THE INTERNET OF THINGS IN HEALTH CARE

Loading...

THE INTERNET OF THINGS IN HEALTH CARE Word count: 17.240

Janni Illegems Student number: 01202331

Supervisor: Prof. dr. Steve Muylle Commissioner: Nils Van den Steen

Master’s Dissertation submitted to obtain the degree of: Master of Science in Business Engineering

Academic year: 2016 - 2017

THE INTERNET OF THINGS IN HEALTH CARE Word count: 17.240

Janni Illegems Student number: 01202331

Supervisor: Prof. dr. Steve Muylle Commissioner: Nils Van den Steen

Master’s Dissertation submitted to obtain the degree of: Master of Science in Business Engineering

Academic year: 2016 - 2017

Confidentiality agreement PERMISSION I declare that the content of this Master’s Dissertation can be consulted and/or reproduced if the sources are mentioned. Janni Illegems Signature:

i

Acknowledgments First of all, I would like to thank prof. dr. Steve Muylle for giving me the opportunity to study this interesting subject. Furthermore, Nils Van den Steen for providing me his excellent guidance and motivation during this dissertation. I also would like to thank dr. Ann Ackaert, Frederic Vannieuwenborg, Koen Wauters, the product manager of Healthy and Bart Van Pee for their cooperation. Last but not least, I want to say thanks to Raf Illegems, Sonja Meersman, Nancy De Keermaeker, Luc Smet and Marie Buysse for their continuous moral support during this dissertation.

ii

Table of contents Chapter 1 Introduction ........................................................................................................ 1 1.1

The Internet of Things in health care .................................................................................... 1

1.2

Research question................................................................................................................ 2

Chapter 2 Literature review ................................................................................................. 3 2.1

Internet of Things ................................................................................................................ 4

2.1.1 2.1.2 2.1.3 2.1.4

2.2

Business models ................................................................................................................ 10

2.2.1 2.2.2 2.2.3 2.2.4 2.2.5 2.2.6

2.3

Enabling technologies ........................................................................................................................... 4 Definition ............................................................................................................................................... 5 Related concepts ................................................................................................................................... 7 Example ................................................................................................................................................. 9 Business models in academic literature .............................................................................................. 10 Business Model Ontology .................................................................................................................... 10 Other frameworks for business modeling ........................................................................................... 13 Impact of the Internet of Things on business models ......................................................................... 16 IoT as an ecosystem of partners ......................................................................................................... 19 Challenges for IoT ................................................................................................................................ 20

Health care ........................................................................................................................ 21

2.3.1 2.3.2 2.3.3 2.3.4

The health care system in Belgium ..................................................................................................... 21 Recent trends in health care ............................................................................................................... 24 Business models in health care ........................................................................................................... 28 IoT in health care ................................................................................................................................. 29

Chapter 3 Methodology .................................................................................................... 30 3.1

Research objective ............................................................................................................. 31

1.2

Research design .................................................................................................................. 31

3.1.1 3.1.2 3.1.3

3.2

Case study subjects ............................................................................................................ 32

3.2.1

3.3

Qualitative research ............................................................................................................................ 31 Exploratory study ................................................................................................................................ 31 Case study ........................................................................................................................................... 31 Company information ......................................................................................................................... 33

Data collection ................................................................................................................... 34

3.3.1

Semi-structured interview .................................................................................................................. 34

3.4

Data analysis ..................................................................................................................... 35

3.5

Reliability and validity of the research ................................................................................ 35

Chapter 4 Results .............................................................................................................. 37 4.1

The Internet of Things from a Flemish entrepreneurial perspective ..................................... 38

4.2

Specialized applications ..................................................................................................... 38

4.3

The development of self-learning algorithms ...................................................................... 39

iii

4.4

Collaboration with other companies ................................................................................... 40

4.5

Privacy, security and data monetization ............................................................................. 41

4.6

Modeling tools in health care ............................................................................................. 42

4.7

Business model innovation in health care ........................................................................... 43

4.8

Technology in health care .................................................................................................. 44

Chapter 5 Discussion ......................................................................................................... 45 5.1

The Internet of Things in health care in Flanders ................................................................. 46

5.2

Technological restrictions ................................................................................................... 47

5.3

Focused applications .......................................................................................................... 47

5.4

Health care and government regulations in Belgium ........................................................... 48

5.5

Business models for IoT in health care ................................................................................ 49

5.6

An ecosystem of partners ................................................................................................... 49

Chapter 6 Conclusion......................................................................................................... 51 Chapter 7 Limitations and further research ........................................................................ 52 Bibliography ........................................................................................................................ 53 Appendix 1 .......................................................................................................................... 58 Interview guide ............................................................................................................................ 58

iv

List of abbreviations B2B: Business-to-business BMO: Business model ontology CPS: Cyber-physical system EHR: Electronic health record IoT: Internet of things M2M: Machine-to-machine NFC: Near field communication OECD: Organization for economic co-operation and development RFID: Radio-frequency identification SaaS: Software as a service WSN: Wireless sensor network

v

List of tables Table 1 Comparison of the building blocks in the Business Model Canvas and The Business Model Ontology .................................................................................................................................................................... 12 Table 2 Comparison of Johnson's et al. (2008) and Osterwalder's et al. (2005) business model framework .................................................................................................................................................................... 13 Table 3 Comparison of Gassmann's et al. (2013) and Osterwalder's et al. (2005) business model framework .................................................................................................................................................................... 14 Table 4 Comparison of Morris's et al. (2005) and Osterwalder's et al. (2005) business model framework .................................................................................................................................................................... 15

vi

List of figures Figure 1 Concepts related to IoT .................................................................................................................. 8 Figure 2 Nine Business Model Building Blocks. Adapted from “Clarifying business models: origins, present, and future of the concept,” by Osterwalder, A., Pigneur, Y., & Tucci, C. L., 2005, Communications of the Association for Information Systems, 15(1), p. 10. Copyright [2005] by the Association for Information Systems....................................................................................................................................................... 11 Figure 3 Business Model Canvas. Reprinted from Business Model Generation. A Handbook for Visionaries, Game Changers, and Challengers (p.44), by Osterwalder, A., & Pigneur, Y., 2010, John Wiley & Sons. Copyright [2010] by Alexander Osterwalder.............................................................................................. 12 Figure 4 Business model definition – the magic triangle. Reprinted from The St . Gallen Business Model Navigator (p.3), by Gassmann, O., Frankenberger, K., & Csik, M, 2013. Copyright [2013] University of St. Gallen, Institute of technology management. ........................................................................................... 14 Figure 5 Comparison of business model levels of Osterwalder, Pigneur, & Tucci (2005) and Morris, Schindehutte, & Allen (2005). Adapted from “Clarifying business models: origins, present, and future of the concept,” by Osterwalder, A., Pigneur, Y., & Tucci, C. L.d ., 2005, Communications of the Association for Information Systems, 15(1), p. 5. Copyright [2005] by the Association for Information Systems. ...... 16 Figure 6 IoT Ecosystem. Adapted from Building the Internet of Things Implement New Business Models, Disrupt Competitors, Transform Your Industry (Emerging IoT Ecosystem section, Chapter 3), by Kranz, M., 2016, Wiley. Copyright [2017] by Maciej Kranz. ........................................................................................ 19 Figure 7 Overview of the health system in Belgium. Reprinted from “Belgium: Health system review,” by Gerkens, S. & Merkur, S., 2010, Health Systems in Transition,12(5), p. 17. Copyright [2010] by Sophie Gerkens and Sherry Merkur. ...................................................................................................................... 23 Figure 8 Problems in health care ............................................................................................................... 24 Figure 9 The outcome measures hierarchy. Reprinted from “What is value in health care?,” by Porter, M., 2010, New England Journal of Medicine, 363(26), p. 2479. Copyright [2010] by Massachusetts Medical Society. ....................................................................................................................................................... 27 Figure 10 Recent trends in health care ...................................................................................................... 28 Figure 11 Elements of disruptive innovation. Adapted from The Innovator's Prescription: A Disruptive Solution for Health Care (p. xx), by Christensen, C. M., Grossman, J. H., & Hwang, J., 2009, McGraw-Hill. Copyright [2009] by Itzy. ............................................................................................................................ 29 Figure 12 Potential factors that influence the development of IoT in healthcare in Belgium................... 50

vii

Chapter 1 INTRODUCTION 1.1 THE INTERNET OF THINGS IN HEALTH CARE The title of this dissertation is very general and doesn’t say a lot about its purpose. So first some background information is necessary to give an idea why this subject resulted in a dissertation. The Internet of Things is a very popular topic today and is commonly viewed as all kinds of objects that communicate using the Internet. For most consumers, this is noticeable when buying multimedia devices. It’s difficult these days to find a television or radio that isn’t connected to the internet. This trend will most likely continue in the years to come by connecting more and more objects. This will result in huge networks and intensify the Big Data trend that started in the same way as the Internet of Things. To understand this concept, the first step is to look at how it works without going too far into the technical details. Next, a definition is needed to have a shared understanding. This is addressed in the first part of the literature review. An interesting angle is to see how companies will use this technological concept to create value for their customers. A useful technique for this purpose is looking at the business models of these companies. This is the second part of the literature review, business models for companies that use Internet of Things applications in their value proposition. The academic literature on this topic turned out to be scarce since it’s such a recent concept. Looking at business models of companies that use the Internet of Things is a broad subject for a dissertation of a year and a half. Limiting this topic to a specific sector would narrow this down and provides the opportunity to look at it in more detail. In the literature, there is often an overview of different sectors with the biggest possibilities for the Internet of Things. Health care is the sector that is always mentioned and has some very promising opportunities (Atzori, Iera, & Morabito, 2010; Holler et al., 2014; Vermesan et al., 2015). An overview of this sector and identifying these opportunities is the last part of the literature review. The academic literature on business models for the Internet of Things isn’t numerous, the literature on Internet of Things business models in healthcare is even less than that. This creates an interesting research opportunity that will be addressed in this dissertation. Equally important will be recognizing the difficulties, challenges and restrictions for this concept in this setting.

1

1.2 RESEARCH QUESTION The initial idea for a research question was to map the business models of startups in Flanders that developed an Internet of Things application in health care. Using a popular business model framework, the Business Model Ontology (Osterwalder, Pigneur, & Tucci, 2005), this could give an insight into the suitable business models for the Internet of Things in health care. However, while doing the interviews, this didn’t turn out to be a valuable contribution. The application of the startups differed much more from the proposed definition than initially assumed and the development of the Internet of Things in healthcare is going slow. Completely mapping their business model wouldn’t be very useful. That’s why a change in research question was necessary. The big gap between the reality in this sector in Flanders and the expectations from the literature gives rise to the following research question:

Which factors affect the development of IoT in healthcare in Belgium? Why is the progress of startups in creating an Internet of Things application (which fits the proposed definition) advancing with difficulty? The objective is to get an insight into how an IoT application in healthcare in Belgium looks like today. The focus will be on identifying potential difficulties, challenges and restrictions for this concept in this specific setting. This research is performed by doing an exploratory case study.

2

Chapter 2 LITERATURE REVIEW Synopsis My review of the existing literature is focused on three central topics: Internet of Things (IoT), business models and health care. The IoT section first discusses the necessary technologies for IoT. This is followed by a definition that includes some key components a system should possess to label it as IoT. An IoT application that fits the definition connects heterogeneous objects that are embedded with intelligence. This allows the autonomous interaction of these objects. The created data is integrated and analyzed by a cloud structure. During this entire process, the focus is on automation (minimizing the need for human input). Then some related concepts are compared with IoT: smart devices, a wireless sensor network, machine-to-machine communication and cyber-physical systems. An illustration of a future IoT application is given at the end of this section. A section on business models is included since this concept is an important topic when introducing new technologies. The nine building blocks of the Business Model Ontology by Osterwalder, Pigneur, & Tucci (2005) are used as a framework to describe a company’s business model. This section starts with a description of this framework, followed by the reasoning why this one is chosen. Next, the use of business models for IoT is outlined, based on the current (limited) literature on this topic. Three perspectives are discussed. The business models of companies that will implement IoT in other companies or create IoT systems, the transformation of current business models by the impact of IoT and the creation of completely new business models. To finish this section, three current challenges for IoT are identified: interoperability of smart objects, regulatory restrictions of the government and finding the right business model. The healthcare section starts with an overview of the health system in Belgium. Subsequently, some opportunities for IoT in healthcare are described, based on current problems and recent trends in this sector. The lack of business model innovation and the possibility of IoT as a technological enabler to disrupt this sector (Christensen, C. M., Grossman, J. H., & Hwang, 2009) are discussed as a final part of this chapter.

3

2.1 INTERNET OF THINGS The term Internet-of-Things (IoT) is used for a lot of different concepts and applications because there doesn’t exist a strict definition for this term. To create a common understanding, my personal view on this term is illustrated in the form of a definition. The following section first includes a basic explanation of the technologies used for IoT to get and understanding of how it works. Then the definition is presented, followed by concepts related to IoT and an example of what an IoT application could look like in the future.

2.1.1 Enabling technologies The first technology is often mistaken for IoT. Smart devices, devices connected to a network or other devices and have the ability to interact (Wikipedia, 2016). For example, a television or a fridge connected to the internet and had an ‘upgrade’ with applications that basically offer the same possibilities as a web browser but with a better interface for the user. Referring to these devices as IoT is incorrect because smart devices are only a small part of this concept. It’s even better to use the term smart objects or smart things and not just devices. Based on the definition of Miorandi, Sicari, De Pellegrini, & Chlamtac (2012); anything with the following characteristics can be used for IoT: 

Uniquely identifiable



Able to communicate (can be discovered, receive and reply to messages)



Can perform basic computations

An optional characteristic is the sensing and/or actuation capabilities (Miorandi et al., 2012). Over the last few years, there has been made significant progress in the technology of sensors; they are more economical, easy to install and cheaper than ever. In short, the ideal opportunity to make any object smarter. This is one of the reasons why IoT became such a popular topic in recent years. However, when using the definition of Miorandi et al. (2012), sensing capabilities aren’t a necessary characteristic for smart things. Secondly, the technology that will support the interaction of the smart things is a combination of mostly wireless networks: Bluetooth, Near Field Communication (NFC), Wi-Fi, Wireless sensor network (WSN), low-power wide-area networks (LPWA) or long range wide area networks (LoRa), etc. (Gubbi, Buyya, Marusic, & Palaniswami, 2013; Openshaw et al., 2014; Schatsky & Trigunait, 2011). The third useful technology for IoT applications is radio-frequency identification (RFID). RFID technology enables the design of microchips for wireless data communication and is actually a great improvement of the traditional barcode (Gubbi et al., 2013). The RFID reader can read different tags at once, require no line-of-sight contact and provide writing capabilities next to the reading capabilities (He & Zeadally, 2015).

4

The RFID reader provides the necessary power to communicate the signal from the RFID tag so no batteries have to be installed (Gubbi et al., 2013). These tags can store sensitive data since it identifies objects automatically. However, improving the security of the RFID tags is still an important subject of a lot of research papers since it requires some extra fine-tuning. Privacy and security are more difficult in the IoT context because not only users but also unauthorized objects could access data (Miorandi et al., 2012). The fourth technology, cloud computing, is necessary to aggregate and analyze all the (big) data. It’s clear this won’t be an easy task. The data from the smart things can be of any kind: temperature, pressure, altitude, motion, proximity to something else, biometrics, sound, etc. and the network consists of diverse objects with variable data formats so the integration of all this data is one of the biggest challenges for IoT (Davenport & Lucker, 2015). The actions performed by this technology are a fundamental part of IoT. Cloud computing is a platform that allows on-demand network access to computing services (Geng, 2017). This platform works in the background and is used to receive data from the smart things, analyze and interpret this data and provide web-based visualizations for the user (Gubbi et al., 2013). This is a very interesting part for companies because this will create a market with a lot of opportunities to create value for users of IoT applications. The analysis of this data in the cloud will typically be performed by big data analytics and machine learning algorithms (Geng, 2017). Machine learning is a form of artificial intelligence and allows these algorithms to improve themselves by learning from their data input. In recent years, Big Data has been a hot topic but the amounts of data created by IoT applications is even greater. These huge amounts of data first have to be transported from various sources and locations over a network, stored in a centralized database, and then analyzed by the cloud (Schatsky & Trigunait, 2011). This will require some time and a significant increase in storage volume. There is already a technology that provides a solution for this time-consuming process, edge computing1. Big companies such as Dell, Cisco and Hewlett-Packard Enterprise have developed devices; mainly gateways, routers, and servers that use this technology (Schatsky & Trigunait, 2011). These devices preprocess data such that information is transferred to the cloud instead of data.

2.1.2 Definition There are a lot more interesting technologies that will be used for IoT, but in my opinion, the technologies mentioned in the previous section are the most important ones today. IoT is a broad concept and not easy to define, but in my opinion, IoT is a synonym of the following process:

1

Also called distributed computing, ubiquitous computing, fog computing and pervasive computing (Davenport & Lucker, 2015)

5

The Internet of Things is the connection of heterogeneous, (everyday) objects embedded with intelligence (e.g. computing capabilities, a unique identifier and communication abilities (Miorandi et al., 2012)) which allows them to interact and exchange data (Cambridge Advanced Learner’s Dictionary & Thesaurus, 2017). This interaction and data exchange may be performed by different communication channels and can be direct. However, at some point, the collected data passes through a cloud structure where it’s analyzed. This process is autonomous, thus without human intervention. The process runs in the background and doesn’t continuously ask for human confirmation. At the end of the process, information can be communicated to the user in more advanced service applications but the focus should be on minimizing human input. In current technology, the communication is mainly oriented from devices to the end-user and the Internet is used to connect these end-user devices. The focus of IoT, on the other hand, is on the autonomous interaction of smart objects without using humans as an intermediate station (Miorandi et al., 2012). The connected smart objects interact, collect and share data in the background, which in the end creates useful information for an end-user or influences the end user's actions. My impression of IoT is similar to what Kevin Ashton said in 1999, when he was the first person to use the term IoT: “Today computers and, therefore, the Internet are almost wholly dependent on human beings for information. Nearly all of the roughly 50 petabytes (a petabyte is 1,024 terabytes) of data available on the Internet were first captured and created by human beings by typing, pressing a record button, taking a digital picture or scanning a bar code. … The problem is, people have limited time, attention and accuracy all of which means they are not very good at capturing data about things in the real world. If we had computers that knew everything there was to know about things using data they gathered without any help from us we would be able to track and count everything and greatly reduce waste, loss and cost. We would now when things needed replacing, repairing or recalling and whether they were fresh or past their best.” (Ashton, 2009, para. 2) According to my definition, an IoT application needs to use a cloud infrastructure for the analysis of data. However, it’s not necessary that every smart object is directly connected to the internet. When this is the case, some authors talk about an alternative interpretation of IoT, the “Intranet of Things” (Holler et al., 2014; Minerva, Biru, & Rotondi, 2015). This condition isn’t included in my definition because in my

6

opinion, whether every smart object is directly connected to the internet or not shouldn’t determine if something is an IoT application.

2.1.3 Related concepts The definition presented in the previous section can now be used the compare different concepts related to IoT but have some different characteristics. A clear distinction between these concepts allows the proper classification of applications. The use of IoT in the future could result in a mix of information for the end-user and actions from smart things. If the ultimate objective is the automation of every possible process, including human behavior, applications will directly influence human actions by actuators in the network instead of communicating information to the user (Miorandi et al., 2012). This ultimate objective is more associated with something called a cyber-physical system (CPS). CPS and IoT are very similar concepts, they use the same kind of technology (wireless networks, sensors, cloud computing, etc.) and are used for the automation of processes. They only difference is that for CPS, the focus is on the actuation of objects, controlling physical entities (e.g. logistics and production systems) and when talking about IoT, the focus is on the network structure used for the interaction of objects which allows the collection and integration of data (Minerva et al., 2015). This network structure is needed for the actuation of different objects in CPS systems. A WSN is a network of autonomous sensors that send their data through the network to a central location (Minerva et al., 2015). An IoT system can use a WSN for the collection of data in a lot of applications, but not every IoT system will use one since there are many other possibilities. The collection of data is only the first step of an IoT system, this data must be analyzed and transformed into valuable information or shared with other objects. The sensors used for the WSN provide the possibility to make any object smart and the huge progress in these sensors is probably the main innovation that started the IoT evolution. Another concept related to IoT is ambient intelligence; an environment with sensing and computing abilities, which can interact with humans. This concept is different from IoT since it only supports some predefined capabilities in a closed environment (e.g., a room, a building), is focused on human interaction and the used objects don’t necessarily have to be connected to each other (Miorandi et al., 2012). This is different from IoT because an important aspect of IoT is minimizing human input. Machine-to-machine (M2M) communication is a simplified version of IoT. The focus of M2M is on connecting devices and provide the possibility to remotely access data from these devices. This data is processed in a service management application to achieve productivity gains, reduce costs, and increase safety or security (Holler et al., 2014). The data isn’t integrated into other processes, only takes place on the level of the machines because the machines don’t necessarily have to be connected to a cloud platform. It’s more a direct, one-way form of communication. Data in IoT applications comes from

7

heterogeneous objects in many formats and is then integrated without human intervention, this is different in M2M applications. IoT can support the same services as M2M but has much more capabilities because data in IoT applications can be used for other purposes thanks to the web-based technologies (Holler et al., 2014). A schematic overview of all the different concepts is provided in Figure 1. 

Cloud computing and edge computing are technologies used for an IoT system.



An IoT system uses smart objects. This isn’t the same as smart devices, these are just devices connected to the internet and equipped with an interface for some predefined capabilities to interact. However, a smart object can offer the same capabilities as a smart device so that’s why the structures overlap in the overview. When the used objects are sensors connected to the internet, this is called a WSN. These structures also overlap because not every smart object is a WSN. As mentioned by Miorandi et al. (2012), smart objects don’t necessarily have sensing capabilities.



The same explanation applies for ambient intelligence and M2M. An IoT system can offer the same capabilities as these concepts but IoT is more than that.



IoT and CPS are two different concepts but there is an overlap. A CPS application will use IoT technology in most cases, for the collection of data and the interaction of smart objects. But the focus of CPS is on the actuation of objects, which isn’t the case for IoT.

Smart Objects WSN

Cloud/Edge Computing

Smart Device

Indirect interaction

IoT M2M

Cyber Physical

Systems Ambient Intelligence

Direct interaction Figure 1 Concepts related to IoT

8

2.1.4 Example The following example is based on a video of IBM Social Media (2010) and illustrates future applications as well as the difference between IoT and CPS from my point of view: My alarm clock goes off at the right time in the morning because I have set a meeting in the calendar of my smartphone and the alarm clock looked up which ferry I need to take to be there on time. The alarm is much louder today because the sensors in my mattress were indicating that I was in a sound sleep. In the meanwhile, the bathroom heater was turned on half an hour before my alarm clock went off and the coffee machine is preparing my coffee. My car is already running because the weather forecast indicated freezing overnight and the ice on my windshield should be removed (that’s why my alarm went off five minutes early so I have time to scrape off the rest). When I leave my house, every unnecessary device (including the lights) is turned off and the door locks behind me as I walk onto my driveway without the single push of a button. The GPS in my car already set an alternative route because there was a traffic accident but there’s no need to rush since I got an audio message in my car that the ferry was arriving five minutes late. The interaction in this example happens mostly between smart objects. The user’s input is minimized. There is no need for human confirmation every time a piece of information is processed. The actions of the smart objects are autonomous, unlike what happens mostly today. In current applications, most actions of a smart device are controlled by user input. In IoT, all the relevant data from one object is used by another object which in its turn creates data that can be used by other objects and so and so on. This illustration is a CPS application since the focus is on the actuation of objects: 1. Alarm clock goes off

4. Car started running

2. Bathroom heater turned on

5. Unnecessary devices are turned off

3. Coffee machine starts

6. Door locks

All these actions are controlled automatically by a system that enables the interaction and data exchange of the heterogeneous smart objects, while the data is analyzed in a cloud structure. This controlling system is an IoT application. When using the term IoT in the following sections, this refers to the concept defined and illustrated in this section.

9

2.2 BUSINESS MODELS As explained in section 2.1.1 and 2.1.2, IoT can be described as a concept that aggregates different technologies. The importance of business models for new technologies can be illustrated by a quote of Chesbrough. “A mediocre technology pursued within a great business model may be more valuable that a great technology exploited via a mediocre business model” (Chesbrough, 2010, p. 1). Since the foundation of IoT is the use of different technologies, this statement also applies for IoT. To follow the advice of Chesbrough, this entire section is devoted to the use of business models for IoT. The Business Model Ontology (Osterwalder, 2004) is used as business model framework for this dissertation, this is explained in section 2.2.2. Other business modeling frameworks are compared to this framework in section 2.2.3. This is followed by a discussion on the impact of IoT on business models (Section 2.2.3). In Section 2.2.5, IoT is described as an ecosystem of partners and the current challenges for IoT are mentioned in Section 2.2.6.

2.2.1 Business models in academic literature At the moment, there still doesn’t exist a consensus on a framework for business models. Every framework has a certain level of detail and focusses on different aspects. In their review of academic literature on business models; Massa, Zott, and Amit (2010) explain that researchers still haven’t reached an agreement on a definition for a business model since it’s used for different concepts: “Throughout our review, we have shown that the business model has been used to address different concerns in different contexts and in different management areas. Scholars have used the same term (i.e., business model) to explain and address different phenomena such as ebusiness types, value creation or value capture by firms, and how technology innovation works.” (Massa et al., 2010, p. 22) However, Massa et al. (2010) did also conclude business models are now widely recognized as a tool for analyzing how firms do business and business models try to explain the value creation and capturing of a company. So in my own words: business models can be useful to examine how firms do business and create value but it’s important to make sure the business model framework addresses the right concept in the context it is meant for.

2.2.2 Business Model Ontology One of the papers reviewed in the research of Massa et al. (2010) is ‘The business model ontology—A proposition in a design science approach’, by Osterwalder (2004). The Business Model Ontology (BMO) (Osterwalder, Pigneur, & Tucci, 2005) was the theoretical basis used to create the Business Model Canvas

10

(Osterwalder & Pigneur, 2010), the most famous and widely used technique for the representation of a business model. In the review, Osterwalder’s BMO is classified as a business model for e-Business. This means the use of information technology in organizations and internet-based business. Since IoT fits this classification, this technique is applied in the right context to describe the business model of companies that use IoT. Osterwalder’s definition for business models is the following: “A business model is a conceptual tool that contains a set of elements and their relationships and allows expressing the business logic of a specific firm. It is a description of the value a company offers to one or several segments of customers and of the architecture of the firm and its network of partners for creating, marketing, and delivering this value and relationship capital, to generate profitable and sustainable revenue streams.” (Osterwalder et al., 2005, p. 10) This definition mentions the BMO’s nine building blocks for business models. These are listed in the following table: Pillar

Business Model Building Block

Product

Value Proposition Target Customer

Customer Interface

Distribution Channel

Relationship

Value Configuration Infrastructure Management

Core Competency

Partner Network

Cost Structure Financial Aspects Revenue Model

Description Gives an overall view of a company’s bundle of products and services. Describes the segments of customers a company wants to offer value to. Describes the various means of the company to get in touch with its customers. Explains the kind of links a company establishes between itself and its different customer segments. Describes the arrangement of activities and resources. Outlines the competencies necessary to execute the company’s business model. Portrays the network of cooperative agreements with other companies necessary to efficiently offer and commercialize value. Sums up the monetary consequences of the means employed in the business model. Describes the way a company makes money through a variety of revenue flows.

Figure 2 Nine Business Model Building Blocks. Adapted from “Clarifying business models: origins, present, and future of the concept,” by Osterwalder, A., Pigneur, Y., & Tucci, C. L., 2005, Communications of the Association for Information Systems, 15(1), p. 10. Copyright [2005] by the Association for Information Systems.

The famous Business Model Canvas can be used as a visualization tool for these building blocks.

11

Figure 3 Business Model Canvas. Reprinted from Business Model Generation. A Handbook for Visionaries, Game Changers, and Challengers (p.44), by Osterwalder, A., & Pigneur, Y., 2010, John Wiley & Sons. Copyright [2010] by Alexander Osterwalder.

Some building blocks were rearranged in the Business Model Canvas. The value configuration building block is divided into a key resources and key activities component. The core competency building block is absorbed by the key activities component. The other building blocks only got a new name as shown in Table 1. ‘Business Model Generation. A Handbook

‘Clarifying business models: origins,

for Visionaries, Game Changers, and

present, and future of the concept’ by

Challengers’ by Osterwalder & Pigneur

Osterwalder et al. (2005)

(2010)

Target Customer



Customer Segments

Distribution Channel



Channels

Relationship



Customer Relationships

Value Configuration



Key Resources

Core Competency



Key Activities

Partner Network



Key Partners

Revenue Model



Revenue Stream

Table 1 Comparison of the building blocks in the Business Model Canvas and The Business Model Ontology

12

The journal article of Johnson, Christensen, & Kagermann (2008) and the journal article of Morris, Schindehutte, & Allen (2005) are the other two most popular2 pieces of academic literature with frameworks to describe business models, besides the publication of Chesbrough & Rosenbloom (2002). In the following section, these frameworks and the similarities with Osterwalder’s BMO are illustrated. The paper of Gassmann, Frankenberger, and Csik (2013) is also discussed since this paper is used in section 2.2.4.

2.2.3 Other frameworks for business modeling These three other frameworks can also be used for defining the characteristics of a business model, but in my opinion, they only use fewer components (building blocks) or they are just formulated differently. The first framework comes from an article published in Harvard Business Review’s must-reads on strategy. In ‘Reinventing your business model’ by Johnson et al. (2008), four components are used to construct a business model: Customer Value Proposition, Profit Formula, Key Resources and Key Processes. This is illustrated in Table 2. ‘Reinventing your business model’ by Johnson et al. (2008)

‘Clarifying business models: origins, present, and future of the concept’ by Osterwalder et al. (2005) Product

Customer Value Proposition Customer Interface Profit Formula Key Resources Key Processes

Financial Aspects Value Configuration

Table 2 Comparison of Johnson's et al. (2008) and Osterwalder's et al. (2005) business model framework

This framework discusses three of the four pillars of the BMO (Product, Customer Interface and Financial Aspects). However, the Customer Interface pillar is less detailed and included in the Customer Value Proposition next to the Product pillar. The Financials Aspects pillar corresponds to the Profit Formula component. The Infrastructure Management pillar is less detailed since it only includes the Value Configuration building block (BMO) and is divided into a separate Key Processes and a Key Resources component.

2

Based on the number of citations on Google Scholar

13

The second framework is created by Gassmann, Frankenberger, and Csik (2013) in their research paper on the basic patterns of business models. A business model is described using four dimensions: the Who, the What, the How, and the Value.

Figure 4 Business model definition – the magic triangle. Reprinted from The St . Gallen Business Model Navigator (p.3), by Gassmann, O., Frankenberger, K., & Csik, M, 2013. Copyright [2013] University of St. Gallen, Institute of technology management.

This corresponds to Osterwalder's description of a business model: “The business model is an abstract representation of the business logic of a company. And under business logic I understand an abstract comprehension of the way a company makes money, in other words, what it offers, to whom it offers this and how it can accomplish this.” (Osterwalder, 2004, p. 14) Only the part of financial viability (Value?) wasn’t included in this description but was mentioned in the Financial Aspects pillar of the BMO. Since Osterwalder’s BMO is based on this definition, this framework is a less detailed version of the BMO. The pillars have another name but the underlying building blocks aren’t described (Table 3). ‘The St . Gallen Business Model Navigator’ by (Gassmann et al., 2013)

‘Clarifying business models: origins, present, and future of the concept’ by Osterwalder et al. (2005)

What?

Product

Whom?

Customer Interface

How?

Infrastructure Management

Value?

Financial Aspects

Table 3 Comparison of Gassmann's et al. (2013) and Osterwalder's et al. (2005) business model framework

14

The third framework comes from a research paper published in the Journal of Business Research by Morris, Schindehutte, and Allen (2005). They created a less popular framework for characterizing the entrepreneur’s business model around the same time period as Osterwalder et al. (2005). The biggest similarities between this paper and the publication of Osterwalder et al. (2005), can be found where they define business models on three different levels. Osterwalder’s first type of business model is called the Overarching Business Model Concept or a level-1 business model and is placed on the highest hierarchical level. This type of business model is the BMO. It’s a metamodel for business models that defines which elements are needed for a business model. This corresponds to the foundation level in the research of Morris et al. (2005). In this research, they define six components to construct a business model compared to Osterwalder’s four pillars (Table 4). ‘The entrepreneur’s business model: Toward a unified perspective’, by Morris et al. (2005)

‘Clarifying business models: origins, present, and future of the concept’ by Osterwalder et al. (2005)

How will the firm create value?

Product

For whom will the firm create value?

Customer Interface

What is the firm’s internal source of advantage?

Infrastructure Management

How will the firm position itself in the marketplace? How will the firm make money? What are the entrepreneur’s time, scope, and size ambitions?

 Strategy Financial Aspects  Strategy

Table 4 Comparison of Morris's et al. (2005) and Osterwalder's et al. (2005) business model framework

These six components are formulated as questions. These questions are very general and should help companies for developing a business model (Morris et al., 2005). There are also some strong similarities between the questions and Osterwalder’s four pillars. Only question four and six are defined as strategy by Osterwalder. The next level of Osterwalder is Taxonomies or level-2 business model and the proprietary level for Morris’s research. On this level, the common characteristics of a real-world business model are described (the level-3 business model/ Rules level). It’s a type of business model that has some similarities with other companies (e.g. the Razor and Blade business model). This is mostly what is meant when the term business model is used in general. When the general term ‘business model’ is used in this dissertation, this refers to the level-2 business model. The lowest level of business models is called Instance Level or

15

level-3 business model by

Osterwalder et al. (2005)

Osterwalder and the Rules level in Morris’s research. This level

Overarching (Level-1) Business Model

Morris et al. (2005)

Foundation level

explains how a company worked out the building blocks or questions of the highest level in practice. For example,

Taxonomies (Level-2)

Proprietary level

Instance (Level3)

Rules level

how Gillette implemented the Razor and Blade model. As mentioned, it’s my opinion that all these concepts are very similar to each other. The BMO framework is most Figure 5 Comparison of business model levels of Osterwalder, widely used in practice in the form of Pigneur, & Tucci (2005) and Morris, Schindehutte, & Allen (2005). the Business Model Canvas when

Adapted from “Clarifying business models: origins, present, and future of the concept,” by Osterwalder, A., Pigneur, Y., & Tucci, C. L.d .,

talking about e-business and the 2005, Communications of the Association for Information Systems, generation of new business models. 15(1), p. 5. Copyright [2005] by the Association for Information Since IoT fits the e-business context and

Systems.

this technique addresses the concept it is meant for; describing, transforming and developing new business models. The use of the BMO framework as a tool for describing business models in this dissertation is justified.

2.2.4 Impact of the Internet of Things on business models There are three ways to look at business models and IoT. On the one hand, are the companies that will implement IoT in other companies or intervene somewhere in the creation of IoT systems. They solely use IoT technology as their main value proposition and IoT is used to create straightforward applications (e.g. predictive analytics). On the other hand, there are the currently used business models. These will be transformed by the impact of IoT. The third perspective is the creation of completely new business models. These three perspectives are comparable to the three phases in a study of the McKinsey Global Institute on IoT. The authors predict three phases for suppliers in the IoT technology market, in comparison with the evolution of personal computers and the Internet in its earlier days (Manyika et al., 2015). The first phase is characterized by the domination of companies that produce hardware and infrastructure, the building blocks of the technology. For IoT, this means sensors, cloud computing, connectivity infrastructure, data storage, etc. The next phase involves the creation of core services and software that use a broad platform. In the personal computer and internet evolution, this meant Google and Yahoo’s search engine. In the

16

third phase, new business models based on the technology will dominate the market. Examples from the personal computer and internet evolution include companies like Amazon, Uber, and Airbnb. Where each transition to the next phase involves an increase in the total value of the IoT market (Manyika et al., 2015). Phase 1 According to Manyika et al. (2015), we are currently situated between the first and the second phase. Nevertheless, in these early phases, IoT already offers great opportunities to create value. Maciej Kranz, vice president Strategic Innovations Group at Cisco and previously general manager of the Connected Industries Group at Cisco, a business unit focused on the Internet of Things, has already seen the following paybacks for companies deploying IoT at this stage (these companies were mostly situated in the B2B area) (Kranz, 2016): 

Reduction in costs



Design and creation of new business models



Revenue generation



Implementation of new go-to-market strategies



Increased uptime



Development of new product and service delivery options



Streamlined business processes



More efficient ways to service and support customers



Speedier service and delivery



New insight into product usage and customer information

Now focusing on (one of) these paybacks could already offer different possibilities for companies to create interesting value propositions. For companies that succeed in building a complete business model to support these value propositions, IoT technology could turn out to be a lucrative opportunity. This is the so-called first phase. Phase 2 Based on the research of Gassmann et al. (2013) on business model patterns; Fleisch, Weinberger, and Wortmann (2014) identified two business models patterns that could be commonly used alongside the usage of IoT technology when the IoT technology market evolves to phase two. These business model patterns are based on the existing business model patterns of Gassmann et al. (2013). The digitally charged products business model pattern is quite basic. Digital services are linked to physical products to create new value propositions. Fleisch et al. (2014) define six transformed business models based on the idea of the digitally charged products (e.g. Add-on, Freemium, Pay per Use, etc.). However, their view on IoT is quite different from the definition formulated in section 2.1.2. When using the definitions of related IoT concepts from section 2.1.3, their focus is more on the use of smart things with M2M communications. Fleisch et al. (2014) describe their business models from a single smart thing perspective. The integration of different smart objects isn’t an important aspect in their business models.

17

It’s mainly one smart object with a fixed digital service offered in different forms. The type of service determines the business model. This is quite superficial and in my view, not the true value of IoT. These are business models for smart devices, with applications that provide an interface for the user, to make use of services for the device. The second business model pattern is called Sensor as a Service. The collected data from products can be sold to external interested parties. The value proposition isn’t a product or service, but the data itself. This business model is based on the existing Software as a Service (SaaS) business model. The provider in this model offers on-demand information processing services to the user (Ma, 2007). This business model results from the use of cloud computing in applications. Instead of merely selling the software, the provider offers the use of its IT infrastructure, hardware and all supporting support services (e.g. maintenance, installation, data backup, security) to deliver applications on demand for the user through a network (Ma & Seidmann, 2008; Mell & Grance, 2011). The user only pays a fee for the use of this service. I’ve found similar predictions for variations of existing business models in other research articles. Manyika et al. (2015) see three possible groups of business models: new pricing models, service-based business models and monetization of IoT data. With new pricing models, they mean extensive use of pay-per-use models. The service-based business models are the same as the Digitally Charged Products business models and monetization of IoT is a different name for the Sensor as a Service business model pattern. Based on the seven laws of information from Moody and Walsh (2002), Bucherer and Uckelmann (2011) present four exemplary business model scenarios for the Internet of Things. The value proposition for the business models is always based on information as a value creator (raw data as well as aggregated or processed data). These four business model scenarios are similar to the pay-per-use model, the Sensor as a Service business model pattern and two of the specific digitally charged products business models (Remote Usage and Condition Monitoring; and Product as Point of Sales). A remark about the research of all these authors (in phase two) is that it was mainly focused on the revenue model and the used value propositions don’t fit the definition of IoT used in this dissertation. However, there were only a limited number of papers that identified business models that could be used by companies with IoT value propositions. Phase 3 It’s clear it’s already difficult to predict the impact of IoT on business models. Since IoT itself is still in such an early phase it’s not possible to predict which new business models will be created in the future.

18

2.2.5 IoT as an ecosystem of partners An author with a more similar perspective on IoT is Maciej Kranz. Kranz (2016) is more orientated towards the B2B-environment and argues that it’s not possible for one company to develop a complete IoT solution for every customer. You need partners that bring their expertise and other capabilities into the complete solution. Kranz offers a template of how an IoT ecosystem of partners could look like in Figure 6.

End device and sensors

Analytics applications

Visualization applications

End-device components

Cloud infrastructure and services

Management and applications enablement software

IT Infrastructure

Security

Service providers

Embedded operating systems

IT infrastructure

Vertical products and solutions

Figure 6 IoT Ecosystem. Adapted from Building the Internet of Things Implement New Business Models, Disrupt Competitors, Transform Your Industry (Emerging IoT Ecosystem section, Chapter 3), by Kranz, M., 2016, Wiley. Copyright [2017] by Maciej Kranz.

Other authors that share this opinion are Chan (2015); Westerlund, Leminen, and Rajahonka (2014); Turber, Vom Brocke, Gassmann, and Fleisch (2014). These authors suggest using a different tool than Osterwalder's & Pigneur's (2010) Business Model Canvas since this is focused too much on an individual company’s objectives and thus unable to provide a framework to develop business models for companies that use IoT in their value proposition. This framework isn’t suited for the ecosystem nature of IoT. However, Westerlund et al.(2014) don’t present a new framework, they only explain what it should look like and Chan (2015) and Turber et al. (2014) have a framework that has never been used in practice or other research. This means when using the Business Model Canvas for IoT companies with products or services based on IoT technology, it’s vital to extensively pay attention at the partner network building block because of the need for an IoT ecosystem.

19

2.2.6 Challenges for IoT Now that IoT and business models are discussed sufficiently in the previous sections, the challenges concerning this topic can be explained. There are three main challenges that still must be conquered for IoT to be successful. The first challenge is a technical one. The interoperability of smart objects, the main feature of IoT, still needs to be worked out. All connected objects from different manufacturers should operate fluently to be valuable, this is also important for the exchange of data. This will require open standards and industry-wide interoperability (Kranz, 2016). It’s clear this will take some time, but in the words of Maciej Kranz: “The industry knows how to do it; we did it for the first stage of the Internet and for the cloud. The current task at hand is even bigger and more complex, but I know that the IoT community is up for the challenge.” (Kranz, 2016, Key obstacles section, Chapter 1) There are a lot of possibilities with IoT but this concept can’t completely prove its value until the objects can properly interact with each other. Most companies will possibly try to avoid the collaboration with other companies for as long as possible. IoT hasn’t proven its value yet and collaboration on interoperability poses the risk of sharing trade secrets and other intellectual property. However, as mentioned by Kranz (2016) collaboration and partnerships will be imperative for IoT to succeed. The second challenge is the role of the government. Government regulations are obviously needed in some areas, but too much regulatory restrictions have to be avoided to leave some space for companies to innovate their business model (Kranz, 2016). The problems with Uber and the taxi sector have shown it’s not always easy to find the right balance between new business models and regulations from the past (Schellevis, 2014). The government also has a role to play for concerns about privacy and security. As mentioned, security in an IoT environment will be a difficult assignment since there is a trade-off between the fluent interaction of different smart objects and preventing unauthorized access of these objects to sensitive data. On the other hand, privacy and security are still a concern today in the world of smartphones, social media, and other applications while these industries seem to operate quite profitably. The third and last challenge concerns the subject of this entire section. IoT is an interesting technological evolution, but it is only relevant if it provides value for its users. This is a big technological change with a lot of opportunities, but it’s not easy to find a good value proposition and the right business model to support it. Big Data for example is a concept that’s been around for some time now and companies are still having some difficulties trying to figure what to do with it.

20

2.3 HEALTH CARE No one will contest that health care is a complex sector. There are a lot of different stakeholders with their own objectives and the structure differs in every country since it’s mostly determined by extensive government regulations (De Cleyn et al., 2015). Since there are such big differences between health systems, only the health system in Belgium is discussed; followed by general trends and problems in health care. Section 2.3.3 describes the use of business models in health care. The potential role for IoT as a technological enabler of disruptive innovation (Christensen, C. M., Grossman, J. H., & Hwang, 2009) is explained in section 2.3.4.

2.3.1 The health care system in Belgium The complexity of this sector is illustrated in Figure 7 on page 23, which gives an overview of the health system in Belgium. In the following paragraphs, a simplified description of the Belgian health system is given. Only the general structure and the most important financing activities are discussed since the political activities and the organization of all the public authorities isn't relevant for this dissertation. The following explanation is completely based on ‘The Health Systems in Transition’s report’ on Belgium’s health system by Gerkens & Merkur (2010). The main stakeholders in the health-care system are the government, sickness funds, health-care providers and patients. The Belgian health system is based on the principle of social insurance. The federal government is responsible for the regulation and financing of the mandatory health insurance; the financing of hospital budgets and heavy medical equipment (e.g. CT and MRI scanners); and the registration of pharmaceuticals and their price control. The mandatory health insurance is mainly financed through subsidies from the federal government and social security contributions. The social security contributions are related to income. The sickness funds are non-profit-making, non-commercial organizations and in charge of the mandatory health insurance. Their major task is the reimbursement of health service benefits to their members. To fulfill this task, they receive a budget from the government to finance the health-care costs of their members. All individuals entitled to health insurance must join or register with a sickness fund. Usually, the sickness funds also offer additional private insurance packages to get a refund of all medical costs that aren’t covered by the mandatory health insurance.

21

Health-care providers can be divided into two groups: 

Health institutions such as hospitals, homes for the elderly, nursing homes, day centers, laboratories, outpatient clinics, etc.



Health professionals such as physicians, dentists, psychiatrist, physiotherapists, pharmacists, nursing practitioners, nursing auxiliaries, midwives, paramedical practitioners, etc. Health professionals are generally organized as self-employed professionals (except nurses and midwives).

The main payment mechanism is the fee-for-service model. There are two systems for payments: 

Direct payment, where the patient pays for the full cost of the service and then obtains a reimbursement from the sickness fund for part of the expense.



A third-party payer system, where the sickness fund pays the provider directly and the patient is only responsible for paying any co-payments, supplements or non-reimbursed services.

Generally, the direct payment system applies to ambulatory care and the third-party payer system applies to inpatient care and pharmaceuticals. As outlined in section 2.2.6, the role of the government and a solid business model will be important concerns for companies developing an IoT application. This challenge will be even greater for companies that will use their application in health care. The government is a dominant factor in this sector and as will be explained in section 2.3.3, developing a supporting business model with the outdated fee-forservice model is still a problem for many organizations. In the following section, the recent trends in the health-care sector are discussed to get an idea where the opportunities for IoT applications are.

22

Figure 7 Overview of the health system in Belgium. Reprinted from “Belgium: Health system review,” by Gerkens, S. & Merkur, S., 2010, Health Systems in Transition,12(5), p. 17. Copyright [2010] by Sophie Gerkens and Sherry Merkur.

23

2.3.2 Recent trends in health care The most popular topic concerning the health-care system in Belgium, but also in the US, UK, etc. is the continuous rise in health expenditure. This is caused by a variety of factors. The health-care sector is characterized by significant innovations in terms of medical technology, treatments, etc. However, the cost aspect of these innovations remains a serious problem in this sector. In 2013, the share of gross domestic product allocated to health spending in Belgium was 10.2%, compared to the OECD3 average of 8.9% (OECD, 2015). The three problems where IoT can have the biggest impact are the aging population, increase in patients with a chronic disease (e.g. asthma, diabetes, obesity, etc.); and the general continuing inefficiency of this system.

Aging population

Problems in health care

Cost increase chronic diseases Inefficiency administration procedures

Figure 8 Problems in health care

Islam et al. (2015) discuss the use of IoT for remote health monitoring of patients with a chronic disease. These patients need frequent follow-ups for their condition. Reducing the need for face-to-face visits in the doctor’s office is one way of countering this rise in medical costs. Using an IoT application, doctors can check-up on patients by simply looking at the data of their patient’s vitals. A similar application can be used to support the elderly. By monitoring senior citizens and providing emergency notification systems, it allows them to live at home independently for longer periods (Geng, 2017). Reducing the number of admissions in homes for the elderly can be another way of reducing costs.

3

The Organisation for Economic Co-operation and Development (OECD), an intergovernmental economic organization with 35 member countries

24

There already exist a Belgian startup that created an application in support of this problem, the Zembro (2017) bracelet is a wearable that functions as a personal alarm system for the elderly. The inefficiency problem in health care is mostly situated in the administration procedures. More specifically the interoperability between health-care actors is a concern for a lot of countries. In Belgium, the first step in solving this problem was the launch of the digital platform eHealth in 2008. This platform gives the different stakeholders secured digital access to all health information and applications at a central location to avoid duplication of data (Gerkens & Merkur, 2010). A second initiative enrolled in a lot of countries is the Electronic Health Record (EHR). The EHR is used to standardize information exchange between patients. Since recently, Google (Google Fit), Microsoft (Microsoft HealthVault) and Apple (Healthkit) now offer a web-based service for personal health records. This gives patients the control back of their own health data. They are able to store, manage and easily share their health data (medical history, allergies, vaccinations, test results, insurance information, etc.) with the necessary health-care providers (Steinbrook, 2008). This trend is called patient engagement or citizen health empowerment (De Cleyn et al., 2015). Patient and family engagement can be defined as: “Patients, families, their representatives, and health professionals working in active partnership at various levels across the health-care system— direct care, organizational design and governance, and policy making— to improve health and health care” (Carman et al., 2013, p. 224). The aggregation of all data in a simplified overview could be an interesting value proposition for an IoT application to support this patient engagement. The third popular trend in the health-care industry to increase efficiency is value-based health care. Sickness funds in a lot of countries are trying to shift the basis for reimbursement from the number and type of procedures to the quality of care provided. However, this change is happening slowly because reducing volume (and increasing value) cuts into short-term profits of the health-care providers (Kaiser & Lee, 2015) and this requires significant organizational changes. An interesting follow-up question is: ‘What Is Value in Health Care?’. This is exactly the question Michael Porter tried to answer in a Journal Article in The New England Journal of Medicine (Porter, 2010). According to Porter (2010), value should be defined around the customer, in this case, the patient. And value is defined as the health outcomes achieved relative to the costs incurred. In a complex sector like health care, this has some difficulties: “The proper unit for measuring value should encompass all services or activities that jointly determine success in meeting a set of patient needs. These needs are determined by the patient’s medical condition, defined as an interrelated set of medical circumstances that are best addressed in an integrated way.” (Porter, 2010, p. 2478)

25

Outcomes in health care are very specific for every medical condition, involves multiple interdependent sources of care and can take a long time to become visible. However, according to Porter, the chart in Figure 9 can be used to measure the outcomes of any medical condition. The incurred costs should contain the total amount necessary for the entire process to treat a patient’s medical condition instead of the cost of individual services. This includes the costs for the use of shared resources. These costs have to be assigned to individual patients, based on the actual resources used to deliver the care to the patient (Porter, 2010). Of course, when health-care providers take the lead to deliver more value to their patients, the other stakeholders (government and sickness funds) should make sure they follow and align their main tasks (reimbursements and regulations) to create a complete system of value-based health care. Other trends that illustrate the ongoing concerns in health care include the focus on prevention instead of treatment and a holistic approach to treating patients. The current health system is still focused on illness while prevention is a much more cost-effective approach and the same is true for a holistic approach to treat patients (Deloitte, 2017). Including factors such as social and mental variables in the treatment helps to eliminate the root causes of a patient’s medical condition. IoT can be very useful for these trends. For prevention, predictive analytics can be used. The aggregation of heterogeneous sources of data from the patient will allow a holistic approach. The relevant trends in health care that offer opportunities for IoT are summarized in Figure 10.

26

Figure 9 The outcome measures hierarchy. Reprinted from “What is value in health care?,” by Porter, M., 2010, New England Journal of Medicine, 363(26), p. 2479. Copyright [2010] by Massachusetts Medical Society.

27

Digital platform Electronic Health record

Hollistic approach

Trends in health care Patient Engagement

Prevention

Value-based health care

Figure 10 Recent trends in health care

2.3.3 Business models in health care Gerkens & Merkur (2010) didn’t discuss the private companies that produce goods and services (biotechnology, pharmaceuticals, medical devices, etc.) for the health-care system in their review. However, in my opinion, they are the fifth stakeholder. While these companies are known for their product innovations, business model innovations have been quite scarce (Christensen, C. M., Grossman, J. H., & Hwang, 2009). To overcome the challenges mentioned in the previous section, these companies will need to become more successful in the use of the business model concept rather than just pushing their innovations forward towards consumers. The value-based health-care and patient engagement concept can be used as a starting point towards business model innovation, but so far, companies in this sector have failed to implement it successfully. According to Christensen, C. M., Grossman, J. H., & Hwang (2009); there are two important factors that could explain this phenomenon. Namely the extensive regulations and the reimbursement systems of the complex health-care sector. Van Limburg et al. (2011) share the same view on this subject, a lagging legislation and financial complexity are mentioned as reasons why this sector is so resistant to change. Other reasons mentioned by van Limburg et al. (2011) are a deficiency of publications that illustrate how business models can be created in health care and the lack of business-like thinking in general. This is a difficult topic in this sector. In their research, they specifically focus on the importance of using businesses models when developing technologies in the health industry. The authors believe this is the main reason why currently, new technologies often create extra side processes instead of replacing old processes

28

characterized by inefficiency. This statement confirms that Chesbrough’s quote from section 2.2 is also relevant for technology in health care and highlights the importance of business models in health care.

2.3.4 IoT in health care Finding the right business model for IoT applications is one of the current challenges for IoT, the combination with health care will only make this challenge more difficult. IoT is still an unknown area and health care is a complex sector with a lot of government involvement, which makes business model innovation in this sector difficult. However, if companies can successfully create business models to support their IoT application, two of the three enables for disruptive innovation are present (Christensen, C. M., Grossman, J. H., & Hwang, 2009)(Figure 11). IoT is a technological enabler and business model innovation is the second enabler. The third enabler is a value network and defined by the authors as “a commercial infrastructure whose constituent companies have consistently disruptive, mutually reinforcing economic models” (Christensen, C. M., Grossman, J. H., & Hwang, 2009, p. xx).

Technological enabler

Regulations and standards that facilitate change Value network

Business model innovation

Figure 11 Elements of disruptive innovation. Adapted from The Innovator's Prescription: A Disruptive Solution for Health Care (p. xx), by Christensen, C. M., Grossman, J. H., & Hwang, J., 2009, McGraw-Hill. Copyright [2009] by Itzy.

Of course, one company alone can’t completely disrupt health care. A good interaction with the different care providers and the insurers is needed such that a value network exists through which care is delivered. This is also called a commercial ecosystem by the authors, which corresponds to the need for an IoT ecosystem as explained in section 2.2.5. So again, the partner network building block will be an important aspect when developing a business model for companies with an IoT application in health care. Besides these three elements, the authors mention the need for regulatory reforms and new industry standards to facilitate change in the new disruptive industry. This is similar to the first and second challenge for IoT, mentioned in section 2.2.6.

29

Chapter 3 METHODOLOGY Synopsis The research design is a qualitative, exploratory case study. The cases are three Flemish startups: Partheas, Healthy and Ectosense. An overview of these companies is given alongside a description of their application. Semi-structured interviews were used to collect the data since this provides some advantages for the research. The analysis of the data is based on the interview guide which offered an overview to discover differences, similarities and trends in the interviews. To finish this chapter, some remarks about the reliability and validity of the research are discussed.

30

3.1 RESEARCH OBJECTIVE The initial objective of this research was to see how the business model of startups in healthcare with a value proposition based on IoT looks like. The current objective is to identify the difficulties, challenges and restrictions startups with an IoT value proposition in healthcare in Belgium are experiencing. By exposing the similarities, differences and trends between the answers of three startups. This will give an overview of the current progress of these startups in developing an IoT application.

1.2 RESEARCH DESIGN Based on the structure of Saunders, Lewis, & Thornhill (2015); the research design consists of the following three parts: 1) Methodological choice: qualitative research design 2) The purpose of your research design: exploratory study 3) Research strategy: case study

3.1.1 Qualitative research The methodological choice is straightforward since the subject of the research question is qualitative in nature and there are only a few companies that use IoT applications specifically in health care because it’s still an unknown domain, which limits the possibilities for quantitative research.

3.1.2 Exploratory study The purpose of the research design is to gain insight into difficulties companies are having with developing an IoT application in health care. This will require open questions because IoT is a recent phenomenon and it’s unclear how companies in this setting currently operate since there are only a limited number of them. This exploratory perspective has the advantage that it’s possible to change direction, based on new insights while conducting research (Saunders et al., 2015), which turned out to be necessary.

3.1.3 Case study An effective research strategy to conduct an exploratory study (in business research) is a case study (Dul & Hak, 2008). Yin (2013) argues that a case study is a preferred strategy in situations where: 1) The main research questions are “how” or “why” questions. 2) The researcher has little or no control over behavioral events. 3) The focus of the study is a contemporary phenomenon.

31

For this research, these situations are present. The applied case study strategy will be based on multiple cases such that it’s possible to compare the findings of these cases to each other. This can clarify which findings are unique and which are common across the startups (Bryman & Bell, 2011). The time horizon of this research is a “snapshot” taken at a particular time, this makes it a cross-sectional study according to Saunders et al. (2015).

3.2 CASE STUDY SUBJECTS The subjects of the cases are three Flemish startups. There are several reasons why Flemish startups are chosen as the research population. The first reason is based on the insights of Christensen, C. M., Grossman, J. H., & Hwang (2009). The authors explain that in the history of disruptive innovations in health care, the technological enablers mostly emerge from laboratories of leading institutions, while the business model innovations were created by new entrants to the industry. The second reason comes from my own experience that most startups are more willing to cooperate to a dissertation of a student than bigger companies. Flemish startups are used due to the practical reasons for doing the interviews and because health-care systems are very different in other countries which makes it difficult to compare foreign startups in this sector. My search for these startups was based on health tech startups in general and involved different channels: 



IoT conventions 

Internet of Things Convention Europe (http://iot-convention.eu/nl/home/)



The IoT Solutions World Congress (http://www.iotsworldcongress.com/)

Health-care conventions 





HEALTH&CARE (http://www.health-care.be/nl?ReturnUrl=%2fnl)

Incubators 

Start it @ KBC (http://startit.be/)



The Birdhouse (http://www.gobirdhouse.com/)

Start-up communities 

Startups.be (http://data.startups.be/dashboards/home)



Healthstartup (http://www.healthstartup.eu/)



Referrals of other startups



Imec, a research and innovation hub in nanoelectronics and digital technologies. I had a meeting with dr. Ann Ackaert and Frederic Vannieuwenborg. Two researchers at imec in the domain of digital health. They made some suggestions on possible startups.

32

The three startups used for the case study are the companies with the closest fit to the definition of IoT presented in section 2.1.2. This comparison was based on the information on the company website. The fit with the definition was based on terms such as: ‘information/data from multiple sources’, ‘third-party sensor input’, ‘wearables’, ‘aggregation/integration of data’, ‘sensors’, ‘algorithms’, ‘cloud-based’, ‘new technology’, ‘IoT’, ‘RFID’, ‘platform’, ‘mobile’. In the following section, a brief description of the three startups is given.

3.2.1 Company information

Partheas is a Flemish IT company, founded by a group of experienced IT and business consultants in 2012. The company developed a solution (Partheas Flow) to manage the physical flow of patients in the ambulatory care in 2014. The application uses RFID-tags to trace the patients throughout the hospital. Based on historical data and the planned appointments, a waiting time is calculated and communicated to the patient using monitors. The patients are automatically called in and guided towards the consultation room based on an algorithm that considers different parameters: ‘Is the patient present in the waiting room?’, ‘Does the patient have multiple appointments?’, ‘Does the patient need urgent care?’, etc. The application is already (partially) implemented in several hospitals in Belgium. The interviewee was Koen Wauters, managing director at Partheas. http://www.partheas.com/

Healthy, the second startup, preferred to be anonymous. That’s why a fictional company name is used to refer to this startup. Healthy is a Flemish startup founded in 2015. The idea for founding the company started with a survey of health-care providers. The results indicated there is an urgent need for a solution of the following three problems: too many administrative tasks, insufficient time to inform colleagues about patients’ status and a lack of real-time patient data to provide accurate care. Healthy is currently focused on care homes for the elderly and is developing a mobile service platform to address these issues. Using mobile devices, the application takes over a big part of the administration of the care personnel by giving an overview of the patient’s file, real-time data and the remaining operational tasks. Healthy is also working on an application that automatically detects incidents with patients. The interviewee was the product manager of the company.

33

is a Flemish startup based in Leuven and founded in 2015. The company created a scalable solution for the diagnosis of sleep disorders, especially sleep apnea. This solution consists of a very small wearable that sends sensor data and sound recordings to a mobile device; and a cloud software platform that analyzes this data to provide a diagnostics report. A beta version of the software is already available and the hardware component will be available in the fall of 2017. The interview was with Bart Van Pee, the CEO. http://www.ectosense.com/

3.3 DATA COLLECTION Semi-structured interviews are used as data collection technique. For each start-up, one of the founders of the company was interviewed in their office. They possess the necessary information of the company to answer the questions in the interview. The interviews were recorded to make sure the focus was on listening to the interviewee and asking the right questions. These recordings were transcribed afterward. A personal interview is used because the person of interest will more likely agree to be interviewed, rather than by telephone.

3.3.1 Semi-structured interview Semi-structured interviews are a useful technique in this research design since this is a flexible interview process. A list of questions on specific topics is prepared, this list is called an interview guide, but this doesn’t mean the interviewer will have to ask all these questions and follow the initial order of these questions (Bryman & Bell, 2011; Saunders et al., 2015). The interviewee can speak freely so it’s possible some questions are already answered in other questions and a better opportunity can arise to ask followup questions which were initially structured somewhere else. Based on the answers of the interviewee, the interviewer has the possibility to ask questions that were not included in the interview guide. In the end, the interviewer must make sure all the necessary information is answered such that the results can be compared. The interview guide is divided into five parts. Firstly, an introduction with general questions about the interviewee and the company, followed by a part on business models that is based on the nine building blocks of the BMO (Osterwalder et al., 2005). There are seven main questions, each is based on a building block and then probing questions are provided in case a more detailed description is needed (the value configuration and core competency building block are addressed in the follow-up questions of the partner network). The next two parts contain questions about health care in Belgium and IoT. The last part has two concluding questions. Some questions were added based on the outcome of the previous interview(s). An overview of the question framework can be found in Appendix 1 on page 58.

34

3.4 DATA ANALYSIS The initial approach was to make use of coding to analyze the transcriptions of the interviews. However, this wasn’t necessary since the data was already properly structured. Some patterns were already clear after listening to the interviews twice. Furthermore, the three interviews were all transcribed and reread in a short period, the interviewees answered the questions to the point and the interview guide provided a good supporting structure. These factors facilitated the search for differences, similarities and trends across the interviews without coding the data.

3.5 RELIABILITY AND VALIDITY OF THE RESEARCH Ellinger, Watkins, & Marsick (2005) summarize the findings of Yin (2003) on validity and reliability when designing and conducting case study research. There are four criteria to assess the quality of the research design: 

Construct validity concerns defining the correct measures for the concepts that are studied.  The practical guide ‘Business Model Generation’ by Osterwalder & Pigneur (2010) was used to the make sure the questions in the interview can be used to get a view of the business model of the cases. However, the overall objective of this exploratory research doesn’t concern measuring a phenomenon. The outcome of this research is unknown. Based on the information from the interviewees, the objective is to discover similarities, trends and differences between the startups.



Internal validity is about the accuracy of a causal relationship between two variables.  This issue isn’t for relevant for an exploratory case study (Ellinger et al., 2005).



External validity concerns the ability to generalize the findings of this research to other relevant settings.  This is a problematic concept in case study research and even more so in this research because it’s conducted in a very specific setting (Flemish startups with an IoT application in health care) with a small sample size. As explained by Saunders et al. (2015), by providing a full description of the research question, research design, context, results and interpretations, the reader is provided the opportunity to judge the generalizability of the study to another setting.

35



Reliability relates to whether the used data collection techniques and analysis would produce the same results if they were repeated on another moment in time or by a different researcher (Saunders et al., 2015).  The use of more than one interviewer and data analyst would reduce errors and biases in this study, however, this isn’t possible in this dissertation.

36

Chapter 4 RESULTS Synopsis The findings of similarities and differences between the startups are discussed in this chapter. The results of the interviews are illustrated with questions and extracts from the answers. The following eight trends were found: 1. The applications of the startups don’t fit the proposed definition of this concept. Their application can be categorized as one of the related concepts of IoT. 2. The value propositions of the startups are all specialized. Either in terms of target customers or in number of capabilities. 3. The development of a self-learning algorithm is a serious challenge. 4. The startups now don’t have meaningful partnerships with other companies, they are focused on the software and the feedback of the end-user. 5. Privacy and security aren't a difficulty. 6. The use of models (including the business model canvas) as a tool differs between the interviewees. 7. Government regulations have a strong influence on this sector. 8. Health care is a difficult sector for the implementation of (new) technology.

37

4.1 THE INTERNET OF THINGS FROM A FLEMISH ENTREPRENEURIAL PERSPECTIVE The interviewed companies see possibilities with IoT but their application doesn’t fit my definition of this concept. They are also rather skeptical when it comes to other companies that mention the use of IoT in their applications on their website. In their experience, there are a lot of companies that give the impression they’ve created an IoT application. However, in reality, these companies are still working out the idea of an IoT application or they tested a proof-of-concept, but they are far from an application that is ready for use. The numbers behind the quotes are used to refer to the startups: Partheas [1], Healthy [2] and Ectosense [3]. Have you ever heard of the Internet of Things? ‘Yes, of course. It’s a buzzword that we sometimes use ourselves. However, our application doesn’t use this kind of technology … In the future, we should be able to use this technology to further develop our processes, however, at the moment we can’t do much with it.’ [1] How would you describe the Internet of Things? ‘Sensors with smart analytics is what it should be. But at the moment it’s only a label used for marketing purposes. I’m rather skeptical when someone uses this term to describe their product.’ [3] Do you see more opportunities for your company with this technology? ‘… at the moment, it is still unclear in many cases how to connect different data sources to each other and how to get useful information out of it.’ [2]

4.2 SPECIALIZED APPLICATIONS A second observation is that all three startups currently have a specialized application. They are either specialized in terms of targeted customers or provide an application with not too many different capabilities. All three of them believe it is best to start off with a specialized service/product to create a solid business instead of pushing a large product into the health-care market with capabilities that aren’t absolutely necessary for their customers.

38

You are now focusing on patients with a very specific disorder, are you planning on expanding the application to other disorders as well? ‘There are other disorders that we could detect using the same data but this isn’t our objective at the moment. During the next five years, the market of this specific disorder will be worth four billion euro so we want to build a solid business in this market first.’ [3] During my literature review, I have noticed that a lot of startups are currently working on a platform for the communication between health-care providers because health care in Belgium is still very fragmented. Are you also looking to develop a solution for this problem? ‘Yes, but this will be at a later stage. We first want to make sure that we have a product that works on its own. From our experience with previous projects, we learned that you can’t just push one big product into the market and expect that all health-care providers will start using it together. That’s why we first try to create a very good service for just one health-care provider...’ [2] How do you offer the service to your customers and how does the payment take place? ‘…our intention is to first place our product into the market to generate sales and traction.’ [1]

4.3 THE DEVELOPMENT OF SELF-LEARNING ALGORITHMS Two of the three startups are currently working on a self-learning algorithm that can provide additional value to their customers in the near future. However, this turns out to be quite the challenge. The other startup already uses a clinically proven, not self-learning, algorithm for diagnostics. Have you also built the software for the data analytics yourselves? ‘… we are currently working on a project to analyze all the data we receive from our customers and get information out of it by using a self-learning algorithm to improve our customer’s processes even further...’ [1] Can you tell me something more about your product, how does it work? ‘… we are currently building a third part of our product. This should become an analytics dashboard.’ [2] Do you use the data you receive from the customer for other applications or is it only used in the current product/service? ‘In the long term, we want to use the data we collect from our customers to make predictions about patient’s health. But to make this application work, we first need larger flows of data, of course.’ [2]

39

4.4 COLLABORATION WITH OTHER COMPANIES The three startups don’t really collaborate with other companies. One of the startups has to outsource the testing of their product since these are clinical studies but the relation with this company doesn’t go any further than that. The startups are very focused on the software aspect of their application, which they developed completely on their own. They either outsource the development of the hardware components or let the customer decide which supplier is used for the hardware components. They sometimes give recommendations but this is only based on the technological needs of their software. As long as they can implement their application using these hardware components, it doesn’t matter who the supplier is. This is also the case for the maintenance of the system. The supplier is responsible for the hardware and the startup takes care of the software component. The most significant collaboration is based on the feedback of their customers and the end user. Two of the three startups make use of government programs and incubators for coaching about the health-care sector and developing a business in general. How are the sensors being produced and do you have agreements with suppliers? ‘We completely outsourced the design and initial production of the sensor.’ [3] Have you collaborated with other companies or research centers for the development of the company? ‘All the intellectual property is created by our company but we outsource the clinical studies that are needed to test and benchmark our product because laboratories have the necessary infrastructure to perform these studies…’ [3] Do you see opportunities to work together with other companies to further expand your services? ‘Not for the moment.’ [3] What are the main costs for the company? ‘… another part is the software development. We do the analytical part of the intellectual property completely ourselves (the algorithms) but we outsource the development of the web portal, apps, etc.’ [3] How do you collaborate with other companies when you use their hardware? ‘We don’t have fixed hardware partners. Our interest is the software ...’ [1]

40

4.5 PRIVACY, SECURITY AND DATA MONETIZATION At the moment, there aren’t any serious problems for the startups regarding data security or the privacy of their users, the patients of their customers. Securing their application isn’t more complex than it is for other companies working with apps, wearables and data transfer over a (wireless) network. The only difficulty is that some hospitals refuse to work with a cloud-based implementation so everything has to be implemented on the hospital’s internal network. The startups are aware that privacy is a sensitive issue in health care but there are clear regulations that explain what companies have to do when they receive data from their users. None of the startups offer data to other organizations. So you use the data mainly internally and not for external purposes? ‘Yes, definitely … I think strict regulations are on the way for 2018 so we are not planning on offering data to external parties.’ [2] How do you deal with the privacy and security of the data?  ‘We took on a lawyer who is responsible for these complex matters.’ [2]  ‘Hospitals are correctly very cautious when it comes to their patient’s privacy. When we want to implement our application using a cloud system, the hospitals almost always refuse to use this system. Even though it would be more beneficial for us and the hospital in terms of cost. For the time being, we use the internal hospital network to implement our application so we don’t have any issues regarding the privacy of the patients…’ [1]  ‘… of course, you make sure that the data of patients isn’t just available and sufficiently protected. Certain employees aren’t allowed access to all the data, this is also embedded in the quality management system. This is typical for this sector, it’s a challenge but not really a problem.’ [3] Your system is cloud-based, is this a problem for the rest homes when it comes to privacy of their residents? ‘We want to make health care mobile. With mobile we mean using mobile devices, that’s why we need a cloud solution. The legislation for hospitals is much stricter than it is for health-care centers. At the moment, we don’t have any problems with privacy and hopefully, it stays that way.’ [2]

41

Do you use the data you receive for other applications as well or are you planning on selling this data to other parties? ‘I have never seen someone successfully doing data monetization outside its own company. Data is worthless when it isn’t enriched with information and I don’t think another party would pay us for this kind of data. It’s only useful to improve your own algorithms…’ [3]

4.6 MODELING TOOLS IN HEALTH CARE The interviewees have a different view on the use of the Business Model Canvas. According to one interviewee, this model has been of use for him. But for the other two, this didn’t provide any value. One correspondent even questions the suitability of this tool in the health care sector. The interviewees also have opposite views about the use of models in general. Two correspondents see the use of these models as a tool to think about the company’s business model, competitors, and strategy. According to another interviewee, these models don’t add much value when thinking about your business and strategy when you have the size of a startup. Have you used models like for example the Business Model Canvas to get a good overview of the company?  ‘Yes, we’ve already used the Business Model Canvas … This was actually quite useful to look at things in a systematic way and improve the positioning of the company.’ [1]  ‘Yes we have, although models like that aren’t really relevant for applications in a health-care environment. There are other models out there that are better suited for health care… Health care is very complex and has some inherent difficulties that are related to the market. For example, the decoupled buyer-user decision system. The physician will be the one who decides for the patient, but he will not be the buyer for that patient since this is the insurer’s responsibility. That’s why the Business Model Canvas doesn’t apply here.’ [2]  ‘Yes, we often see this in obligated workshops. We realize we should think about things like suppliers, competitors, value proposition, etc. But in my opinion, if you can’t figure this out by yourself, a Business Model Canvas won’t help you that much further.’ [3] Do you use other tools to help set out the strategy of the company such as SWOT, Balanced Scorecard and Porter five forces?  ‘Yes, I think it’s a must to look at SWOT occasionally. The Balanced Scorecard can be useful, but you need a bigger company and much more projects for that. This is less useful for us at the time since we’re still a startup.’ [1]

42

 ‘Yes, this mainly on the request of investors. Apart from that, I think these tools are very useful to get a good overview of our strategy and the future of the market we are in. However, when you’re busy, these tools are easily forgotten.’ [2]  ‘No nothing. I think that these models are only useful for companies that need a high-level perspective and have a revenue that exceeds five million euro. When this is the case, I think it’s important to think about your competitors but as a startup, your main concern should be being fast.’ [3]

4.7 BUSINESS MODEL INNOVATION IN HEALTH CARE At the moment, the startups don’t see much possibilities for business model innovation in health care. They think the current system in Belgium limits their options. Some changes are necessary to increase the feasibility. Where are the biggest possibilities for technology in health care? ‘The problem in digital health care today is that most innovative ideas get support, but there isn’t a framework and financial incentive system from the government for health care.’ [2] Do you do this to make a reimbursement possible? ‘In our model, we do not consider the use of reimbursement because this would slow down the process significantly…’ [3] Do you have anything to add to this interview? ‘… I think that the influence of the government partly guides your business model towards a certain way of doing business. Suppose we could charge patients for an extra fee because we give them more information. This would be very convenient for us. Of course, this isn’t possible at the moment and not at all our intention to do so…’ [1] Do you think that for example in the US, there are more possibilities in terms of business model because health care is largely privatized there? ‘Yes, I think so … Because health care and, in particular, the hospitals are financed and set up differently than an ordinary company, the implementation of a system like ours is not really encouraged. If our system could be used by companies in other sectors, the implementation of our system would go much quicker since we have such a good ROI. The system in the US works completely different. This doesn’t mean that I am an advocate of the privatization of hospitals,

43

but this means that systems like ours are being implemented faster. Whereas in Belgium they only try to do this since recently.’ [1]

4.8 TECHNOLOGY IN HEALTH CARE The startups explained it’s difficult to convince their customers to implement their application. They mentioned the health-care sector, in general, is a difficult market and the implementation of technology goes very slow. The startups can present a good ROI but the purchase procedures of health-care providers are rigid. Could you introduce the company briefly? ‘…currently, all documentation about patient care is still written on paper and then entered into a computer manually…’ [2] How do you offer the application to these customers?  ‘To sell an application like this one, you have to visit the customer more than five times...’ [1]  '... it's a new product and the hospital sector is not really known to invest quickly in this kind of applications because it also requires a lot of adjustments from the hospital. We drastically change their way of working. … The sale cycle in health care is very slow ...’ [1] During my literature review, I have read that the use of new technologies often advances with difficulty in health care. Are you also confronted with these difficulties? ‘The implementation is very difficult. The adoption is generally easier; with adoption, I mean that the end user will use our product. We have fully developed our product while consulting the end users so this will not be a problem…’ [2] How big do you estimate this market? ‘It’s a very difficult ad slow market…’ [2]

44

Chapter 5 DISCUSSION Synopsis This chapter is about the interpretation of the results and a reflection on the literature review to answer the research question: ‘Which factors affect the development of IoT in healthcare in Belgium?’. While giving an insight into how an IoT application in healthcare in Belgium looks like today. Using the definition for IoT from the literature review, none of the applications of the startups can be classified as an IoT application. When looking at the related concepts, the application of Partheas can be described as ambient intelligence and the application of Ectosense and Healthy as M2M applications with a smart device that uses cloud computing. The term IoT is now used a marketing tactic to create the impression this technology is already available. The interviewees think that more advanced applications (IoT applications) will become available at some point, but this will require some time. The development of self-learning algorithms turns out to be a technological challenge at the moment. This limits their capabilities, so they are focusing value proposition on a specific problem in health care. A difficulty for the startups are the inherent characteristics of the sector. It’s a complex market where technology is implemented slowly and the government is a dominant aspect. This pushes the business model of the startups in a certain direction, which limits the innovation possibilities. The size and the specific regulations of the Belgian Market are also an important factor here. An ecosystem of partners now isn’t important for the startups. The feedback of the customers and enduser is most important. The interviewees confirmed the importance of the value creating aspect when thinking about their service application since it’s not possible to just push it into the market and see what happens. The use of models (including the Business Model Canvas (Osterwalder & Pigneur, 2010)) for thinking out the strategy and business model differs between the interviewees. An overview of the findings of potential contributing factors to the development of IoT in healthcare in Belgium is given at the end of this chapter.

45

5.1 THE INTERNET OF THINGS IN HEALTH CARE IN FLANDERS The interviewed startups don’t have an application that fits my definition of IoT. When using the definitions of related concepts from the literature review, the system of Partheas can be categorized as ambient intelligence. Using RFID-tags and monitors, the hospital becomes a sensing environment with an algorithm that computes the expected waiting time. This system partly automates the flow of patients inside the hospital (closed environment) with some predefined capabilities. The application of Ectosense is a diagnostic smart device connected to a smartphone which sends data to a cloud platform where it is analyzed. This satisfies a big part of the IoT definition. However, the data comes from one smart object, there is no integration of data and the data flows mostly in one direction. This makes it a M2M application with a smart device that uses cloud computing. This is also true for the application of Healthy. The data from the patients is processed in a mobile service application that uses cloud computing. But the data isn’t integrated into other processes and is mainly oriented towards human interaction. These are useful and interesting applications, but not the advanced IoT applications that were envisioned in the literature review. Based on the comments of the interviewees, the impression is given that an IoT application that fits my definition isn’t available at the moment in health care in Belgium [Result 1]. The startups that were chosen for the interviews had the most similarities with the proposed definition from the literature review. Nevertheless, it’s difficult to prove this assumption since this selection was based on the information of the startups’ websites. The search for startups was based on health tech and not solely on IoT since it’s possible that companies don’t use this term to present their application, even when it is an IoT application according to the proposed definition. However, the sample of interviews is limited and the search was extensive but not exhaustive since some startups will not publicly show their application if it’s not finished yet. At the moment, the term IoT can be used as a marketing tactic. This tactic is called ‘vaporware’; advertising a product that isn’t available yet or doesn’t even exist. Companies use this tactic to create the impression IoT technology is already available to have a rapid adoption when they’ve finished their development (Schilling, 2012). This is possible since there doesn’t exist a strict definition of IoT. The interviewees confirmed there are some opportunities to use IoT applications in health care but there are still some technological restrictions. Still, they believe more advanced applications will become available in the future, but it will take more time than what is generally assumed.

46

5.2 TECHNOLOGICAL RESTRICTIONS The first challenge for IoT explained in the literature review, concerns the interoperability of smart objects (Kranz, 2016). This difficulty wasn’t confirmed in the interviews, but another challenge was identified. The development of (self-learning) algorithms turns out to be much more difficult in practice than what is mentioned in the literature [Result 3]. The use of (self-learning) algorithms in IoT applications is commonly mentioned, but the challenges for developing these sophisticated algorithms are rarely explained in the reviewed literature (Holler et al., 2014; Manyika et al., 2015; Vermesan et al., 2015). One of the possible reasons could be the lack of available data. The entrepreneurs are confronted with some sort of "Chickenand-egg" problem. To have a good working self-learning algorithm, you need a lot of data, otherwise, it isn’t valuable. Their customers don’t gather a lot of data since they can’t do anything with it but the startups need information to feed it into their algorithm. The more data it gets, the better it becomes. Another important factor is that these algorithms must be clinically approved in this sector. The developers of these algorithms are still figuring out how they can turn different sources of data into valuable information which from my point of view, is the crucial part of an IoT application.

5.3 FOCUSED APPLICATIONS At the moment, the startups are focusing their value proposition on a specific problem in health care [Result 2]: 

Improving the life of patients with a chronic condition.



Simplify/automate administrative tasks of the care staff.

Johnson et al. (2008) declared precision is the most important feature of a value proposition and the administration in health care was indicated as one of the three problems where IoT can have the biggest impact in the literature review. Nevertheless, these two categories of applications are so-called lowhanging fruits, a service or product that can be sold easily (Investopedia, 2017). The startups are aware their current strategy has an expiration date and are trying to develop additional software services to improve their offerings. The problem is that these additional services need self-learning algorithms. A possible future scenario is that companies will continue to grow their services and expand their specialized value proposition to other problems in health care. This will take some time to do this successfully because a sufficient user base will be necessary to obtain enough data and further improve the service. If however, companies can succeed in this step, the real possibilities of IoT will surface. And in the future, the integration of these applications or the continuous extension of their service could result

47

in an actual IoT application that fits the definition from the literature review. As explained in the review, these more advanced value propositions will require proper business models in support.

5.4 HEALTH CARE AND GOVERNMENT REGULATIONS IN BELGIUM The statements in the literature review from Johnson et al. (2008) and van Limburg et al. (2011) about the health care sector were mostly confirmed for Belgium. It’s a complex market and technology is implemented slowly [Result 8]. The government regulations are an important factor here. The reimbursement systems and investment decisions are connected to the government and insurers. This creates an obstacle for business model innovation [Result 7]. The government is aware a serious change is urgent. That’s why in 2016 the Minister of Health, Maggie de Block started a pilot project to test a few health applications and mobile devices to determine which regulations, legal framework and remuneration models are necessary for health-care organizations. However, awaiting the results of this pilot project and the period until the actual changes are made can be fatal for the startups. That’s why for some startups, it could be more beneficial to direct their application towards foreign markets. The market in Belgium is quite small and the applications in health care need a lot of adjustments to comply with the different regulations in Flanders and Wallonia. In the literature review, security was cited as a serious difficulty since all the connected objects could now be a potential threat for unauthorized access (Miorandi et al., 2012). However, at the moment this isn’t the case for the startups [Result 5]. The startups take the same measures as other companies in an ITenvironment. The security of data probably isn’t a problem because there isn’t a lot of autonomous interaction between objects in their application and definitely not on a scale envisioned in the literature. The same is true for the privacy of the users. In the words of Bart Van Pee, one of the interviewees: "Privacy is a challenge but not a real issue”. In May 2018, the General Data Protection Regulation will become effective. This is a new European regulation about the use of personal data (European Commission, 2016). These regulations clearly state what companies can and cannot do with the data from their users. None of the startups offer data to other organizations. The data isn’t valuable for other parties when they have to comply with these privacy regulations. So now, data monetization isn’t a useful resource of revenue for them.

48

5.5 BUSINESS MODELS FOR IOT IN HEALTH CARE Building a supporting business model was defined as the third challenge for IoT in the literature review. Because a technological evolution is only relevant if it provides value for its users, this is similar to what Chesbrough (2010) mentioned in his quote. The Business Model Ontology of Osterwalder et al. (2005) was chosen in the literature review as a framework to represent the business model of the startups. However, describing the complete business model of the startups didn’t turn out to be a useful contribution from the answers in the interviews. The interviewees are aware it’s important to consider the value creating aspect when developing their service application. They confirmed the days of pushing any health-care product or service into the market are over. However, the use of models in general seems to be something personal, especially when the company has the size of a startup [Result 6]. The contribution of the Business Model Canvas from Osterwalder & Pigneur (2010), the business model visualization tool based on the BMO, during the development process was limited for some interviewees. To answer the initial research question, a brief explanation of the business models of the startups is given. The value proposition is already discussed in section 3.2.1; a description of all the other building blocks would be too extensive since this isn’t the objective. That’s why the description is based on the revenue model. The interviewed startups can be divided into two. Healthy and Partheas use a variation of the Software as a Service business model and are situated in phase 1 of the IoT technology market. Partheas charges their customers a licensing fee for the use of their software, based on the number of waiting rooms. Healthy sells their application with a set-up cost and then licenses its software based on the number of patients. Ectosense can be described using the digitally charged products business model pattern of Fleisch et al. (2014) since it’s a single smart object with M2M communication capabilities.

5.6 AN ECOSYSTEM OF PARTNERS An important subject of the literature review was the use of a (commercial) ecosystem of partners to launch an IoT application in health care (Christensen, C. M., Grossman, J. H., & Hwang, 2009; Kranz, 2016). However, collaboration clearly is not the most important aspect of the business model for the startups at this point in time [Result 4]. The interviewees are more focused on their customers and end user feedback while developing their application. For Healthy, a survey of care personnel was even the start of their company. This lack of collaboration could change in the future if they expand their capabilities and start to integrate other applications into their value proposition (this means moving towards the proposed definition of

49

IoT). Since this could create additional challenges for which they will need the help of other companies. This could be vital if they want to remain ahead of their competitors in the future. To conclude, an overview of potential contributing factors of the development of IoT in healthcare in Belgium is given in Figure 12. The green factors have a positive influence while the red factors slow down the development of IoT in healthcare in Belgium.

Self-learning algorithms Customer/ End user feedback

Complexity of the sector

IoT in health care in Belgium

Privacy and security

Belgian market

Technology implementation

Government regulations

Figure 12 Potential factors that influence the development of IoT in healthcare in Belgium

50

Chapter 6 CONCLUSION Based on the academic literature, IoT and related concepts were defined in the literature review to classify the applications of the startups. This classification exposed a serious gap between the reality in this sector in Flanders and the expectations for IoT from the literature. This created the research opportunity to identify potential factors that influence the development of an IoT application in health care This dissertation identified seven potential factors. These were found by conducting case studies of three startups in Flanders. Two factors are linked to the technology; four factors are based on the healthcare market in Belgium and the other factor relates to the business model. Firstly, self-learning algorithms are more difficult in practice than what is assumed in the literature. The startups are still figuring out how to transform data into valuable information using this concept. Secondly, privacy and security aren’t more difficult for these applications than it is for others. The third and fourth factor are the complexity of the health care sector and the accompanying government regulations. This makes it a real challenge for the startups to develop an innovative business model to support their application. Especially the missing revenue framework is of importance here. This is thoroughly discussed in the literature, but not many results have followed. Fifthly, the startups confirmed that health care is still a difficult sector for technology. The implementation goes slow which has a negative influence on the progress of their operations. The sixth factor is the Belgian market for health tech and potentially has a similar effect. The market is small and the application needs different modifications to comply with the different governmental regulations. The last factor concerns the use of tools to set out the strategy of the company and developing a supporting business model. This was different for the interviewees but was considered crucial in the literature review. Customer and end-user feedback are used as the most important attribute to create a value-adding application. The combination of these factors could influence the progress of these startups in developing an IoT application that fits the proposed definition. This provides an interesting research opportunity to confirm these factors in other startups and bigger companies in health care in Belgium. It could also be meaningful to see if some of these factors are also present for startups with an application in other sectors.

51

Chapter 7 LIMITATIONS AND FURTHER RESEARCH The limitations of this research are mainly the use of Flemish startups as case study subjects. Some of the encountered difficulties could be inherent to their size and structure. The difficulties that were identified are also partly influenced by the perspective of the entrepreneurs since they were the main source of information. It’s possible some of the influencing factors aren’t experienced by other entrepreneurs with a tech startup in health care. That’s why a further and deeper study of these factors in other startups and bigger companies is needed to confirm or contradict these proposals. It’s also important to mention that the interviews were conducted in Dutch and the data analysis was based on these transcriptions. The extracts were translated so it’s possible that some nuances may have gotten lost in translation which could be important for the interpretation of the results.

52

BIBLIOGRAPHY Ashton, K. (2009). That “Internet of Things” Thing. RFiD Journal, 4986. Retrieved from http://www.rfidjournal.com/articles/view?4986 Atzori, L., Iera, A., & Morabito, G. (2010). The internet of things: a survey. Computer Networks, 54(15), 2787–2805. Bryman, A., & Bell, E. (2011). Business Research Methods (3rd Revise). Oxford University Press. Bucherer, E., & Uckelmann, D. (2011). Business Models for the Internet of Things. Architecting the Internet of Things, (July), 253–277. http://doi.org/10.1007/978-3-642-19157-2 Cambridge Advanced Learner’s Dictionary & Thesaurus. (2017). The internet of things. Retrieved February 16, 2017, from http://dictionary.cambridge.org/dictionary/english/internet-of-things Carman, K. L., Dardess, P., Maurer, M., Sofaer, S., Adams, K., Bechtel, C., & Sweeney, J. (2013). Patient and family engagement: a framework for understanding the elements and developing interventions and policies. Health Affairs (Project Hope), 32(2), 223–231. http://doi.org/10.1377/hlthaff.2012.1133 Chan, H. C. Y. (2015). Internet of things business models. Journal of Service Science and Management, 8(4), 552. Chesbrough, H. (2010). Business model innovation: Opportunities and barriers. Long Range Planning, 43(2–3), 354–363. http://doi.org/10.1016/j.lrp.2009.07.010 Chesbrough, H., & Rosenbloom, R. S. (2002). The role of the business model in capturing value from innovation : evidence from Xerox Corporation ’ s technology spin-off companies. Industrial and Corporate Change, 11(3), 529–555. http://doi.org/10.1093/icc/11.3.529 Christensen, C. M., Grossman, J. H., & Hwang, J. (2009). The Innovator’s Prescription: A Disruptive Solution for Health Care. New York: McGraw-Hill. Davenport, T. H., & Lucker, J. (2015). Running on data. Deloitte Review, (16), 5–15. De Cleyn, S., De Vos, K., Kallstenius, T., Van Bruystegem, E., Van Daele, W., & Vermeulen, S. (2015). Citizen health empowerment. iMinds Insights, 40. Retrieved from http://www.iminds.be/en/gaininsights/iminds-insights/citizen-health-empowerment

53

Deloitte. (2017). 2017 Global Health Care Outlook. Retrieved from http://www2.deloitte.com/us/en/pages/life-sciences-and-health-care/articles/global-health-caresector-outlook.html Dul, J., & Hak, T. (2008). Case study methodology in business research (First edit). Elsevier. Ellinger, A. D., Watkins, K. E., & Marsick, V. J. (2005). Case study research methods. In R. A. Swanson & E. F. Holton (Eds.), Research in organizations: Foundations and methods of inquiry (First edit, pp. 327– 350). Berrett-Koehler Publishers. European Commission. (2016). Protection of personal data. Retrieved May 1, 2017, from http://ec.europa.eu/justice/data-protection/ Fleisch, E., Weinberger, M., & Wortmann, F. (2014). Business Models and the Internet of Things. Bosch IoT Lab White Paper, (SEPTEMBER), 19. http://doi.org/10.13140/RG.2.1.3824.2008 Gassmann, O., Frankenberger, K., & Csik, M. (2013). The St . Gallen Business Model Navigator. Retrieved from http://www.alexandria.unisg.ch/224941/1/St. Gallen Business Model Navigator.pdf Geng, H. (2017). Internet of Things and Data Analytics Handbook. John Wiley & Sons. Gerkens, S., & Merkur, S. (2010). Belgium: Health system review. Health Systems in Transition, 12(5), 1– 266. Gubbi, J., Buyya, R., Marusic, S., & Palaniswami, M. (2013). Internet of Things (IoT): A vision, architectural elements, and future directions. Future Generation Computer Systems, 29(7), 1645– 1660. http://doi.org/10.1016/j.future.2013.01.010 He, D., & Zeadally, S. (2015). An Analysis of RFID Authentication Schemes for Internet of Things in Healthcare Environment Using Elliptic Curve Cryptography, 2(1), 72–83. http://doi.org/10.1109/JIOT.2014.2360121 Holler, J., Tsiatsis, V., Mulligan, C., Avesand, S., Karnouskos, S., & Boyle, D. (2014). From Machine-toMachine to the Internet of Things: Introduction to a New Age of Intelligence. Retrieved from http://store.elsevier.com/product.jsp?isbn=9780124076846 IBM Social Media. (2010). The Internet of Thingss. Retrieved April 26, 2016, from http://www.youtube.com/watch?v=sfEbMV295Kk Investopedia. (2017). Low-Hanging Fruit. Retrieved May 1, 2017, from http://www.investopedia.com/terms/l/low-hanging-fruit.asp

54

Islam, S. M. R., Kwak, D., Kabir, H., Riazul Islam, S. M., Daehan Kwak, D., Humaun Kabir, M., … Kyung-Sup Kwak, K. S. (2015). The Internet of Things for Health Care : A Comprehensive Survey. Access, IEEE, 3, 678–708. http://doi.org/10.1109/ACCESS.2015.2437951 Johnson, M. W., Christensen, C. M., & Kagermann, H. (2008). Reinventing your business model. Harvard Business Review, 86(12). Kaiser, L. S., & Lee, T. H. (2015). Turning Value-Based Health Care into a Real Business Model. Harvard Business Review, 1–5. Retrieved from http://hbr.org/2015/10/turning-value-based-health-careinto-a-real-business-model Kranz, M. (2016). Building the Internet of Things Implement New Business Models, Disrupt Competitors, Transform Your Industry. Wiley. Ma, D. (2007). The Business Model of “Software-As-A-Service.” IEEE International Conference on Services Computing (SCC 2007), 701–702. http://doi.org/10.1109/SCC.2007.118 Ma, D., & Seidmann, A. (2008). The Pricing Strategy Analysis for the “Software-as-a-Service” Business Model. In International Workshop on Grid Economics and Business Models (pp. 103–112). Las Palmas de Gran Canaria, Spain: Springer Berlin Heidelberg. http://doi.org/10.1007/978-3-54085485-2 Manyika, J., Chui, M., Bisson, P., Woetzel, J., Dobbs, R., Bughin, J., & Aharon, D. (2015). The Internet of Things: Mapping the value beyond the hype. McKinsey Global Institute, (June), 144. Massa, L., Zott, C., & Amit, R. (2010). The business model: Theoretical roots, recent developments, and future research. IESE Business School-University of Navarra, 43. Mell, P., & Grance, T. (2011). The NIST Definition of Cloud Computing Recommendations of the National Institute of Standards and Technology. Nist Special Publication, 145, 7. Minerva, R., Biru, A., & Rotondi, D. (2015). IEEE- Towards a Definition of the Internet of Things (IoT). Miorandi, D., Sicari, S., De Pellegrini, F., & Chlamtac, I. (2012). Internet of things: Vision, applications and research challenges. Ad Hoc Networks, 10(7), 1497–1516. http://doi.org/10.1016/j.adhoc.2012.02.016 Morris, M., Schindehutte, M., & Allen, J. (2005). The entrepreneur’s business model: Toward a unified perspective. Journal of Business Research, 58(6), 726–735. http://doi.org/10.1016/j.jbusres.2003.11.001

55

OECD. (2015). How does health spending in Belgium compare? OECD Health Statistics 2015. Retrieved from http://www.oecd.org/belgium/Country-Note-BELGIUM-OECD-Health-Statistics-2015.pdf Openshaw, E., Hagel, J., Wooll, M., Wigginton, C., Brown, J. S., & Banergee, P. (2014). The internet of things ecosystem: Unlocking the business value of connected devices. Deloitte. Retrieved from http://www.deloitte.com/assets/Dcom-UnitedStates/Local Assets/Documents/TMT_us_tmt/us_tmt_IoTEcosystem_062014.pdf Osterwalder, A. (2004). The Business Model Ontology - A Proposition in a Design Science Approach. Retrieved from http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.134.8520&rep=rep1&type=p df%5Cnhttp://www.stanford.edu/group/mse278/cgi-bin/wordpress/wpcontent/uploads/2010/01/TheBusiness-Model-Ontology.pdf Osterwalder, A., & Pigneur, Y. (2010). Business Model Generation. A Handbook for Visionaries, Game Changers, and Challengers. John Wiley & Sons. Retrieved from http://books.google.com.au/books?id=fklTInjiPQAC&printsec=frontcover&dq=intitle:Business+Mo del+Generation+A+Handbook+for+Visionaries+Game+Changers+and+Challengers&cd=1&source=g bs_api%5Cnpapers2://publication/uuid/2DC52F55-F67F-429D-BC0D-E78D85AED7E7 Osterwalder, A., Pigneur, Y., & Tucci, C. L. (2005). Clarifying business models: origins, present, and future of the concept. Communications of the Association for Information Systems, 15(1), 1–43. Porter, M. E. (2010). What is value in health care? New England Journal of Medicine, 363(26), 2477– 2481. http://doi.org/10.1056/Nejmp1011024 Saunders, M., Lewis, P., & Thornhill, A. (2015). Research Methods for Business Students (7th New ed). Pearson Education Limited. Schatsky, D., & Trigunait, A. (2011). Internet of Things Dedicated networks and edge analytics will broaden adoption. Deloitte University Press, 307–323. Schellevis, J. (2014). Belgische rechter verbiedt taxi-app Uber in Brussel. Retrieved February 7, 2017, from http://tweakers.net/nieuws/95410/belgische-rechter-verbiedt-taxi-app-uber-in-brussel.html Schilling, M. (2012). Strategic Management of Technological Innovation: Fourth Edition (Fourth). McGraw-Hill Higher Education. Steinbrook, R. (2008). Personally controlled online health data--the next big thing in medical care? The New England Journal of Medicine, 358(16), 1653. http://doi.org/10.1056/NEJMp0801736

56

Turber, S., Vom Brocke, J., Gassmann, O., & Fleisch, E. (2014). Designing business models in the era of Internet of Things: Towards a reference framework. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8463 LNCS, pp. 17–31). http://doi.org/10.1007/978-3-319-06701-8_2 van Limburg, M., van Gemert-Pijnen, J. E. W. C., Nijland, N., Ossebaard, H. C., Hendrix, R. M. G., & Seydel, E. R. (2011). Why business modeling is crucial in the development of eHealth technologies. Journal of Medical Internet Research. http://doi.org/10.2196/jmir.1674 Vermesan, O., Friess, P., Guillemin, P., Giaffreda, R., Grindvoll, H., Eisenhauer, M., … Tragos, E. Z. (2015). Internet of Things beyond the Hype : Research , Innovation and Deployment. Building the Hyperconnected Society - IoT Research and Innovation Value Chains, Ecosystems and Markets, 15– 118. Westerlund, M., Leminen, S., & Rajahonka, M. (2014). Designing Business Models for the Internet of Things. Technology Innovation Management Review, 4(7), 5–14. Retrieved from http://timreview.ca/article/807 Wikipedia. (2016). Smart device. Retrieved November 23, 2016, from http://en.wikipedia.org/wiki/Smart_device Yin, R. K. (2003). Case Study Research: Design and Methods. Applied Social Research Methods (Vol. 5). Sage publications. Yin, R. K. (2013). Case study research: Design and methods (5th Revise). Sage publications. Zembro. (2017). Zembro personenalarm. Retrieved May 1, 2017, from http://www.zembro.com/

57

APPENDIX 1 INTERVIEW GUIDE Introduction 1. Could you introduce yourself first by giving an idea of your background? a. Your education, previous employers? 2. Could you introduce the company briefly? a. When was the company founded? b. How was the company founded? c. How did the idea start? 3. What’s your position in the company? a. How long have you been working for this company?

Business Model 4. Which services/products does the company offer? Value proposition a. Which problem is solved by this application/What is the added value? b. When will it be available on the market? c. Does the company also implement the application completely? d. What are the main components of the product/service? 5. Does the company collaborate with other companies, research centers, etc.? Partner network a. How would you describe the importance of this collaboration? b. Is this collaboration essential for the development of the company? c. Does the company make use of suppliers? Value configuration/Core competency i. Which sensors are used? 1) Do you produce these sensors yourselves? ii. Which software does the company use? 1) Did the company completely develop this on their own? 6. Who are the targeted customers? Target customer a. Have you identified different customer segments? b. Are you planning on expanding to other customer segments? c. How big do you estimate this market? d. Is the company focused on Belgium or also on other countries?

58

7. How does the company offer the application to the customers? Distribution channel a. Through the website, vendors, etc.? b. Will this be the same for future customers? 8. Do you have contact with the customer after the initial transaction? Relationship a. For example for maintenance b. Does the company offer additional services? 9. What are the main costs for the company? Cost structure a. Licensing, personnel, etc.? 10. What are the main sources of revenue for the company? Revenue model a. How does the payment take place? 11. How is the company financed? a. Did you make use of incubators or other organizations to start the company? 12. Does the company use the data received from the customers for other applications or is it only used in the current product/service? a. Do you see possibilities to sell the data to other parties? b. Does the company use the data for improvement of the current application? c. Do you see more opportunities for the data in the future? 13. Have you used models to analyze the company? a. For example Business Model Canvas, Business Model Triangle, etc. 14. Have you used tools to determine the strategy of the company? a. For example SWOT, five forces of Porter, Balanced Scorecard, etc.? 15. How would you describe the business model of the company? 16. Who are the competitors of the company? a. Do you expect serious competition from multinationals for example? b. Is an acquisition a possible exit scenario?

Health care in Belgium 17. Does the company make use of a platform to centralize the information between patient and health-care providers? a. Did you develop this platform yourselves? b. Why or why not? 18. Do you collaborate with sickness funds to make reimbursements possible? 19. Have you tried to participate in the pilot project of Maggie De Block? 20. Is the company confronted with difficulties to launch the application?

59

a. For example by government regulations 21. How do you deal with the privacy and security of the data? 22. Are changes from the government necessary to facilitate the operations of the company? a. In terms of general regulations? b. Regulations in the healthcare sector?

Internet of Things 1. Have you ever heard of the Internet of Things? a. Did you hear about this term before starting the company? 2. How would you describe the Internet of Things? 3. Would you describe the application of the company as an IoT application? 4. Do you see (more) possibilities with this technology for the company? 5. How will technology evolve in health care? a. Which applications will be possible in the near future? b. Do you think collaboration with other organizations will be necessary to realize this evolution?

Conclusion 1. Do you have anything to add to this interview? 2. Did I forget to ask any questions?

60

Loading...

THE INTERNET OF THINGS IN HEALTH CARE

THE INTERNET OF THINGS IN HEALTH CARE Word count: 17.240 Janni Illegems Student number: 01202331 Supervisor: Prof. dr. Steve Muylle Commissioner: Ni...

2MB Sizes 0 Downloads 0 Views

Recommend Documents

The Internet of Things - ITU
Nov 16, 2005 - Machine-to-machine communications and person-to-computer communications will be extended to things, from

The Internet of Things - Cisco
However, as businesses, governments, standards bodies, and academia work together to solve these challenges, IoT will ..

Architecting the Internet of Things
specified subscription to content tailored to an individual's interests. We expect to retrieve personalised information,

From the Internet of Computers to the Internet of Things
to Internet services. An Internet of Things makes computing truly ubiquitous – a concept initially put forward by Mark

internet of things - Accenture
Apr 20, 2017 - Battle. DIGITAL CONTENT & IOT. SERVICES AGGREGATOR. Operate as Platform Provider of Ecosystem Services. I

Internet of Things - Ericsson
By producing what we in the project call the 'Internet of Things Comic Book' we wanted to explain IoT for everybody the

Extending the Internet of Things to the Future Internet - Hindawi
world one has to consider how legacy technologies would integrate into the IoT using IPv6. While it would be convenient

The Internet of Things in the Cloud - Home
2009. Year 2009. 1. China Premier. Wen Jiabao visited Wuxi. Internet of ings. R&D Center. 2. Mr. Wen Jiabao urged to est

The Internet of Things - Oliver Wyman
The Internet of Things, or IoT, is everywhere and the phenomenon is accelerating. Many industries will see profound effe

Securing the Internet of Things - NICS
R. Roman, P. Najera, and J. Lopez, “Securing the Internet of Things”, IEEE Computer, vol. 44, pp. 51 -58, 2011. http