UNIVERSITY OF VAASA 

SCHOOL OF MANAGEMENT 

 

 

 

 

 

 

 

 

 

 

 

Tarja Vuorela 

 

VALUE CO-CREATION AND POTENTIAL BENEFITS  

THROUGH BIG DATA ANALYTICS: 

 HEALTH BENEFIT ANALYSIS 

 

 

 

 

 

 

 

 

Master’s Thesis in 

Strategic Business Development 

 

 

 

 

 

 

VAASA 2018





1 

 

TABLE OF CONTENTS  

page 

 

LIST OF FIGURES AND TABLES 5 

ABBREVIATIONS 7 

ABSTRACT 9 

1. INTRODUCTION 11 

1.1. Background of the study 11 

1.2. Research gap 13 

1.3. Objectives and research questions 14 

1.4. Thesis structure 16 

2. DATA-DRIVEN VALUE CO-CREATION IN HEALTHCARE 17 

2.1. Value-based healthcare 17 

2.1.1. Health benefit analysis and care gap 19 

2.1.2. Individual patient’s and population health management 20 

2.2. Value and value co-creation in service systems 24 

2.2.1. Concept of value 24 

2.2.2. Service science and service-dominant (S-D) logic 26 

2.2.3. Healthcare ecosystem and value co-creation practices 28 

2.2.4. IT-enabled and data-driven value co-creation 32 

2.3. Big data 34 

2.3.1. Definitions of big data 34 

2.3.2. Identified challenges and critical success factors 39 

2.3.3. Taxonomy of analytics 42 

2.3.4. Big data analytics in healthcare 44 

2.4. Big data analytics-enabled transformation model 45 

2.4.1. Components of BDET model 47 

2.4.2. Big data-enabled transformation and value co-creations practices 51 

3. METHODOLOGY 54 

3.1. Research method 54 

3.2. Sampling and case selection process 56



 

  



3 

 

3.3. Data collection and analysis 57 

3.4. Validity and reliability 61 

4. EMPIRICAL FINDINGS 62 

4.1. Case: The Health Benefit Analysis tool 62 

4.2. Health Benefit Analysis tool explanatory variables 65 

4.2.1. Big data analytics components 66 

4.2.2. Big data analytics capabilities 68 

4.2.3. Summary of explanatory variables 71 

4.3. Affected or transformed value co-creation practices 73 

4.3.1. Meaningful use of electronic health records 76 

4.3.2. Evidence-based medicine 78 

4.3.3. Multidisciplinary cooperation 80 

4.3.4. Clinical resource integration 81 

4.3.5. Network collaboration 82 

4.3.6. Network knowledge creation 84 

4.3.7. Personalized care 85 

4.3.8. Effects in healthcare ecosystem actors 87 

4.4. Potential benefits and performance 88 

4.4.1. Identified benefits 90 

4.4.2. Value generated to stakeholders 95 

5. CONCLUSIONS 98 

5.1. Key findings 98 

5.2. Theoretical implications 103 

5.3. Managerial implications 105 

5.4. Suggestions for future research 106 

5.5. Limitations 106 

LIST OF REFERENCES 107 

APPENDIX 1. Examples of results generated with Health Benefit Analysis tool 115 

APPENDIX 2. Step by step description of health benefit analysis 118 

APPENDIX 3. Interview structure 120 





 5 

 

 

LIST OF FIGURES AND TABLES 

 

Figure 1. Three perspectives to potential benefits and value 15 

Figure 2. The structure of a value for money triangle 23 

Figure 3. Healthcare ecosystem 29 

Figure 4. The wide range of sources for big data 38 

Figure 5. Critical success factors for big data analytics 40 

Figure 6. The continuum of data to information to knowledge 41 

Figure 7. Big data analytics-enabled transformation model 46 

Figure 8. The extended big data analytics-enabled transformation model 52 

Figure 9. Resources, functionality, and outputs of the Health Benefit Analysis tool 67 

Figure 10. The most evident identified paths-to-value chains 99 

 

 

Table 1. Typology of co-creation practices and their indicative measures in healthcare 30 

Table 2. Summary of the seven big data V’s definitions and characteristics 35 

Table 3. IT-enabled transformation practices with examples 48 

Table 4. The benefit dimensions with examples of subdimensions 50 

Table 5. Information gathering through preliminary discussions 58 

Table 6. List of interviewees, time schedule and duration of interviews 59 

Table 7. Value co-creation practice sub-elements affected by using the HBA tool 73 

Table 8. Expected potential benefits and performance that generate value to stakeholders 89 



  



 7 

 

 

ABBREVIATIONS 

 

AI Artificial intelligence 

BDET Big data analytics-enabled transformation 

CDS  Clinical decision support 

EBMeDS Evidence-Based Medicine Electronic Decision Support  

EHR  Electronic health record 

EMR Electronic medical record 

IS Information system 

IT Information technology 

GPS Global positioning system 

HBA  Health Benefit Analysis 

ML  Machine learning 

NNT  Number needed to treat 

ODA Self Care and Digital Value Services 

PHR Personal health record 

PTB Potential to benefit 

RFID Radio frequency identification 

SAMK Satakunta University of Applied Sciences 

STAR Socio technical allocation of resources 

TUT Tampere University of Technology 





 9 

 

 

UNIVERSITY OF VAASA 

School of Management 

Author:      Tarja Vuorela  

Topic of Thesis: Value co-creation and potential benefits 

through big data analytics: 

 Health Benefit Analysis 

Name of supervisor:    Karita Luokkanen-Rabetino 

Degree:   Master of Science in Economics and  

Business Administration 

Major Subject:    Strategic Business Development 

Year of Entering the University:  2016 

Year of Completing the Master’s Thesis: 2018         Pages: 121 
 
ABSTRACT 

Big data analytics in healthcare context is often studied from a technical point of view. In the 

field of strategic management, researchers have indicated a research gap in how big data 

analytics create business value. This study examines how big data and advanced analytics 

generate potential benefits and business value for the healthcare service provider, and value 

for the individual patients and population health. In addition, the effects of advanced analytics 

to the value co-creation practices and actors in healthcare ecosystem are studied. The 

theoretical framework used for the purpose is the big data analytics-enabled transformation 

model which is adapted to answer the research questions. The study is conducted as a single 

case study. The studied case is the Health Benefit Analysis (HBA) tool. The empirical data 

is collected in eight semi-structured interviews with participants of the tool development 

project.  
 
Using the HBA tool reveals several paths-to-value chains. The most evident path shows how 

using advanced analytics affects the personalized care practice by enabling a more interactive 

service process between the health professionals and patients. It denotes a business scope 

redefinition as patients are now being interpreted as essential actors in the value co-creation 

of their own health outcomes. The benefits that arise from the advanced analytics are of 

several dimensions; operational, managerial, strategic, and organizational. Using the HBA 

tool generates strategic business value for the healthcare service provider as a differentiator 

that contributes to gaining competitive advantage compared to other service providers not 

using this innovation. Value emerges for the individual patient as improved patient 

experience and better health outcomes. Population health gains most value from the reduced 

health inequalities.  
 
The evolving value co-creation practices set requirements for the healthcare ecosystem actors 

as they need to conform to new practices with patients and other professionals from other 

sectors and levels of the ecosystem. The healthcare work and service culture need to develop 

and adapt to new tools, related processes, and a more diversified professional base, including 

health analysts and other new professionals. To conclude, it can be claimed that advanced 

analytics of healthcare big data contributes to the shift to value-based healthcare. 
 
KEYWORDS: value co-creation, big data analytics, value-based healthcare, health benefit  

analysis, healthcare ecosystem.  





 11 

 

 

1. INTRODUCTION 

 

Digitalization has rapidly become reality for many industries by disrupting old business 

models. New and enhanced value co-creation practices yield possibilities to offer increased 

value and various potential benefits for the business owners, individual customers, and entire 

customer segments. Digital transformation has also had its effect and changed the way of 

thinking and way of working in the healthcare industry. This change is still ongoing, and the 

development of digital healthcare services, related medical products and equipment, as well 

as the electronic information systems in the field, continue to evolve in the future.  

 

 

1.1. Background of the study 

 

According to Barlow (2016: 3, 13), the medical knowledge base is growing exponentially, 

since more data is collected about the patients and new medical information is published 

every day, which no single human being can keep up with, leading to a situation where 

doctors and other care staff need help to succeed with this highly complex field of life 

sciences. According to Obermeyer & Lee (2017: 1209), every patient is a “big data” 

challenge, as medical knowledge is expanding rapidly, and patients are older with more 

coexisting illnesses and medications. Further, Barlow (2016: 1 – 3) claims that the traditional 

labor-intensive healthcare transforms into more knowledge-driven and data-intensive 

practice where the newer healthcare delivery models depend on user-friendly, real-time big 

data analytics, artificial intelligence (AI) and machine learning (ML) tools, and that millions 

of individual patients may benefit from the improved capabilities on diagnosis and 

treatments. Also, according to Rose and Burgin (2014: 11), real-time big data analytics has 

the potential to enable well-timed interventions to get customers, i.e. patients the right care, 

at the right time, and in the right venue. This, in turn narrows the potential gaps in healthcare 

delivery, and in that means, generates potential savings by improving the operating 

conditions, as well as competitiveness of the healthcare service providers. The improved 

capabilities can become an asset on population level as well, for example with disease 

management and epidemics tracking, such as hotspots of Malaria, or by providing 

estimations of influenza activity as Google Flu Trends does (Sahay 2016: 420, 426; 

Raghupathi & Raghupathi 2014: 8). Moreover, it can help with ensuring the needed proactive 



 12 

 

 

care delivery and interventions for specific groups of patients suffering from similar health 

problems, e.g. heart failure or hypertension (Kunnamo 2017).    

 

To become one of the game changers and to contribute to the discussion, healthcare service 

providers should invest in investigating the value and potential benefits of big data and 

advanced analytics, for example how data analytics could enhance the offered healthcare 

services and ensure that these services optimally meet the patients’ needs and improve their 

health condition. From a business development point of view, it should be evaluated how the 

use of advanced analytics can create value for the actors in the healthcare ecosystem, as well 

as for the end customers, i.e. the specific populations and individual patients. 

 

Regarding the development of information technology in healthcare (E-Health) in the 

European Union, scholars have indicated some inequalities between the member states, as 

many of them, usually the richest ones, have been able to invest more in the development, 

while some countries have not. Therefore, the European Council urged in 2013 for the 

reduction of the digital gap in healthcare amongst the member states. (Quaglio, Dario, 

Stafylas, Tiik, McCormack, Zilgalvis, D’Angelantonio, Karapiperis, Saccavini, Kaili, 

Bertinato, Bowis, Currie & Hoerbst 2016: 314). The European Union (2016) also carried out 

a systematic literature review and consultation of experts to identify examples of the use and 

value of big data analytics in the practice of public healthcare and telemedicine, and to 

identify whether there is a need for policy recommendations to develop and support the use 

of big data in public health. This review also confirms that the increasing availability of data 

and technical progress combined with limited financial resources, stakeholders in public 

health as well as the scientific community are open to the opportunities offered by big data 

applications not only for the health of the individual but also for the health of the whole 

population. Moreover, the review indicates that the use of big data might improve also the 

performance and outcome of healthcare systems. (European Union 2016: 22, 25.)  

 

The most important lesson the European Union (2016: 55) learned in its review was that 

raising awareness of the added value of big data in health is needed quite urgently by 

stimulating a continuous open dialogue with all stakeholders and patient groups. 

Consequently, the public discussion around the major potential benefits and challenges of 

big data in health is in full flow and will continue in the coming years.  

 



 13 

 

 

In order to achieve any potential benefits and expected value from advanced and big data 

analytics, there are, however, many challenges on the way. These issues need to be studied 

and answered to, which makes the effects of data analytics to value co-creation practices in 

the field of healthcare an interesting research topic. Moreover, according to McKinsey & 

Company reports (Manyika, Chui, Bughin, Dobbs, Bisson & Marrs 2013: 11) big data 

applied with disruptive technologies like Internet of Things, cloud, next generation genomics, 

and advanced robotics, are expected to increase significantly and become a trillion-dollar 

business by 2025. It is also estimated to reduce healthcare spend in US of $300 billion to 

$450 billion (Groves, Kayyali, Knott, Van Kuiken 2013: 8 – 9). Big data is of great 

significance in optimizing the costs of public and private health systems, and it also promotes 

healthy lifestyles and activities, helping people to avoid chronic diseases (Chen, Ma, Song, 

Lai & Hu 2016: 830), so it definitely pays off to resolve as many challenges as possible in 

the years to come.  

 

 

1.2. Research gap 

 

Studies have indicated that there are challenges with matching the capacity of healthcare 

units with the need of care of patients, which requires to develop systems that increases the 

accessibility for care and better match the supply and demand (Nordgren 2011: 304). Such 

systems can be considered as platforms for value co-creation opportunities for healthcare 

service providers and healthcare consumers, referred to as patients (Andrews, Sahama & 

Gajanayke 2014: 375). Moreover, the study of Andrews et al. (2014: 378 – 379) indicates 

some promising effects caused and value created by using digital resources in healthcare 

service setting. Therefore, there is demand for additional studies regarding value co-creation 

models of digital healthcare services. Bardhan and Thouin (2013: 447) also indicate the 

importance of examining the information technology enabled capabilities and the impact of 

these capabilities on the process and quality outcomes in order to explain how benefits can 

be derived from adapting information technology in healthcare. 

 

To gain further understanding of the nature and scope of value co-creation, Lusch, Vargo & 

Gustafsson (2016: 2060 – 2961) conclude in their research on transdisciplinary service 

ecosystems, that more opportunities to study service ecosystems based on digital platforms, 

including for example computer and information sciences, should be considered. This makes 



 14 

 

 

sense, since such service ecosystems involve in value co-creation not only human actors, but 

also organizational and digital artifacts. According to Storbacka, Brodie, Böhmann, Maglio 

& Nenonen (2016: 3008), actor engagement as microfoundation for value co-creation has 

become a major research stream in strategic management, as it is empirically more 

observable than value co-creation itself. Storbacka et al. (2016: 3013) indicate several 

research gaps related to actor engagement, one particularly concerning the role of machine 

actors, for example advanced algorithms which are predicted to play a much bigger role in 

service ecosystems in the future.  

 

According to Demirkan, Bess, Spohrer, Rayes, Allen & Moghaddam (2015: 734) big data is 

a business priority that has the potential to change the competitive landscape of today’s 

globally integrated economy by providing innovative solutions and new ways to transform 

processes, organizations, entire industries, and even society. However, the research of big 

data and big data analytics have so far concentrated mostly on the technical side, whereas the 

business value, as well as managerial and strategic views especially in the field of healthcare 

has not yet been sufficiently explored (Wang, Kung & Byrd 2016: 1; Sivarajah, Kamal, Irani 

& Weerakkody 2017: 279 – 280; Wang & Hajli 2017: 295). Uncovering the potential value 

and benefits for various stakeholders, governments need to invest time, resources, visioning, 

and planning on how to successfully implement big data technologies (Archenaa & Mary 

Anita 2015: 313) and in future research on how recent advancements on information 

technology and big data analytics systems can be effectively exploited in healthcare services 

(Sakr & Elgammal 2016: 57). 

 

 

1.3. Objectives and research questions  

 

Value co-creation practices have a major role in how value is created. Since the digitalization 

has its effect in value co-creation practices and involved actors, and because the research of 

the value of advanced data analytics and big data analytics in healthcare industry is lagging 

compared to other industries, it is worth to conduct a research on this topic. Thus, the 

objective of this study is to discover how using big data and advanced analytics create 

potential benefits for healthcare service providers and consumers, i.e. the specific populations 

and individual patients, and how it affects in value co-creation practices in the healthcare 

ecosystem.  



 15 

 

 

These research objectives are studied by answering to the following research questions: 

 

RQ 1.  How does big data and advanced analytics generate potential benefits and value for 

healthcare service providers, individual patients, and population health?  

 

RQ 2. How does big data and advanced analytics affect the value co-creation practices and 

actors in a healthcare ecosystem? 

 

Finding out how big data and advanced analytics create value and benefits for various 

stakeholders, contributes the healthcare service organizers and providers to indicate and 

analyze potential care gaps and plan how to narrow the deficits by optimizing healthcare 

services timely and cost-effectively for those patients who benefit most from them.  

 

 

Figure 1. Three perspectives to potential benefits and value. 

 

Specifically, the study is concentrated on the potential benefits and value that arise from 

introducing advanced analytics of healthcare big data as an actor in the value co-creation 

practices. The value is viewed from three perspectives; the healthcare service provider’s, 

population health’s and individual patient’s (Figure 1). The study discusses also how 

advanced analytics affects the value co-creation practices and respective actors in the 

healthcare ecosystem. It is also possible to identify new types of actors emerging in the 

healthcare ecosystem due to the use of big data and advanced analytics. This may provide 

insights for management in how to develop its resources and needed competencies.  



 16 

 

 

The theoretical contribution of this study fills in the stated gap in research of the impact of 

big data analytics in the field of healthcare, using the concept of value co-creation from the 

discipline of strategic management. Also, the theoretical framework used for analyzing the 

paths-to-value chains and specific benefit dimensions, which is primarily intended for 

revealing the business value of big data analytics solely from the healthcare service 

organizer's or provider's perspective, will be extended so that it also sheds light on the 

potential benefits for individual patients and population health. The managerial implications 

include the affected value co-creation practices which need to be considered when planning 

to introduce advanced analytics or big data analytics in healthcare context. Moreover, the 

discovered potential benefits for the stated stakeholders are presented, and a summary of 

identified challenges and opportunities are discussed.   

 

 

1.4. Thesis structure 

 

The thesis first introduces the background to the topic and discusses the potential areas of 

research interests and needs in the field, explains research objectives, and presents the 

research questions. The second chapter introduces the context of the study that is value-based 

healthcare through health benefit analysis, as well as discusses the characteristics of 

individual and population health management. The literature review covers the concepts and 

principles of value, value co-creation, service systems, as well all as characteristics of typical 

co-creation practices and actors in a healthcare ecosystem. The theoretical part continues 

with discussing big data and big data analytics. Last, the framework for analyzing the value 

and potential benefits of big data analytics is introduced and extended to cover not only the 

business value, but also the value for population health and individual patients. 

 

The theoretical part is followed by the methodology, which introduces the used research 

method and explains the background and reasons why that method was selected. Also, the 

collection, handling and analyzing methods of the empirical data is explained in detail. 

Finally, in the results chapter the case is introduced, and the results of the analyzed empirical 

data is reported. The results are compared to the theory, which is the base for conclusions. 

The conclusions consist of key findings, theoretical and managerial implications, as well as 

ideas for future research. 



 17 

 

 

2. DATA-DRIVEN VALUE CO-CREATION IN HEALTHCARE 

 

This chapter first introduces the context of the study, i.e. information technology (IT) enabled 

value-based healthcare through health benefit and care gap analyses. It also describes the 

purpose and target groups of the analyses, as well as discusses briefly the users, that primarily 

are professionals representing various healthcare service providers.  

 

Further, it introduces through literature review the characteristics of value and value co-

creation, related service logics, as well as discusses how co-creation practices and actors are 

perceived in healthcare ecosystems, and raises the potential impact of digital actors, e.g. data 

analytics and algorithms as value co-creators. The chapter continues with introducing the 

concept of big data and advanced analytics, especially in the healthcare context. Finally, the 

theoretical framework for studying the effects and potential benefits through advanced data 

analytics, on which the health benefit analysis is based, is presented. The framework used for 

the purpose is the 'big data analytics-enabled transformation model', which includes selected 

organizational IT-enabled practices treated and examined in this study as value co-creation 

practices. The model includes also specific benefit dimensions, which are explored not only 

from the business value point of view, but also to find out what kind of value propositions it 

creates for individual patients and population health management. Therefore, specifically for 

this study, an applied version of the big data analytics-enabled model is developed and 

introduced. 

 

 

2.1. Value-based healthcare 

 

Healthcare is often perceived as expensive and inefficient. Moreover, healthcare service 

delivery outcomes are of varied quality. Therefore, there is room for improvement and need 

for changing the focus of how success of healthcare is measured. Instead of monitoring 

healthcare efficiency with the number of patient visits or number of performed tests and 

procedures, the interest should be in the effectiveness of medical and care interventions, for 

example on quicker patient recoveries, fewer readmissions to the hospital, lower infection 

rates, and fewer medical errors. In other words, healthcare should be valued by its outcomes, 

which is referred to as value-based healthcare. The goal of value-based healthcare is to lower 

the healthcare costs and improve the quality and outcomes of healthcare service delivery. 



 18 

 

 

(Cosgrove 2013.) Porter & Teisberg (2006: 4 – 5, 155) argue that the primary goal of each 

healthcare provider must be excellence in patient value. They also agree with that value of 

healthcare can only be measured over the care cycle, not by individual procedures, services, 

office visits or tests, and raise a concern about the fact that among doctors there is a lack of 

overall perspective on the care-cycle, and that navigating in the care-cycle is challenging for 

patients. Therefore, they suggest paying attention to current practices and test how they 

contribute to creating value for the patients, instead of focusing on short term costs and 

battling over who pays what. Moreover, Porter & Teisberg (2006: 8) claim that value-based 

competition among healthcare service providers cuts out the inefficiency and quality 

problems that plague healthcare services and motivates the poorly performing service 

providers to improvement.   

 

Porter (2010: 2477 – 2478) has also argued that achieving value for patients should be the 

overarching goal of healthcare delivery. Further, value in healthcare is created by the health 

outcomes which can be evaluated on individual patient or population level. The goal is what 

matters for patients and unites the interest of all actors in the system. Value should also define 

the framework for performance improvement in healthcare delivery.  

 

Shifting the focus to value-based healthcare means that healthcare service providers must 

solve a challenging puzzle, because they are expected to reduce variation in quality and 

produce improved outcomes at lower costs. However, this can be viewed as an opportunity 

for development. For example, value-based care teams can be established, unnecessary 

practice variation can be eliminated by developing evidence-based care paths across diseases, 

comprehensive care coordination can be improved so that patients move seamlessly through 

the system, as well as unnecessary visits in health centers and hospitalizations can be reduced. 

(Cosgrove 2013.) In order to develop more qualified and value-based healthcare services, 

information technology platforms and data-driven solutions play a major role (Cosgrove 

2013; Barlow 2016: 1 – 3). One possibility to improve the value of healthcare for individual 

patients and specific populations, is to perform health benefit analyses in order to target the 

interventions and care for those in need, and for those who would gain most benefit out of 

them.  

  



 19 

 

 

2.1.1. Health benefit analysis and care gap  

 

Health benefit analysis and care gap, are two central concepts to understand when discussing 

the health impact of the outcomes of specific interventions. The health benefit analysis has 

according to Kunnamo & Alper (2017) two phases: determining the care gap and calculating 

the health benefit of filling the gap.  

 

According to Kunnamo & Alper (2017: 2 – 3) The health benefit analysis for an individual 

is a list of net impacts of different interventions. For the patient, a health impact can be 

considered as a benefit or a harm. The net benefit or harm of an intervention is the sum of 

net impact of all its outcomes. Further, “the health benefit is the net benefit of an intervention, 

which is calculated by subtracting the sum of all important harms from the sum of all 

important benefits”. On individual level, health benefit analysis acts as a tool for making a 

care plan and shared decisions between the patient and the doctor when making choices 

between alternative interventions (Kunnamo & Alper 2017: 3). The freedom of choice 

regarding the interventions creates value for the patient and means that the doctor and patient 

are practicing shared decision-making which differs from the traditional situation where 

doctor is responsible for the decision-making and risk-bearing when deciding which 

interventions are most beneficial and impactful for the patient’s health. (McGuire, Henderson 

& Mooney 1988: 39, 46, 48; Jung & Padman 2015: 302). However, it is often claimed that 

patient is deemed to be in an asymmetric relation to healthcare providers and thus incapable 

of making purposeful choices based on sufficient knowledge, as well as that in case the doctor 

fails to give patients information about alternative interventions, the patients possibility to 

choose is strongly impaired (Nordgren 2011: 309). In this respect, the health benefit analysis 

seems to be promising. 

 

On the population level, the health benefit analysis helps the healthcare service providers to 

allocate resources for medical services and interventions that provide the largest health 

benefit for the population in the most cost-effective manner. Thus, the health benefit analysis 

for a population is a care gap analysis listing how many people would benefit from each 

intervention complemented by numbers that indicate the average health impact of each 

intervention. (Kunnamo & Alper 2017: 3 – 4.) Such innovative use of advanced analytics is 

comparable to other innovative internet technologies which can be valuable for underserved 

populations as with this technology care providers can reach patients who otherwise would 



 20 

 

 

not have access to healthcare service they could benefit from (Jung & Padman 2015: 299 – 

300).  Practical examples of the described health benefit analyses (Kunnamo & Alper 2017) 

are presented in Appendix 1. 

 

The concept of care gap in healthcare context seems often to be related to a specific medical 

condition or population, such as patients with heart disease, children, women, the elderly, or 

an ethnic group. A general definition of a care gap, provided by The Free Medical Dictionary 

(2017), according to which health care gap is “a disparity between healthcare needs, and the 

healthcare services, especially as it applies to the medically indigent”. This definition 

supports the concept of health and well-being gap which Sitra (2017) describes being the care 

gap between the needs of an individual patient and the healthcare services offered or available 

for the individual patient or a group or a population of similar patients. The gaps can also be 

caused by patient’s hidden needs which currently can be found only randomly when the 

patient is visiting a doctor for another reason. Care gap analysis, as well as the plans how to 

implement it, are also discussed by Lehto and Neittaanmäki (2017: 19 – 21) in their report 

regarding the Finnish health data environment. However, they do not provide any 

comprehensive definition to the care gap or care gap analysis concepts, therefore the concepts 

and definitions provided by Kunnamo & Alper (2017) and Sitra (2016) are applied 

accordingly in this study. 

 

2.1.2. Individual patient’s and population health management 

 

An individual person’s health and wellbeing depends on many factors, such as the person’s 

overall health condition, living and health behavior habits. In cases of sudden illness or long-

term illness, the person becomes a patient, or a consumer of healthcare services run by public 

or private healthcare providers. The individual patient is treated with interventions ordered 

or recommended by doctors and healthcare professionals. The recommended interventions 

and treatments are based on the medical professionals’ expertise and assessments on what 

would be the most beneficial for the patient’s health. In addition, to get well, or improve their 

health condition, the patients also themselves have to take responsibility for their care with 

following the recommendations and possibly make some changes in their lifestyle. 

According to Batalden, Batalden, Margolis, Seid, Armstrong, Opipari-Arrigan & Hartung 

(2016: 509), in such situation health outcomes, in good and bad, are co-produced between 

doctor and patient, and emphasize the importance of effective communication so that a shared 



 21 

 

 

understanding of the problem and mutually acceptable care plan can be created. This is 

supported also with the empirical evidence on that informed and activated patients may be 

effective in facilitating good health outcomes at lower cost (Batalden et al. 2016: 509). 

 

Porter (2010: 2478) argues, that value for an individual patient is created by providing 

combined efforts over the full cycle of care and that benefits, and outcomes of care depend 

on how successfully the practices and interventions are integrated. However, according to 

Health Level Seven International1 (2013: 6 – 7), care planning and coordination of care 

delivery over time and across multiple settings and disciplines has long challenged the 

healthcare community due to the complexity of chronic conditions, increased number of 

interventions, and care across multiple sites. Therefore, it recommends healthcare service 

providers to use digital and standardized integrated care plans, which cover all conditions 

and treatments of an individual patient. Also, Nordgren (2009: 124) points out the importance 

of care coordination in order to avoid ineffective use of healthcare capacity and staff, 

decreased accessibility and long waiting periods to healthcare, as well as the risk of offering 

inadequate care for an individual patient.  

 

Population health has been defined in literature in different ways, because it is, according to 

Kindig (2007: 139 – 140), a relatively new term and there has not been agreement whether it 

refers to the concept of health or to the field of study of health determinants. Therefore, 

Kindig and Stoddart (2003) and Kindig (2007) have studied the concept thoroughly to be 

able to provide a suggestion for how to determine population health and how to define the 

concepts related closely to it. As a result, they define population health as “the health 

outcomes of a group of individuals, including the distribution of such outcomes within the 

group” (Kindig & Stoddart 2003: 381; Kindig 2007: 143).  

 

Some authors have chosen other viewpoints to define population health, and for example 

Dunn and Hayes (1997: S7) in their turn determine population health as “the health of a 

population as measured by health status indicators and as influenced by social, economic, 

and physical environments, personal health practices, individual capacity and coping skills, 

                                                 
1 Health Level Seven International (HL7, http://www.hl7.org/) is a not-for-profit, ANSI-accredited 

standard developing organization dedicated to providing a comprehensive framework and related 

standards for the exchange, integration, sharing, and retrieval of electronic health information that 

supports clinical practice and the management, delivery, and evaluation of health services. 



 22 

 

 

human biology, early childhood development, and health services”. (Kindig 2007: 143 – 

145.) 

 

Population health outcomes are discussed extensively in the literature as well, and for 

example Kindig (2007: 148) provides a definition for the outcomes as “all possible results 

that may stem from exposure to a causal factor from preventive or therapeutic interventions; 

all identified changes in health status arising as a consequence of the handling of a health 

problem”. Further, health outcome measures can be classified for example to mortality rate, 

morbidity, disability, health status, and quality of life.  

 

Evans and Stoddart (1990) define population health outcomes and their distribution in the 

population with specific health determinants such as social environment (e.g. income, 

education occupation), physical environment (e.g. clean air and water, urban design of 

neighborhoods), genetic endowment, individual response (behavior/habits and biology), 

health care (access, quantity and quality of health care services), disease, health and function, 

well-being, and prosperity. (Kindig 2007: 153.)  

 

According to Kindig (2007: 142), population refers to a group of individuals, in contrast to 

the individuals themselves, organized into many different units of analysis, depending on the 

purpose of the research. A population can be for example a geographic region, nation, 

community or a group of employees, disabled persons, or ethnic groups (Kindig & Stoddart 

2003: 381).  

 

To create value and by that means benefit for the whole society, healthcare organizations 

need, as their core responsibility, to improve the health of populations and individual 

patients’ experience of care, but at the same time also reduce the cost per capita of healthcare 

(Kindig & Isham 2014: 7). Triple Aim, framework developed by Institute for Healthcare 

Improvement2, is a contemporary concept striving to fulfill the mentioned three aspects. 

Kindig & Isham (2014: 3) propose, that the Triple Aim framework is complemented by 

developing a specific multisectoral community health business partnership model, which 

they claim to be even better for achieving the goals. 

                                                 
2 The Institute for Healthcare Improvement (IHI, http://www.ihi.org/), an independent not-for-profit 

organization based in Cambridge, Massachusetts, is a leading innovator, convener, partner, and driver 

of results in health and health care improvement worldwide. 



 23 

 

 

However, a common challenge for healthcare systems funded through taxation, regardless 

how it is organized, is to decide what services to offer and to whom, within a limited budget 

(Airoldi, Morton, Smith & Bevan 2014: 965). This may cause inequalities among local 

population of a healthcare center, as some patients are offered interventions they need, and 

due to several reasons, some are not. Some patients may be even unnecessarily over-treated. 

To ensure that interventions are delivered in equal manner, and to provide a tool for the local 

health planners in their annual task of allocating fixed budgets to a wide range of types of 

healthcare and improve population health in a specific geographical area, Airoldi et al. (2014: 

965) have developed a model for socio technical allocation of resources (STAR) including a 

value-for-money triangle (Figure 2), which can be used for evaluating the cost-effectiveness 

of interventions and expected benefits, as well as how to improve population health and 

reduce possible inequalities among population.  

 

 

 

Figure 2. The structure of a value for money triangle.  

  



 24 

 

 

The horizontal side of the triangle represents the cost associated with the intervention. The 

vertical side represents the additional expected benefit score. The larger the triangle, the 

larger the health benefit in the population. The higher the triangle, the more cost-effective 

intervention. Additional benefit can be gained from reduction of health inequality. (Airoldi 

et al. 2014: 970; Kunnamo 2016: 69.) 

 

To conclude, value-based healthcare concentrates on the effectiveness of interventions and 

treatments and is measured by health outcomes and improvements in patient’s health instead 

of the number of single visits in health centers. On population level, practicing value-based 

healthcare provides an opportunity to cost-effectiveness and reduction of health inequalities 

among patients. Various analytical methods, such as health benefit analyses can be used to 

reach the underserved patients and populations. The earlier the needed interventions are 

implemented, the better and more cost-effective health outcomes.   

 

 

2.2. Value and value co-creation in service systems 

 

Creation of value has traditionally been the core purpose and central process of economic 

exchange (Vargo, Maglio & Akaka 2008: 145). Regarding value and value co-creation, there 

are several recurring concepts which need to be explained and defined, in order to be able to 

understand and discuss the opinions on the characteristics of value co-creation presented by 

a number of researchers. Therefore, in this subsection, these concepts and related service 

logics and systems are introduced and discussed. Further, the value co-creation practices and 

service ecosystems are presented in the context of healthcare.  

 

2.2.1. Concept of value 

 

Although it is difficult to define and measure, value for customers, according to Grönroos 

(2008: 303), means that after a customer has been assisted by a self-service process or a full-

service process, he or she feels better off than before. For example, in successful healthcare 

service delivery, where the outcome has improved an individual patient’s health, value has 

been created for the customer. Also, Rantala & Karjaluoto (2016: 34) suggest that in the 

healthcare sector, the definition of value and value offering is based on the betterment of the 

patient’s condition. Sometimes value can be measured in financial terms, sometimes it can 



 25 

 

 

be indicated through effects on revenues or wealth or gained through cost savings. Value is 

considered to have also an attitudinal component such as trust, affection, comfort, and 

easiness of use. In some cases, value can also be negative. (Grönroos 2008: 303.)  

 

Value and the nature of it has been debated already in the ancient times as in the 4th century 

B.C. Aristotle distinguished value into use value and exchange value in order to address the 

difference between things and their attributes. Use value, or value-in-use refers to collection 

of substances or things and the qualities associated with these, whereas exchange value, or 

value-in-exchange has been more difficult to explain. In contemporary research value-in-

exchange is referred to as goods-dominant (G-D) logic and exemplified with how value is 

created in the form of goods, e.g. an automobile, which is exchanged in the marketplace for 

money. (Vargo et al. 2008: 146.)  

 

Grönroos (2008: 298, 304) expresses his views on the issue with “when customers are using 

resources they have purchased, value is created as value-in-use”, and continues with a 

statement that is “value-in-exchange in essence concerns resources used as a value 

foundation which are aimed at facilitating customers’ fulfilment of value-in-use”, and draws 

a theoretical conclusion according to which value-in-exchange can exist only if value-in-use 

can be created.    

 

Further, Grönroos (2008: 298) claims that when value is viewed from the value-in-use 

perspective, the customers are regarded as the value creators. When a service provider adopts 

a service logic which enables its involvement in the customer’s value-generating processes, 

the service provider can become a co-creator of value with its customers. Hence, value co-

creation is a value generating process carried out in interaction between the supplier or 

service provider and the customer.  

 

Some scholars refer to the value co-creation phenomenon with different terminology. For 

example, Osborne, Radnor and Nasi (2012: 139) have studied service-dominant approach in 

public management and agree with that service is an intangible process in which production 

and consumption occur simultaneously, and where users are obligate co-producers of the 

outcome. Co-production as a term, however, characterizes more the goods-dominant logic 

which refers to transformation of raw materials into sellable goods. 

 



 26 

 

 

Vargo et al. (2008: 145) define value co-creation as a phenomenon where value is created 

collaboratively in interactive configurations of mutual exchange and refer to these 

configurations as service systems. Moreover, it is stated that the service systems and value 

co-creation are studied within a discipline called service science.  

 

2.2.2. Service science and service-dominant (S-D) logic  

 

In this study, the business value and potential benefits of big data analytics are studied in the 

field of healthcare. To learn how the healthcare services are arranged and how healthcare 

works, the interaction between healthcare service providers, patients and other possible 

actors can be explored. Service science is the study of service systems and of value co-

creation within complex constellations of integrated resources. A service system is an 

arrangement of resources, such as people, competences, technology (e.g. algorithm) and 

information (derived from data) connected to other systems by value propositions. The 

purpose of a service system is to make use of its own resources to improve its circumstances 

and enhance that of others. In service systems, value is defined in terms of improvement in 

system well-being. In other words, value is not created until the well-being of a customer has 

improved in some way. (Vargo et al. 2008: 145, 149 – 150.) The definition of service science 

is applicable when examining healthcare service arrangements consisting of populations and 

patients as customers, and for example health centers with their professionals, equipment and 

supporting information systems as service providers.  

 

According to Vargo et al. (2008: 145), service is determined as the application of 

competences, such as knowledge and skills. In addition, Grönroos (2008: 300), brings into 

focus the fact that service can be viewed from three aspects, stating first that service is an 

activity, a process where someone does something to assist someone else, and then examining 

service from customer’s perspective as value creation, and from service providers perspective 

as business logic.  Further, Grönroos (2008: 299) separates the service logic from customer’s 

and service provider’s perspectives in the following way: 

 

“1. When using resources provided by a firm together with other resources and 

 applying skills held by them, customers create value for themselves in their 

 everyday practices (customer service logic). 



 27 

 

 

  2. When creating interactive contacts with customers during their use of goods and 

 services, the firm develops opportunities to co-create value with them and for 

 them (provider service logic).” 

 

Service-dominant (S-D) logic, developed and introduced by Vargo et al. (2008), emerged 

when the traditional goods-dominant models of value creation concentrating only on firm’s 

output and price were challenged with new alternative perspectives. S-D logic is 

characterized with the notion of value co-creation that suggests that there is no value until an 

offering is used, and that the customer always need to participate in value co-creation (Vargo 

et al. 2008: 148). Grönroos (2011: 282 – 283) has however challenged this view of S-D logic 

and argues that the value creating activities of the service provider and customer cannot be 

included in the same analysis, and suggests that the service provider’s value creation is an 

all-encompassing process which is separate from the customer’s creation of value-in-use. 

Further, in case there is a mutual value-in-use creation between the service provider and 

customer, Grönroos (2011: 290 – 291) insists that it happens in “one merged coordinated 

interactive process”, and states that “if there are no direct interactions, no value co-creation 

is possible”. 

 

Interestingly, Rantala & Karjaluoto (2016: 34) agree with the definition of value co-creation 

where both parties create mutual value via cooperation, but also address the new mode of 

interaction in value co-creation as the digitization of the services is transforming the scene. 

The idea that value can be created only in direct interaction, is challenged as digitization of 

services has changed the traditional service-process thinking and technology has enabled 

both parties to act independently and not necessarily simultaneously (Rantala & Karjaluoto 

2016: 36). The new ways of interaction in value co-creation through digital platforms 

transforms value co-creation according to Rantala & Karjaluoto (2016: 40) so remarkably, 

that they suggest a paradigm shift in definitions of interaction and time-dependency. 

 

As indicated, the exploration of value co-creation has over the years raised lively debate 

among scholars. Thus, the definitions of it as well as the determining factors of S-D logic 

have now been encapsulated by Vargo & Lusch (2017: 47) and Lusch et al. (2016: 2957) in 

form of five axioms: 

 

“ 1. Service is the fundamental basis of exchange 

   2. Value is co-created by multiple actors, always including beneficiary 



 28 

 

 

   3. All social and economic actors are resource integrators  

   4. Value is always uniquely and phenomenologically determined by the           

 beneficiary     

   5. Value co-creation is coordinated through actor-generated institutions and      

 institutional arrangements”. 

 

Conclusively it can be argued, that in a service situation, value is co-created rather than 

created and delivered by one actor, and that the S-D logic represents a dynamic continuing 

narrative of value co-creation through resource integration and service exchange (Vargo & 

Lusch 2017: 47). Moreover, since the S-D logic represents a dynamic and continuing 

narrative of value co-creation, it has increasingly been introduced to new disciplines, such as 

data analytics and cognitive computing. Also, the continued research of value co-creation has 

resulted in studies where various service ecosystems have become more frequent units of 

analysis for value co-creation, for example in healthcare. (Vargo & Lusch 2017: 47, 58, 62.)  

 

2.2.3. Healthcare ecosystem and value co-creation practices  

 

Ecosystems are, in biological literature, defined as communities of organisms interacting 

over time and space with other organisms and other elements in the system. Markets, 

economies, and similar human systems are comparable with natural ecosystems because they 

change and emerge similarly over time. Interestingly, to capture this systemic dynamism, S-

D logic has identified the concept of service ecosystem. (Lusch et al. 2016: 2958.) 

 

A service ecosystem is defined by Lusch et al. (2016: 2958) as “a relatively self-contained, 

self-adjusting system of resource-integrating actors connected by shared institutional 

arrangements and mutual value creation through service exchange”. Moreover, when the 

five axioms of the S-D logic are coupled with the concept of service ecosystem, Vargo & 

Lusch (2016) formulate the process of value co-creation in the following way (Lusch et al. 

2016: 2958): 

 

“Value co-creation occurs through (social and economic) actors, involved in resource 

integration and service exchange, enabled and constrained by institutions and 

institutional arrangements, establishing nested and interlocking service ecosystems 

of value co-creation, which serve as the context for future value co-creation 

activities.” 

 



 29 

 

 

 

As indicated in Figure 3, healthcare ecosystem consists of multiple levels and actors being 

more complex than a relationship between solely a doctor and patient (Frow et al. 2016: 27). 

The healthcare ecosystem is divided into four levels each consisting of various actors, such 

as people and organizations. The mega level involves government agencies defining the 

aspects of health policy, while on macro level e.g. state health authorities determine the 

allocation of funding. On meso level operate the hospitals and local health support agencies, 

and on micro level the co-creation practices involve doctors, nurses, and patients with their 

families. (Frow et al. 2016: 27.) 

 

 

 

 

 

Figure 3. Healthcare ecosystem (adapted from Frow et al. 2016: 27). 

 

As stated, the broader mega level issues concern mostly the policies set by the national 

governments. In this context, the ongoing healthcare reform in Finland can be addressed. 

According to the current political debate, the governmental actors on the mega level suggest 



 30 

 

 

that after the reform is implemented, there will be a new actor on the macro level responsible 

for organizing the healthcare service, and on the meso level, there will be public and private 

healthcare service providers.  

 

Value co-creation practice is a resource integration process which involves actors sharing 

their resources during collaborative activities and interactions. For example, sharing 

resources such as electronic health records between hospitals and health centers results in 

more informed and relevant treatment for patients. The important role of practices is to link 

the actors within an ecosystem, as well as realize benefits for the actors and ensure the well-

being of the service ecosystem, and finally for the benefit of the patient. (Frow, McColl-

Kennedy & Payne 2016: 24, 26.)  

 

One purpose of the study by Frow et al. (2016: 30) was to identify and create a typology of 

co-creation practices in the field of healthcare and analyze whether the impact of the 

identified practices can be considered as positive or negative. They identified altogether eight 

co-creation practices (Table 1) that support the ecosystem well-being. This study highlights 

practices CP3, CP5, CP6 and CP7, because they might be positively impacted when health 

benefit analyses (cf. subsection 2.1.1.) are used as part of healthcare services. 

 

Table 1. Typology of co-creation practices and their indicative measures in healthcare (Frow 

et al. 2016: 31 – 33). 

 

 

Co-creation practices Examples of indicative measures 

CP1: Practices that endow actors 

with social capital 

Density or volume of interactions 

Degree of bonding, bridging, and linking actors 

Actor proximity in direct or intermediated 

interaction 

CP2: Practices that provide an 

ecosystem with shared language, 

symbols, signs and stories 

Extent that dissemination of symbols, signs and 

stories within ecosystem 

Extent of use/dissemination of symbols, signs and 

stories 

CP3: Practices that shape an actor’s 

mental model 

Change in co-creation practices/behavior/activities 

Change in actors’ worldview of their role within 

the whole ecosystem 



 31 

 

 

Extent of adoption of customer-centered practices 

(e.g. patient-centered care model) 

CP4: Practices that impact the 

ecosystem, created or constrained by 

the physical structures and 

institutions that form their contexts 

The extent to which rules, norms, and procedures 

change over time together with their impact 

How changes to a structure or institution impact 

existing practices and support new practices 

CP5: Practices that shape existing 

value propositions and inspire new 

ones 

Extent of actor perceived change in focus or in 

direction of value proposition 

Articulation of new propositions 

Extent to which existing and new value 

propositions follow best-practice guidelines 

CP6: Practices that impact access to 

resources within an ecosystem 

Extent to which actors offer access opportunities 

and platforms for resource sharing 

Extent to which resources are shared by all actors 

within an ecosystem 

CP7: Practices that forge new 

relationships, generating interactive 

and/or experiential opportunities 

Extent to which practices create opportunities for 

forging new relationships within the ecosystem 

Extent to which actors engage in new co-creation 

practices 

Extent, strength and intensity of relationships 

within ecosystem 

CP8: Practices that are intentionally 

co-destructive creating imbalance 

within the ecosystem 

Defection of actors from the ecosystem 

The growth of new ecosystems that supersede 

original ecosystem 

Extent of conflicting roles of actors who belong to 

multiple ecosystems 

 

 

In addition, patients can also become active co-creators for their own health services when 

they are given more responsibility for maintaining their own health for example by eating 

healthier foods, exercising, and practicing self-care (Nordgren 2009: 121). However, there 

has been claims that patients are deemed to be in asymmetric position in relation to the 

healthcare providers, and thus incapable of making proper choices based on sufficient 

knowledge (Nordgren 2011: 309). In practice, according to Nordgren (2011: 309), there is a 

lack of system, which informs patients of the available options in terms of treatments, as well 

as risks and quality, and doctors fail to give patients information about alternative treatments 

on regular basis which lead to situations where the patients’ possibilities to choose are 



 32 

 

 

strongly impaired. Hence, as suggested by Frow et al. (2016: 32), practices that inspire new 

value propositions, may solve the problem in form of e.g. health benefit analysis. 

 

Further, Payne, Storbacka & Frow (2008: 93 – 94) suggest that professionals can even teach 

the patients certain value co-creation behaviors by for example communicating expectations 

on how patients can actively participate in the co-creation of value. This can be compared to 

a situation where healthcare professional suggests a patient to stop smoking as an intervention 

to improve his health condition. If patient agrees and acts as suggested, value co-creation can 

be claimed to happen. 

 

Frow et al. (2016: 35) argue, that value co-creation practices have a central role in shaping 

an ecosystem, and that the co-creation practices they identified are especially relevant to 

healthcare and to the emerging trend toward putting the patient at the center of processes and 

structures related to their well-being. Again, in the context of the Finnish healthcare reform, 

the plan is, that on the micro level, the patient-centered approach will be in focus, thus 

providing the patient a freedom of choice regarding the healthcare service provider.  

 

Regardless of the national setup of the healthcare ecosystem, within it there are multiple 

interactions that occur with each level and across levels. Many actors are also involved 

directly and indirectly, within and across these ecosystem levels. (Frow et al. 2016: 28.)  

 

Besides that, Payne et al. (2008: 83, 88) agree with the discussion regarding the service 

science theme by arguing that the key feature of the service-dominant (S-D) logic is that the 

customer becomes a co-creator of value, they add that value co-creation through 

technological breakthroughs and innovative services offer new ways for service providers to 

engage customers in co-creation of value and customer experiences. For example, a 

healthcare service provider can engage patients in their own care and decision making related 

to it by offering services via new technology, digital platforms or through data-driven 

solutions.  

 

2.2.4. IT-enabled and data-driven value co-creation 

 

According to Storbacka et al. (2016: 3010), the value co-creating actors in an ecosystem can 

be humans or a collection of humans, such as organizations, but if limiting the view only on 



 33 

 

 

human actors, the impact of technologies is ignored. However, service ecosystems are 

increasingly dependent on technology. Technological advancements in information 

technology (IT), such as digitization of services in various disciplines provide significant 

opportunities to study how it impacts the actors in an ecosystem, not only human but also 

other natural and artificial elements such as algorithms, and their interactions within the 

service ecosystem, for example in the field of healthcare. (Lusch et al. 2016: 2960.)  

 

Information technology enables the effective coordination of healthcare services, improve 

patient management, and play a key role in expanding access to healthcare services as it helps 

with connecting patients and health professionals, as well as provides patients an opportunity 

to act as active partners in their treatment (Quaglio et al. 2016: 314). For example, the use of 

electronic health records (EHR) and proving patients access to their own records, as well as 

embedding decision support tools in EHR systems have reportedly generated positive 

impacts on healthcare quality and better healthcare process quality (Bardhan & Thouin 2013: 

439).  

 

Demirkan et al. (2015: 734) agree by arguing that IT enables organizations to improve their 

inter- and intra-organizational collaboration, effectiveness, efficiency, and innovativeness by 

facilitating new types of services and creating possibilities for value co-creation with 

consumers, i.e. patients in the case of healthcare services. For example, a healthcare service 

provider can enable value co-creation with patients by providing an online booking system, 

telemedicine services, as well as improve effectiveness by sharing electronic health records 

among healthcare providers (Andrews et al. 2014: 376). 

 

According to Groves et al. (2013: 7), introducing big data in the service ecosystem in 

healthcare may be even changing the paradigm, as it enables the creation of feedback loop 

which keeps patients informed, provides opportunity to evidence-based care and selecting 

appropriate care provider, leads to sustainable approaches continuously enhancing healthcare 

value in form of cost reductions at the same or better quality, as well as provides opportunity 

for innovation. Consequently, according to Storbacka et al. (2016: 3010) the entities in such 

ecosystems are collections of arrangements of resources, including people, organizations, 

technology and information, e.g. big data, and advanced analytics of it.  

 



 34 

 

 

To conclude, in service ecosystems value is co-created in interaction between various actors. 

In health ecosystem, there are specific value co-creation practices that can be identified and 

evaluated how they are affected or transformed when in addition to the digital actors, such 

as advanced analytics, is introduced in the ecosystem. 

 

 

2.3. Big data 

 

In information technology, data is the source of information and knowledge. The value of it 

lies in its use, which varies over time, place, and context. If data is in isolation, it has no 

meaning or value. Data can be of wide range of value, but it is often found only long after it 

has been collected and organized, or the value of it is understood only after it is lost. 

(Borgman 2015: 3 – 4.)  Since 1990s, companies have been storing large volumes of data 

(Delen 2014: 232), and the amount of collected and stored data from various sources in 

multiple formats has increased exponentially, which has led to the rise of the concept of big 

data (Demirkan et al. 2015: 735). Regarding the value of big data, it has been compared to 

oil of modern business and the glue of collaboration. It is expected to reveal the hidden 

treasures in the bit streams of life. (Borgman 2015: 3). To uncover the value of big data to 

healthcare, Archenaa & Mary Anita (2015: 407 – 408) recommend also governments to 

harness it for use in order to improve quality and minimizing the costs, and to enable value-

based healthcare for patients. 

 

2.3.1. Definitions of big data 

 

Big data means different things to different people, and traditionally the term has been used 

to describe massive volumes of data analyzed by huge organizations such as Google or 

research science projects at NASA (Delen 2014: 231; Demirkan et al. 2015: 734). In 

healthcare, according to the European Union (2016: 11), big data refers to large routinely or 

automatically collected datasets, which are electronically captured and stored. It is reusable 

in the sense of multipurpose data and comprises the fusion and connection of existing 

databases for the purpose of improving health and health system performance. It does not 

refer to data collected for a specific study. 

 



 35 

 

 

The dimensions of big data are in the literature often defined with varying number of big 

data V’s. For example, IBM data scientists (2017) break big data into four V’s: volume – 

scale of data, velocity – analysis of streaming data, variety – different forms of data, and 

veracity – uncertainty of data. Gartner (2017a) in turn, characterize big data the V’s being 

high in nature, e.g. high-volume, high-velocity and/or high-variety information assets that 

demand cost-effective, innovative forms of information processing that enable enhanced 

insight, decision making, and process automation. 

 

As the number of V’s and characteristics of big data presented in the literature are variedly 

defined, a summary of the V’s and respective definitions or described characteristics are 

presented in Table 2. 

 

Table 2. Summary of the seven big data V’s definitions and characteristics. 

 

 

Big data ’V’ Definition / Characteristics Source 

Volume Large amounts of data 

 

Refers to the scale and the size of data 

 

 

The data comes in large amounts 

Berman (2013: xx) 

 

Sakr & Elgammal 

(2016: 50) 

 

Sahay (2016: 420) 

Velocity The content of the data is constantly changing, 

through the absorption of complementary data 

collections, through the introduction of 

previously archived data or legacy collections, 

and from streamed data arriving from multiple 

sources 

 

Represents the streaming data and large-volume 

data movements 

 

Concerns data in motion/streaming data, 

bandwidth, and how fast data is being produced 

and how fast it must be processed to meet the 

needs/demands 

 

The data has a real-time and continuous nature 

Berman (2013: xx) 

 

 

 

 

 

 

Sakr & Elgammal 

(2016: 50) 

 

Demirkan et al. 

(2015: 735) 

 

 

 

Sahay (2016: 420) 



 36 

 

 

Variety /  

Variability 

The data comes in different forms, including 

traditional databases, images, documents, and 

complex records 

 

Refers to the complexity of data in many 

different structures 

 

Concerns data’s many forms (i.e. structured, 

unstructured, text, multimedia, video, audio, 

sensor data, meter data, html and so on) 

 

The data (structured and unstructured) have 

different sources 

Berman (2013: xx) 

 

 

 

Sakr & Elgammal 

(2016: 50) 

 

Demirkan et al. 

(2015: 735) 

 

 

Sahay (2016: 420) 

Veracity Concerns data in doubt, e.g. uncertainty due to 

data inconsistency and incompleteness, 

ambiguities, latency, deception, accuracy, 

quality, truthfulness, or trustworthiness of data 

 

The data can be triangulated from multiple 

sources 

Demirkan et al. 

(2015: 735) 

 

 

 

Sahay (2016: 420) 

Validity The data reflects primary sources of collection Sahay (2016: 420) 

Volatility The data is available over time Sahay (2016: 420) 

Value  

 

Potential of big data to be utilized for 

development 

 

 

Concerns data for co-creation, the relative 

importance of data to the decision-making 

process 

 

The use value of big data represents how it helps 

to address problems and use conditions, while 

the exchange value of big data represents its 

reusable intellectual capital and how it is used in 

multiple contexts to generate value 

United Nations 

Global Pulse  

(2013: 2) 

 

Demirkan et al. 

(2015: 735) 

 

 

Sahay (2016: 421) 

 

 



 37 

 

 

Berman (2013: xx) emphasizes that it is important to distinguish big data from ‘massive data’ 

or ‘lots of data’, and claims that at least volume, variety, and velocity of the V’s must apply 

in order to fulfill the definition of big data.  

 

Delen (2014: 237) claims that velocity may be the most overlooked characteristics of big 

data, but too important to be ignored as when data is created it starts to age and degrade and 

its value proposition becomes worthless, for example, in healthcare, the capability to access 

and process patient data quickly creates more advantageous outcomes for the patient. 

Raghupathi & Raghupathi (2014: 10) agree with this by arguing that real-time big data 

analytics is a key requirement in healthcare. Delen (2014: 238) continues by stating that the 

excitement around big data is created by its value proposition, which promises that by 

analyzing large and feature-rich data, organizations can gain greater business value than they 

could by detecting patterns in small datasets or by using simple statistical methods and 

concludes with statement “big data means big analytics”.  

 

The origin of big data is versatile (IIHT 2013: 6), and according to Delen (2014: 232) big 

data comes from everywhere. As the three-level diagram in Figure 4 depicts, variety and 

velocity, as well as volume of data increases when the traditional databases are first 

complemented with mostly human-generated complicated data from the internet and social 

media and grows even more when machine and sensor generated data is added to the big data 

set.  



 38 

 

 

 

 

 

Figure 4. The wide range of sources for big data (Delen 2014: 233). 

 

For example, in healthcare, big data can be collected in various formats from diverse sources, 

such as internet-based text documents, internet search indexes, sensor networks, social 

networks, global positioning systems (GPS), biology, genomics, biochemical experiments, 

medical records, scientific research, as well as genomic/biomed research (Demirkan et al. 

2015: 735), and further, data in healthcare come in structured format from electronic health 

records (EHRs), or electronic medical records (EMRs), while semi-structured data may 

include instrument readings or converted paper records to electronic health records. In 

addition, structured and unstructured data can be streamed into healthcare systems from 

fitness devices, genetics and genomics, social media, and other sources. (Sakr & Elgammal 

2016: 50.) According to Archenaa & Mary Anita (2015: 409) the big data sources in 



 39 

 

 

healthcare refer to the patient data such as doctors’ notes, laboratory reports, x-ray images, 

national health register data, and even to RFID data of surgical instruments. In addition to 

patient related data, big data in healthcare settings may include also evidence-based medicine 

systems which doctors have traditionally been using to support decision-making (Sakr & 

Elgammal 2016: 56).  

 

The above discussed characteristics and V’s of big data respectively provide challenges to 

consider when planning how to gain value from it. The challenges and critical success factors 

of big data analytics are presented in the following subsection. 

 

2.3.2. Identified challenges and critical success factors 

 

Sivarajah et al. (2017: 263) performed a holistic review on big data and big data analytics to 

gain understanding in the landscape with the objective of making robust investment 

decisions. Based on their review, they concluded that big data challenges can be grouped into 

three main categories based on data lifecycle: data, process, and management challenges. 

 

Data challenges relate to the characteristics of the data itself, e.g. data volume, variety, 

velocity, veracity, volatility, as well as discovery, quality, and dogmatism. Process 

challenges are related to a series of how techniques: how to capture data, how to integrate 

data, how to transform data, how to select the right model for analysis and how to provide 

the results. Management challenges cover for example privacy, security, governance, and 

ethical aspects. (Sivarajah et al. 2017: 265.) Since these challenges are presented on a general 

level, it can be assumed that they are universal and therefore they concern most industries 

utilizing big data and big data analytics, including healthcare.  

 

The success of big data analytics in turn, depends on many critical factors (Figure 5), which 

according to Delen (2014: 240 – 241) need to be clarified and in place before investing in 

any systems or starting the analytics efforts. There should be most of all, a clear business 

need for performing big data analytics, aligned with the current vision and strategy of the 

organization. It is also important to find personnel with analytical skills, choose the right 

analytics tools, and ensure a strong committed sponsorship from the executive level, as 

without it, it would be difficult of even impossible so succeed.  



 40 

 

 

 

 

 

Figure 5. Critical success factors for big data analytics (Delen 2014: 240). 

 

 

Generally, the purpose of various data analytics, such as business analytics is to provide 

insight for problem solving and decision making. Regardless they are not the same, often the 

terms analysis and analytics are used to refer to the same thing. To determine the term 

analytics, Delen (2014: 1) suggests that “analytics is the art and science of discovering 

insight by using sophisticated mathematical models along with variety of data and expert 

knowledge”, and further “these days, analytics can be defined as simply as the discovery of 

meaningful patterns in data”. Moreover, Delen (2014: 3) argues that "analytics is a variety 

of methods, technologies, and associated tools for creating new knowledge/insight to solve 

complex problems and make better and faster decisions”, and “analytics is multifaceted and 

multidisciplinary approach to addressing complex situations”.  

 



 41 

 

 

Hence, big data analytics and modern data mining are relatively new concepts, which also in 

different scope and contexts are often referred to using diverse terminology. Big data 

highlights the challenges related to the increased large data streams, which have been 

addressed with recent advancements in hardware, software, and algorithms. Data mining 

refers to mining corporate data to discover new and useful knowledge to improve business 

and its practices. Data mining plays a key role in analytics, it is “the process of discovering 

new knowledge in the patterns and relationships in large data sets”. (Delen 2014: 1, 4, 14, 

32 – 33.)  

 

 

 

Figure 6. The continuum of data to information to knowledge (Delen 2014: 33).  

 

 

Basically, data can be any data in any format, or even a combination of various data sources, 

i.e. big data. Data is facts, whereas information is organized and processed data, and 

knowledge is information which is contextual, relevant, and actionable (Figure 6). For 

example, according to IIHT (2013: 7) electronic health records coupled with analytical tools 

provide through data mining opportunity to information enabling earlier disease detection, 

more effective outcomes across large populations, and by that means improved population 

health management. 

 



 42 

 

 

2.3.3. Taxonomy of analytics 

 

The analytics are classified according to a simple taxonomy of analytics developed by Delen 

(2014: 16 – 17), into three categories, namely descriptive, predictive, and prescriptive 

analytics, each distinguished by the type of the data used and purpose of the analysis. 

According to Delen (2014: 16) most organizations begin with descriptive analytics and then 

move to predictive analytics and finally to prescriptive analytics which is the most advanced 

level of analytics. Further, he points out that moving from a lower analytics level to a higher 

is not clearly separable, as a business can be in the descriptive analytics level while at the 

same time already use partially either of the more advanced level analytics, too. The 

classification of analytics into descriptive, predictive, and prescriptive is also used by other 

scholars in their research, for example Sivarajah et al. (2017: 266), Wang, Kung, Wang & 

Cegielski (2017: 2 – 3) and Wang & Hajli (2017: 289).  

 

As mentioned, descriptive analytics is the entry level in analytics taxonomy, and it is often 

referred to as business reporting, which provides an answer to questions such as what 

happened and what is happening? Delen 2014: 16). Wang & Hajli (2017: 289) agree with 

this as they state that descriptive analytics provides the ability to describe the data in summary 

form for exploratory insights and to answer to what has happened in the past. Sivarajah et al. 

(2017: 275) in turn state that descriptive analytics is used to identify patterns and create 

reports concerning past behavior, and therefore considered as backward looking and 

revealing only what has already happened.  

 

According to Delen (2014: 16), organizations are mature to move to predictive analytics once 

they are ready to look beyond what happened and able to answer to question what will 

happen? Predictive analytics allow users to predict or forecast the future (Wang & Hajli 

2017: 289; Delen 2014: 17) for a specific variable based on the estimation of probability, and 

it also enables users to develop predictive models to identify causalities, patterns, and hidden 

relationships. Predictive analytics provides the ability to process large volumes of both 

structured and unstructured data, as well as supports the data processing in real-time of near 

real-time. (Wang & Hajli 2017: 289.) Sivarajah et al. (2017: 276) summarize that predictive 

analytics aims to predict the future by analyzing current and historical data.  

 



 43 

 

 

Prescriptive analytics uses optimization-, simulation- and heuristics-based decision-

modelling techniques and tries, according to Delen (2014: 18), to answer to the question what 

should I do? Prescriptive analytics can continually re-predict and automatically improve 

prediction accuracy by taking in new combined structured and unstructured datasets to 

develop more thorough decisions (Wang & Hajli 2017: 289 – 290). Sivarajah et al. (2017: 

277) explain prescriptive analytics with an example where what if simulators help in decision 

making by providing insights regarding plausible options that a business can choose to 

implement in order to maintain or strengthen its position in the market. A comparable 

situation in healthcare could be when analytics is providing insights regarding alternative 

interventions that a patient can choose in order to maintain or improve his or her health. 

 

Descriptive analytics is also called business intelligence (BI) while predictive and 

prescriptive analytics are collectively called advanced analytics. The shift from descriptive 

analytics to the more sophisticated analytics is significant, since it warrants that the analytics 

is advanced. (Delen 2014: 16.)  

 

Moreover, beyond this taxonomy, predictive analytics which employs complex algorithms 

with learning ability, may be even further classified as machine learning (ML), which excels 

at identifying latent patterns and connections that humans are too evolved to perceive 

(Sivarajah et al. 2017: 276; Barlow 2016: 11). Machine learning is a type of artificial 

intelligence (AI) that allows software applications to become more accurate in predicting 

outcomes without being explicitly programmed (TechTarget 2017). According to Gartner 

(2017b), “advanced machine learning algorithms are composed of many technologies, such 

as deep learning, neural networks and natural-language processing used in unsupervised 

and supervised learning, that operate guided by lessons from existing information”. An 

example of artificial intelligence, is IBM’s Watson, which is characterized as “smart” 

machine with human mind, designed to answer questions posed in natural human language. 

Watson is capable to analyze natural language, identify sources, find and generate 

hypotheses, find and score evidence, and merge and rank hypotheses. (Delen 2014: 20 – 21.) 

Similar artificial intelligence applications are developed continuously for various purposes, 

also in the field of healthcare.  

  



 44 

 

 

2.3.4. Big data analytics in healthcare 

 

Many businesses are operating in challenging and complex environments with an ever-

increasing amount of diverse data. Therefore, to gain knowledge and get support for decision-

making based on the best available evidence, businesses have begun to invest in data 

analytics by consciously shifting into data and evidence-driven business practices (Delen 

2014: 4). In healthcare, to gain insight for better informed decisions concerning patients’ 

health, the healthcare service providers have started to use their very large data sets for big 

data analytics, as its potential to improve care and save lives at lower cost have been 

identified (Raghupathi & Raghupathi 2014: 1, 5). Moreover, according to Delen (2014: 24), 

due to the fact that healthcare struggles with an imbalance between demand and supply, and 

increasing prices and decreasing quality, systems that has the ability to help in diagnosing 

and treating patients by analyzing large amounts of data, are needed. 

 

The variety of healthcare data is large, as analytics typically aggregates data from several 

real-time data sources consisting of multiple data formats. Sources used in advanced data 

analytics can be various databases, records and systems, for example electronic health 

records (EHR), clinical decision support systems, as well as web and social media data (e.g. 

clickstream and interaction from Facebook, blogs, health plan websites, smartphone apps), 

machine to machine data (readings from remote sensors, meters and other vital sign devices), 

biometric data (e.g. finger prints, genetics, x-ray and other medical images, blood pressure, 

pulse and pulse-oximetry readings), and human generated data (e.g. email, paper documents). 

Data types can vary from structured data, e.g. traditional electronic health care records, and 

semi-structured data, e.g. the logs of health monitoring devices, to non-structured data, such 

as notes or clinical images. The analytics tools and architecture for structured and 

unstructured big data differ from traditional data management and business intelligence tools 

where the data is assumed to be certain, clean, and precise. (Raghupathi & Raghupathi 2014: 

4 – 5; Wang et al. 2016: 2 – 3.) 

 

Handling data from various sources is challenging because the characteristics of the collected 

data might vary considerably (Wang et al. 2016: 2). For example, as big data is often 

unstructured, messy, and dirty it means that the organizations need to have ability and 

capability to handle the fast arriving data so that it can be converted into actionable insight 

(Delen 2014: 8). Regardless which data sources or formats are used, the veracity of the data 



 45 

 

 

needs to be validated before analyzing it, meaning that it must be ensured that the analyzed 

data is of high quality, truthful, and accurate by making sure that the diagnoses, interventions, 

and outcomes are captured correctly in the data sources (Raghupathi & Raghupathi 2014: 4).  

 

Obermeyer & Lee (2017: 1211) are concerned regarding the data analyzed with highly 

accurate algorithms and bring out some criticism for the veracity of the data. They warn that 

ignoring clinical thinking and relying solely on analytics is dangerous, as the analyzed data 

is always based on human decisions and human mistakes, which can lead to failures. 

Therefore, it is important to consider analytics as thinking partners, not replacements for 

doctors, who need to be trained to utilize analytics to master the complexity of modern 

medicine and patients with more coexisting illnesses and medications (Obermeyer & Lee 

2017: 1209, 1211). 

 

According to Delen (2014: 9), some criticism has been brought out also toward data and data 

analytics, most commonly regarding security and privacy issues due to the risk of breaches 

of sensitive information and leaking or misuse of personal data. Moreover, Gumbus & 

Grodzinzky (2016: 118) raise their concerns of the rise of computational power and cheaper 

and faster devices to capture, collect, store and process data which can lead to “datafication” 

of society and cause discriminatory practices as its side effects, e.g. in a situation where 

analytics lead to harmful or unfair outcomes for individuals or populations.   

 

However, despite of the hurdles in the way, big data analytics is a powerful tool, as it enables 

organizations to gain new insights into organizational knowledge, which can be used in 

decision making and action taking. Using this kind of advanced analytics is not only a matter 

of increased productivity or efficiency, but also of intangible values such as increased 

flexibility and quality improvement. (Wang, et al. 2017: 3 – 4.) 

 

 

2.4. Big data analytics-enabled transformation model 

 

To find out the business value and potential benefits of big data analytics for the healthcare 

industry, Wang & Hajli (2017: 287) developed a big data analytics-enabled business value 

model using the resource-based theory and capability building view. The theory models the 

big data analytics components, capabilities, and benefit dimensions, but does not build a view 



 46 

 

 

on value co-creation practices related to healthcare service delivery. Another theory 

addressing the same subject is the big data analytics-enabled transformation (BDET) model 

(Figure 7) developed by Wang et al. (2017: 2) who supplement the model with a practice-

based view from strategic management to explain how big data analytics can enable 

organizations to develop inimitable practices, which in turn are intended to create business 

value.  

 

 

 

Figure 7. Big data analytics-enabled transformation model (Wang et al. 2017: 2). 

 

The model aims to explain how big data analytics capabilities can create potential benefits 

and business value for a healthcare organization (Wang et al. (2017: 2). The linear progress 

path of the model follows a practice-based view developed by Bromiley & Rau (2014: 1252 

– 1253), who argue that practices are important entities in and of themselves, rather than 

simply indicators for some underlying construct. 

 

Hence, both theories and developed models seek to discover the potential benefits and 

business value of big data analytics in healthcare context, but as the big data analytics-

enabled transformation model by Wang et al. (2017: 2), also includes the practice-based 



 47 

 

 

view, it is more suitable for this study because these practices, even though intended to be 

inimitable, can be viewed from the value co-creation perspective as well. The BDET model 

is explained in more detail in the following paragraphs. 

 

2.4.1. Components of BDET model 

 

Explanatory variables  

 

The explanatory variables of this model refer to the big data analytics capabilities generated 

from big data analytics resources, that are big data analytics architectural components. The 

resources that together build the analytic capabilities consist among others of the data itself, 

managerial and technical skills, and data-driven culture of the organization. Further, the tools 

and functionalities of big data analytics are identified to consist of three architectural 

components, namely data aggregation, data analysis and data interpretation which allow 

users to transform data into evidence-based decisions and informed actions. (Wang et al. 

2017: 2.)  

 

Data aggregation component aims to collect heterogeneous data from multiple sources, e.g. 

data warehouses and databases, and transform it into specific data formats which can be read 

and analyzed. In this phase, according to Wang & Hajli (2017: 289), after data is collected 

and extracted from various sources, data analysis component explains how all kinds of data 

is processed (e.g. data mining or natural language processing) and how analyses are 

performed so that they support evidence-based decision making and meaningful practices in 

healthcare organizations. Data interpretation component generates general clinical 

summaries such as historical reporting, statistical analyses, time series comparisons, provides 

data visualizations and real-time reporting such as alerts, proactive notifications, and 

operational key performance indicators (KPIs), as well as meaningful business insights 

derived from the analytics components. (Wang et al. 2017: 2 – 3, 6.) 

 

To evaluate the analytics capabilities, the model breaks them down into traceability which 

refers e.g. to the capability of tracking medical events and searches in clinical databases; 

analytical capability which refers to the nature of the analysis, e.g. understanding the past 

and current state of variables, causes of occurred medical events and support of real-time 

processing; decision support capability which refers to real-time or near real-time clinical 



 48 

 

 

summaries presented in visual dashboards; and predictive capability which refers to the 

capability to examine undetected correlations, patterns and trends between specific variables, 

compare current and historical data, predict future trends, and provide actionable insights or 

recommendations in readable format (Wang et al. 2017: 6 – 7). Wang & Hajli (2017:290) in 

turn, base their definition in information lifecycle management, and define big data analytics 

capability in the healthcare context as “the ability to acquire, store, process and analyze large 

amounts of health data in various forms, and deliver meaningful information to users, which 

allows them to discover business values and insights in a timely fashion”. 

 

IT-enabled transformation practices 

 

The BDET model explores seven different IT-enabled practices listed in Table 3. The 

practices are classified into localized exploitation, internal integration, business process 

redesign, business network redesign, and business scope redefinition. The two first 

classification levels are evolutionary transformation level practices and the last three are 

considered to be revolutionary transformation level practices. (Wang et al. 2017: 2 – 3, 7.) 

 

 

Table 3. IT-enabled transformation practices with examples (Wang et al. 2017: 7).  

 

 

Classification of IT-enabled 

transformation practice  

Examples 

Localized exploitation: 

1. Meaningful use of EHR  

(electronic health record) practice  

 

Generate lists of patients by specific conditions 

to use for reduction of disparities, research or 

outreach 

Improve care coordination among healthcare 

units through interoperable EHR systems 

Localized exploitation: 

2. Evidence-based medicine practice  

 

Explore the fact from medical events or patient 

treatments to improve specific outcome 

Build holistic view of evidence by insights from 

literature-based data and research studies 

Internal integration: 

3. Multidisciplinary practice  

 

Provide joint decisions regarding treatments to 

patients from a multidisciplinary team 



 49 

 

 

Business process redesign: 

4. Clinical resource integration 

practice  

 

Allocate resources to serve each healthcare unit 

Create centralized information support for 

clinical operation 

Business network redesign:  

5. Network collaboration practice  

 

Build common understanding of healthcare 

service between care providers and other 

stakeholders 

Business network redesign:  

6. Network knowledge creation 

practice  

 

Allow all stakeholders to share information on 

the platforms 

Discover new knowledge by enabling 

stakeholders to collaboratively map ideas 

Business scope redefinition: 

7. Personalized care practice  

 

Create personalized disease risk profile and 

disease and wellness management plan for each 

patient 

 

 

The IT-enabled transformation practices presented in the BDET model can be linked to the 

co-creation typology (Table 1) introduced by Frow et al. (2016: 31 – 33). For example, 

personalized care practice can be linked with the practices that shape an actor’s mental model 

and practices that shape existing value propositions and inspire new ones (CP3 and CP5). 

Also, clinical resource integration practices can be linked with practices that shape an actor’s 

mental model (CP3), as they are affected by how the personalized care practices are arranged.  

In addition, meaningful use of EHR and evidence-based medicine practices can be linked 

with the practices that impact access to resources within an ecosystem (CP6), e.g. in form of 

shared knowledge resources. Multidisciplinary practices, network collaboration practices and 

network knowledge creation practices can be linked to the practices that forge new 

relationships, generating interactive and/or experiential opportunities (CP7), e.g. when 

collaboration between various specialties in hospital or cross-boundary cooperation between 

health and social sectors are developed. 

 

Intermediate outcomes 

 

Wang et al. (2017: 2 – 3) treat intermediate outcomes in the BDET model as benefits. To 

conceptualize the potential benefits, they apply a multidimensional information system (IS) 

benefit framework developed by Shang & Sheddon (2002: 277 – 280) who have identified 

five benefit dimensions in their research. Also, Wang & Hajli (2017: 293) use this benefit 



 50 

 

 

framework and provide some examples on practical benefits. The benefit dimensions and a 

summary of selected examples are listed in Table 4. 

 

Table 4. The benefit dimensions with examples of subdimensions (Shang & Sheddon 2002: 

277; Wang et al. 2017: 3; Wang & Hajli 2017: 290, 293). 

 

 

Benefit 

dimensions  

Description  Examples of subdimensions 

Operational 

benefits  

The benefits obtained from the 

improvement of operational 

activities 

Productivity improvement 

Quality improvement 

Customer service improvement 

Cycle time reduction 

Cost reduction 

Immediate access to clinical data 

for analysis 

Enable proactive treatment before 

the condition worsens 

Managerial 

benefits 

Benefits obtained from business 

management activities, e.g. 

allocation and controlling of 

resources, monitoring of 

operations, and supporting 

business strategic decisions 

Better resource management 

Insights and sound information for 

decision-making and planning 

Performance improvement 

 

Strategic 

benefits 

Benefits obtained from strategic 

activities involving long-range 

planning regarding high-level 

decisions 

Support for business growth 

Support for business alliances 

Building business innovations 

Achieving business competitive 

advantages: cost leadership, 

differentiation, and focus 

Comprehensive view of treatment 

delivery for meeting future needs 

IT 

infrastructure 

benefits  

Sharable and reusable IT 

resources providing foundation 

for present and future business 

applications 

Increased IT infrastructure 

capability 

Reduce of system redundancy 

Transfer data quickly among 

healthcare IT systems 

Simplified IT management 

IT cost reduction  



 51 

 

 

Organizational 

benefits 

Benefits related to 

organization’s focus, cohesion, 

learning, and execution of 

chosen strategies  

Seamless and coordinated patient 

experience delivery 

Changing work patterns 

Facilitating organizational learning 

Cross-functional communication 

and collaboration 

Building common vision 

 

Organizational performance 

 

In the BDET model, organizational performance refers to business value (Wang et al. 2017: 

2). Both Wang et al. (2017: 3) and Wang & Hajli (2017: 290) argue that using Shang & 

Sheddon’s framework helps to understand the potential benefits of big data analytics and 

enhance the understanding of the business value of big data. It also acts as a tool for managers 

to assess the benefits of their firm’s information systems, which means that the model could 

also be used as a general model and guideline for assessment and classification of benefits 

from IT architecture.  

 

2.4.2. Big data-enabled transformation and value co-creations practices 

 

Since one of the objectives of this study is to indicate the possible transformation of selected 

value co-creation practices and evaluate their impacts to the healthcare ecosystem, the IT-

enabled transformation practices of the BDET model are viewed as value co-creation 

practices (cf. Frow et al. 2016: 31 – 33). Additionally, this enables the disclosure of the nature 

of the indicated transformation of the value co-creation practices, which can be considered 

as evolutionary or revolutionary. Another objective of this study is to discover the potential 

benefits and value from several stakeholder groups’ viewpoints. Therefore, in addition to the 

business value perspective, the model is extended with viewpoints of value for individual 

patients and population health. The individual patient’s perspective is important to 

understand in order to be able to develop the personalized patient centric care and motivate 

the patients to participate and acquire a more active role in their own care planning and shared 

decision-making. Regarding the population health perspective, it is essential to understand 

the generated value because it provides insights on how healthcare services should be 

targeted to close the indicated care gaps, which further supports the desired shift into value-



 52 

 

 

based healthcare. Hence, the extended BDET model studies the value of big data analytics 

for three stakeholders. In case the findings show clear paths-to-value, they can also be 

illustrated with this model. 

 

Regarding the explanatory variables, big data analytics resources are studied through 

breakdown into data aggregation, data analysis, and data interpretation, and the capabilities 

in turn through breakdown into traceability, analytical capability, decision support capability, 

and predictive capability (Wang et al. 2017: 6, 10).  

 

To conclude, the original BDET model is extended as explained (Figure 8), and used for 

studying and analyzing the case, i.e. the Health Benefit Analysis tool, developed in a project 

funded and run by Sitra. The case, and the development project are introduced in more detail 

in chapters 3 and 4.  

 

 

 

Figure 8. The extended big data analytics-enabled transformation model (adapted from 

Wang et al. 2017: 2). 



 53 

 

 

The big data analytics-enabled transformation model is aligned with the research questions 

and applied for studying and analyzing transformation practices as value co-creation 

practices and evaluating the performance with business value extended with value for 

individual patients and population health.