheikh, A.
(2011). The Impact of eHealth on the Quality and Safety of Health Care: A Systematic Overview. PLoS Medicine, 8(1), e1000387. doi:10.1371/journal.pmed.1000387
Bodenheimer, T. (2002). Patient Self-management of Chronic Disease in Primary Care. Jama, 288(19), 2469. doi:10.1001/jama.288.19.2469
Boogerd, E. A., Arts, T., Engelen, L. J., & van de Belt, T. H. (2015). “What Is eHealth”: Time for An Update? JMIR Research Protocols, 4(1), e29. doi:10.2196/resprot.4065
Bouchard, T. J., & Propoping, P. E. (1993). Twins as a tool of behavioral genetics. American Psychological Association, 571(7464), 7464.
Catwell, L., & Sheikh, A. (2009). Evaluating eHealth Interventions : The Need for Continuous Systemic Evaluation. PLoS Medicine, 6(8), 1–6.
doi:10.1371/journal.pmed.1000126
Chaudhry, B., Wang, J., Wu, S., Maglione, M., Mojica, W., Roth, E., … Shekelle, P. G. (2006). Systematic Review : Impact of Health Information Technology on Quality , Efficiency , and Costs of Annals of Internal Medicine Improving Patient Care Systematic Review : Impact of Health Information Technology on Quality , Efficiency , and Costs of Medica. Annals of Internal Medicine, 144(January), 742–752.
Christensen, H., & Mackinnon, A. (2006). The Law of Attrition Revised. Journal of Medical Internet Research, 8(3), 20. doi:10,2196/jmir.8.3.e20
