1
Value-based healthcare in practice
An inductive study examining how value-based healthcare
could be applied in practice
Master Thesis, MSc Supply Chain Management University of Groningen, Faculty of Economics and Business
February 12th 2018 Word count: 12.665 by Lysanne Douma Student number: s2211025
First supervisor and first examiner
Prof. dr. ir. C.T.B. Ahaus Second assessor
PhD candidate MSc A.C. Noort Second examiner
2
ABSTRACT
3
PREFACE
4
Content
INTRODUCTION ... 6
THEORETICAL BACKGROUND ... 8
Value-based healthcare ... 8
The importance of value-based healthcare ... 8
What value-based healthcare encompasses ... 8
How to transform healthcare: the value agenda ... 11
The importance of conducting this research ... 13
METHODOLOGY ... 14
Research setting ... 15
Case description and selection ... 15
Selection focus group participants ... 16
Data collection ... 16
Data organization and analysis ... 17
RESULTS ... 19
Interviews ... 19
Thirteen main elements of applying VBHC in practice ... 19
Cluster 1: Patient value ... 19
Cluster 2: costs aspect ... 20
Cluster 3: organization of care ... 22
Cluster 4: Steering (information) ... 26
Cross-case analysis ... 28
Focus group ... 29
Categorization of elements ... 29
Selection of most important elements ... 29
DISCUSSION ... 31
Summary ... 31
Interpretation ... 31
Interpretation of the thirteen main elements ... 32
Interpretation categorization process ... 35
Interpretation of indicated important elements ... 36
Theoretical implications ... 37
Managerial implications ... 38
5
Future research ... 39
Conclusion ... 40
REFERENCES ... 41
APPENDIX A: List of elements focus group in Dutch and English ... 44
Dutch version (used during the focus group discussion) ... 44
English version ... 45
APPENDIX B: Interview-protocol ... 47
Interview-protocol in Dutch ... 47
Interview-protocol in English ... 48
APPENDIX C: Schedule focus group (in Dutch) ... 50
Summarized/visualized version of schedule ... 50
All-encompassing version of schedule ... 52
Slides used during the introduction ... 59
APPENDIX D: Results for each individual case ... 61
APPENDIX E: Categorization of elements and indicated most important elements ... 70
Categorization of elements ... 70
Most important elements ... 70
APPENDIX F: Evaluation focus groups ... 73
6
INTRODUCTION
As costs in healthcare are rising, a new strategy is proposed focusing on value maximization for the patient by accomplishing the best outcomes for patients at the lowest possible costs (Porter & Lee, 2013), instead of focusing on volume of delivered services (Porter, 2010). High value realization for patients should be the main focus within healthcare delivery as the system stakeholders will benefit, namely the providers, payers, patients (Porter, 2010) and service -, technology - and product suppliers (Porter & Teisberg, 2006). Although systematic outcome measurement seems an essential condition in order to improve value (Porter, Larsson & Lee, 2016), there is a lack of outcome measurement in healthcare, compared to process measurements (Porter, 2010). Therefore, outcome measurement has to accelerate in order to unlock the potential improvements by value-based healthcare (VBHC) (Porter et al., 2016), namely efficiency and outcome improvement as well as market share growth (Porter & Lee, 2013). Although VBHC is introduced as an important and promising approach within healthcare, in literature it is still mainly a concept in its infancy.
The aim of this research is to gain a better understanding of how VBHC could be applied in practice. Therefore, this paper addresses the question: How could VBHC be applied in practice? As the healthcare delivery system should be transformed into a system based on value creation for the patient (Porter, 2008; Porter & Lee, 2013), and in practice such a transformation in healthcare implies “a long series of local experiments” (Bohmer, 2016, p. 710), we can learn from organizations that started implementing VBHC into practice. Specifically, VBHC practitioners can provide the necessary information for this research. Therefore, we are interested in the perception of practitioners on VBHC in practice. The knowledge obtained from both literature and practitioners will be extended with reflections on this knowledge by clinicians and non-clinicians interested in applying VBHC. As mostly all providers are confronted with difficulties in measuring (and delivering) value and therefore do not succeed in doing so (Porter, 2010), clinicians and non-clinicians can use the knowledge obtained in this research when they start practicing VBHC.
7
have seen that systematic outcome measurement and improvement of outcomes in healthcare go hand in hand. Even so do the initial efforts of transforming to VBHC and performance improvements. However, it should be noted that the concept of creating value in healthcare is not completely new, as for example Nelson (1996) already focused his research on the use of outcome measurement in order to improve value in healthcare and mentioned that one can think of value as “outcomes in relationship to the total costs of care” (Nelson, 1996, p. 112).
8
THEORETICAL BACKGROUND
Value-based healthcare
Over a decade ago, Porter & Teisberg (2006) introduced the concept ‘value-based competition on results’ within healthcare delivery. Later on, the term VBHC is used by Porter (2008) to describe the way in which healthcare should be delivered. Multiple organizations started with transforming to VBHC (Porter & Lee, 2013). Below, the importance of VBHC is discussed, followed by multiple sections describing what VBHC encompasses. This chapter concludes with the importance of conducting this research.
The importance of value-based healthcare
Healthcare spending is growing in the United States (Fisher, Wennberg, Stukel, Gottlieb, Lucas & Pinder, 2003; Keehan et al., 2015). This rise in costs is a concern for many other countries as well (Porter, 2008; Porter & Lee, 2013). This also holds true for countries with “universal insurance coverage” (Porter, 2008, p. 503). “When many people lack access to primary and preventive care and cross-subsidies among patients create major inefficiencies, high-value care is difficult to achieve” (Porter, 2009, p. 109). So although the universal coverage is a necessary condition for achieving a high value healthcare delivery system as it is necessary for efficiency, it is not sufficient (Porter, 2008). Therefore, value of healthcare delivery should substantially increase. The delivered value should in turn determine insurance costs and what is to be covered (Porter & Teisberg, 2006). By improving the value, “patients, payers, providers, and suppliers can all benefit while the economic sustainability of the health care system increases” (Porter, 2010, p. 2477). Multiple organizations (partly) adopted the VBHC approach with striking results, namely market share growth as well as efficiency-improvements and outcome-improvements (Porter & Lee, 2013). So, VBHC delivery is associated with performance improvements and seems an important approach to improve healthcare.
What value-based healthcare encompasses What is value
9
false savings and might limit healthcare effectiveness. Value can be improved in two ways, namely when one or multiple outcomes are improved without increasing the costs or when costs are lowered without the need to make compromises regarding the outcomes (Porter & Lee, 2013). So when the combination of lowest costs and best outcomes is present, value maximization is achieved. Furthermore, value has to be defined around the patient (Porter, 2010) and therefore the highest value should be achieved for patients (Porter & Lee, 2013).
Measuring value
To measure value, outcome should be divided by costs (Nelson, 1996; Porter, 2010). When outcome measurement is not performed comprehensively, it is difficult to get an understanding of what does and does not improve value (Porter, 2009). Costs represent the total amount of costs of the complete care cycle of the medical condition of the patient (Porter, 2010). The complete care cycle can encompass outpatient and inpatient care, but also rehabilitative care and supporting services all together (Porter & Lee, 2013). Outcomes are specific for each condition (Porter, 2010). Furthermore, they are multidimensional. Therefore, outcome measurement should take place by medical condition, instead of intervention or specialty (Porter & Lee, 2013). So patient groups, having similar needs, should be defined to measure value for these groups (Porter, 2010). Outcomes should be measured over the complete medical condition care cycle, in order to make it possible to improve and optimize the (overall) patient value (Porter, 2008).
10
Although measurements should concern the complete care cycle (Porter, 2009), a decade ago many measures focused on interventions as well as on departments and hospitals (Porter, 2008). Still, the focus shift from volume to value in healthcare remains a slow progress, partially due to the limited outcome measurements performed (excluding survival measurements) that really matter to the patients (Porter et al., 2016). Reasons for this slow progress in the focus shift from volume to value will be discussed in the next section.
Difficulties faced in measuring value
11
specific conditions and patient representatives, is working on standardizations of outcome sets (Porter et al., 2016). Risk factors are outlined by them as well. In the standard sets, patient reported outcome measures (PROMs) are included and they encompass more than fifty percent of the recommended outcomes per set (Lippa et al., 2014). These PROMs are validated instruments and concern the perceptions of patients about their health and the effects of their received healthcare. Currently, the published standard sets by ICHOM cover more than half of the disease burden being present globally (International Consortium for Health Outcomes Measurement, n.d.). Such consistency in collection and reporting of outcome measures is pushing the current limits in comparing between patient populations and therewith improving value (Martin et al., 2015).
Meyer et al. (2012) address another factor that causes difficulty in measuring essential performance measures, namely the increase in mandatory quality measures, as these cause depletion of resources that otherwise could have been devoted to patient value. In the Netherlands, the quality of healthcare is monitored by among others the Dutch Healthcare Authority and the Healthcare Inspectorate (Government of the Netherlands, n.d.). In the USA, healthcare providers need to report quality measures publicly, which are required by multiple external groups (Meyer et al., 2012). However, this number of required reports “has skyrocketed over the past decade and that trend is poised to continue” (Meyer et al., 2012, p. 964). Therefore, Meyer et al. (2012) state that “we must stop the avalanche of an ever-increasing number of mandated quality metrics so we can get to work on using measures that really matter” (Meyer et al., 2012, p. 967). Therewith, Meyer et al. (2012) point to the need for an ideal and always improving measure set in order to strengthen results on both quality and value. Metrics should be guided by the needs of the user regarding performance data. These end-users are according to Meyer et al. (2012) not solely patients, as consumers, the family of the patients, payers and employers are end-users as well.
How to transform healthcare: the value agenda
12
step in order to improve healthcare (Porter & Lee, 2013). Third, for each cycle of care, there should come bundled payments, as this encourages healthcare of high value and teamwork. Fourth, the delivery systems in healthcare should be integrated, in order to eliminate care duplication as well as fragmentation. Fifth, the geographic reach should be expanded, by serving patients not only locally but over a broad geographic area. In this way, value can be increased substantially, as superior providers are able to serve more patients. Finally, an information technology platform should be built in order to enable the previous five agenda items.
The future feasibility of these agenda items cannot be excluded for the Netherlands, as for example: patients can already be served over a broad geographic area, however, they have to pay extra if they choose a healthcare provider that has no contract with their health insurer (Independer, n.d.). Furthermore, there is already the possibility to integrate care to a limited extent as the bundled-payment concept is approved for a few diseases (Struijs & Baan, 2011; Porter & Kaplan, 2016). Moreover, Struijs (2015) states that the bundled payment model in the Netherlands is consistent with the value agenda’s principles by Porter & Lee (2013).
As mentioned before, quite some organizations already started with the transformation to VBHC (Porter & Lee, 2013). The stages achieved by these organizations vary between pilots as well as initiatives performed in individual practices and changes that already involve a couple of items of the value agenda. According to Porter & Lee (2013), transformation to VBHC has to come from the provider organizations and physicians within, as they have the exclusive ability to set up all the interdependent value-improvement steps. Each strategy involves steps that need to be performed sequentially over time (Porter, 2009). Therefore, road maps are necessary, not only to roll out the changes, but in this way actors are also given the time to adjust to the change. Sustained change is a necessary condition for transformation (Bohmer, 2016). Therefore, behavior of individuals, the interactions within the team as well as operations design should all change sustainably. “In practice, health care transformation is a long series of local experiments” (Bohmer, 2016, p. 710).
From measuring to value improvement
13
providers (Porter & Teisberg, 2006). Although multiple outcome dimensions can benefit from improving a single outcome dimension, there also might be tradeoffs in place (Porter, 2010); there can be competing outcomes that should be weighed. The mapping of tradeoffs among different outcome dimensions, as well as searching for ways to make reductions in these tradeoffs, is regarded by Porter (2010) as essential for innovating care.
Shared habits among high-value healthcare organizations
Although organizations that provide high value of healthcare have varying tactics, they have shared habits (Bohmer, 2011). The four common habits are: 1.) specification and planning, 2.) infrastructure design, 3.) measurement and oversight and 4.) self-study. The first habit includes the specification of decisions and activities in advance (Bohmer, 2011). The specification of choices, as well as subgroups and patient pathways does represent an investment in advance planning. The second habit stems from the need to design microsystems (Nelson, Batalden, Godfrey & Lazar, 2011 in: Bohmer, 2011), among others staff and business processes, to match the subpopulations and pathways. The third habit means that high-value organizations not only measure because of external requirements, but they collect more measurements in order to inform staff about the performance (Bohmer, 2011). The fourth habit is that organizations go beyond ensuring consistency of their clinical practices with recent science, as they examine deviance in their own outcomes and care as well (Bohmer, 2009 in: Bohmer, 2011).
The importance of conducting this research
14
METHODOLOGY
In order to answer the research question, an inductive research was performed using the Gioia methodology, which was complemented by two focus group discussions in order to reflect on the findings. In an inductive research, data is collected to discern a pattern within this data (Karlsson, 2016). Conducting an inductive research was appropriate in order to answer the research question, since there is limited, if any, knowledge in literature on the perceptions on VBHC, because VBHC is still a concept in its infancy. In conducting this inductive research, the Gioia methodology was used, which is a systematic approach to grounded theory (Gioia, Corley & Hamilton, 2013). Grounded theory concerns a systematic method for gathering and analyzing data (Strauss & Corbin, 1994). Theory does evolve during the research, because there is a continuous interplay between data collection and data analysis. The focus in a grounded theory research is to unravel experience elements (Moustakas, 1994). These grounded theory researches have often been interview studies (Charmaz, 2011). The grounded theory approach suited well for our research, as the experiences of practitioners helped to give meaning on VBHC, by discussing how VBHC could be applied in practice.
The Gioia methodology was used to enhance systematics, as this method is believed to be balancing two conflicting needs, namely both the inductive development of concepts and meeting high rigor standards (Gioia et al., 2013). The approach “captures concepts relevant to the human organizational experience in terms that are adequate at the level of meaning of the people living that experience and adequate at the level of scientific theorizing about that experience” (Gioia et al., 2013, p. 16). Using the Gioia methodology, prior constructs or theories should not be imposed on the interview participants as an a priori explanation of their experience (Gioia et al., 2013). In this way, new concepts might be discovered rather than that existing concepts get affirmed.
The findings obtained from the analysis of the interview data concerned the perceptions of VBHC practitioners on how VBHC could be applied in practice. Two focus groups categorized the findings (2nd
15
Research setting
The interviews were conducted with employees from three hospitals in the Netherlands and one hospital in the United States, where VBHC is already applied to a certain extent (even though in some cases not explicitly named as such). The interview participants had to have some experience with applying VBHC. This was important as elements of experience were unraveled in order to answer the research question. Both focus group discussions, consisting of participants interested in applying VBHC, were conducted in a hospital in the Netherlands.
Case description and selection
The unit of analysis of the first part of this research where practitioners were interviewed, is a multidisciplinary team consisting of actors involved in applying VBHC to a certain extent. The cases were selected based on the criterion of ‘having experience with applying VBHC’. Twenty interviews were conducted, where possible with different actors per case, ensuring enough variety in the respondents’ functions. The case names and descriptions are shown in Table 1, as well as the function of the interviewed respondents for each case. Cardiology represents an American team. The five other cases are Dutch cases.
Case name Case description Respondents
Palliative care (abbreviated as Palliative in text)
This team is involved in advanced care planning for elderly and its corresponding care pathway. This team had started because it was noticed that treatment often continued too long, without taking into account what the patient considered important.
1. medical specialist / consultant
Oncology This multidisciplinary team is involved with a care pathway for breast cancer patients, which was set-up to put cancer care at a higher level (with more intensive collaboration). The care pathway is developed from the patient’s perspective, and care is organized around this care pathway.
1. case manager; 2. unit head; 3. quality policy officer; 4. medical specialist; 5. diagnostic medical specialist; 6. medical specialist
Medicines This multidisciplinary team is involved in medicine use. This team both monitors and makes policy on effective medicine use, and consists of members who come into contact with expensive medicines.
1. manager; 2. medical specialist; 3. controller; 4. coordinator care department; 5. manager medical support departments; 6. coordinator sales; 7. medical specialist; 8. pharmacist
Cardiology VBHC plays a major role within the cardiovascular program, as implementing VBHC is the biggest role of this program. It is tried to implement evidence-based practice at the best value and the cheapest possible.
1. program manager; 2. medical director of the program; 3. operations manager director of the program Several
conditions (abbreviated as Several in text)
In this hospital, patient/diagnose groups are being defined and it is examined how VBHC can be introduced and worked out for these groups. The respondents have experience with VBHC and PROMs, and therefore relatively much knowledge on applying VBHC. The information obtained in this case does not explicitly belong to one condition.
1. manager of quality and safety; 2. manager of medical departments (simultaneously interviewed)
Hematology The respondent of this case is implementing VBHC in the outpatient clinic regarding hematology. The provided care is multidisciplinary, and currently a dashboard is being developed showing benchmark information for a patient.
1. medical specialist
16
The teams of Palliative, Oncology and Medicines were approached for this research because of their performance in their care trajectories. The team of Cardiology was approached because of its well-organized multi-disciplinary interplay, spot-on clinical data feedback, analysis. Several was approached because of the experience with PROMS. Lastly, Hematology was selected as we were referred to the respondent of this case by one closely involved with VBHC within the same hospital. So although in general all cases were selected with regard to VBHC, the specific reason for selecting these cases slightly differed. Therefore, one might expect some elements will be highlighted more in one case than in another, explaining some possible variety between the cases.
Selection focus group participants
A clinician group and non-clinician group were composed for the focus group discussions. This distinction between the groups was made, because the participants’ background can shape their perspective on the elements of applying VBHC. The participants were selected based on the following criterion: ‘the participant should currently consider to apply VBHC’. The clinician group consisted of four medical specialists and the non-clinician group consisted of a sales manager, quality advisor, professor and a coordinator regarding VBHC.
Data collection
Semi-structured interviews were conducted with twenty-one participants (see Table 1). Before each interview started, the participants were asked whether they would give consent to record the interviews. The interview-protocol, which can be found in Appendix B, was set-up in such a way to try to meet the requirements of both thoroughness by anticipating related issues which should be asked, and avoidance of questions that lead-the-witness (Gioia et al., 2013). Therefore, in order to anticipate the related issues, the main summary of VBHC we derived from literature is represented in the (questions in the) interview-protocol. In this way, broad general elements of VBHC were covered in the interview, if the participant had not mentioned them himself. However, the way the interview-protocol was developed, gave the participant space to mention everything he would like to discuss about VBHC.
17
VBHC as well as the short introduction round at the beginning of the session contributed to the group cohesion. This group cohesion was important, as according to Fern (2001) such a group works together and even more, this group cohesion enhances productivity. This collaboration and productivity was necessary for the focus group discussion, as the participants had to do certain ‘assignments’ in a defined time period.
The discussion was informal in nature and focused around a particular topic (namely the elements of applying VBHC in practice), in compliance with the focus group methodology (Wilkinson, 2004). A focus group environment is considered helpful in order for the focus group participants to discuss their opinions, thoughts, ideas, as well as their perceptions on the topic (Krueger & Casey, 2000 in: Onwuegbuzie, Dickinson, Leech & Zoran, 2009). The group discussion lasted for two hours and was based on a ‘schedule’, namely a series of questions/activities which can be found in Appendix C. During the discussion, the researchers moderated and facilitated the group by asking the questions, by ensuring the discussion kept flowing when necessary, and by enabling each participant to participate fully, as prescribed by Wilkinson (2004).
The interview results used for the focus group discussion stemmed from the interviews with Palliative
– 1, Oncology – 1, Oncology – 2, Oncology – 4, Oncology – 6, Medicines – 1, Medicines – 4 and Cardiology – 1. Each element obtained from these interviews represented either the 2nd order themes
or the aggregate dimension, hence main element. As mentioned, the elements were complemented with elements retrieved from literature. The overlapping elements were combined into one element and some elements were slightly adjusted in order to enhance their understandability for the participants. The resulting number of elements was fifty-one, of which the numbering was randomized in order to avoid any influence on the categorization process. Although not all data from the twenty interviews was used, each main element was covered by at least one digital ‘card’ during the focus group discussion. Furthermore, the amount of elements that practically could be used had reached its final stages. Moreover, data saturation was almost reached, especially because the literature and interview data were combined.
Data organization and analysis
The interview recordings were transcribed into text in order to analyze the data. First, both 1st order
analysis and 2nd order analysis were performed. The 1st order analysis concerned “an analysis using
18
and dimensions (Gioia et al., 2013, p. 18). After it was investigated whether the 2nd order themes could
be distilled into aggregate dimensions (Gioia et al., 2013), the aggregate dimensions were developed. This analysis was presented in a schematic way, by making a data structure showing the elements on how VBHC could be applied in practice. After the overall framework showing VBHC elements was presented and discussed, a cross-case analysis was performed. In this cross-case analysis, the results of both Cardiology and Several were compared with the results of Palliative, Oncology, Medicines and
Hematology, as the first have relatively more experience regarding applying VBHC than the latter. The
19
RESULTS
In the first part of this section, the results from the twenty interviews concerning how VBHC could be applied in practice are provided. Therefore, the complete framework showing the elements of applying VBHC in practice is discussed. Subsequently, the differences and similarities between the cases will be discussed in a cross-case analysis. In the second part of this section, the focus groups results regarding the categorization of elements of applying VBHC and the indicated most important elements are provided.
Interviews
Thirteen main elements of applying VBHC in practice
The interview results showed thirteen main elements (thirteen aggregate dimensions) of how VBHC could be applied in practice, when the elements ‘Plan’, ‘Check’ and ‘Act’ are considered as one combined element, as these are part of the same Plan-Do-Check-Act cycle. The main elements are categorized in four clusters: patient value, costs aspects, organization of care, and steering (information). Figure 1a-1d, representing the overall framework, shows the 2nd order themes and
aggregate dimensions per cluster. The figures representing the 1st order concepts, 2nd order themes
and aggregate dimensions for each individual case can be found in Appendix D. Below, the main elements of how VBHC could be applied in practice are discussed per cluster.
Cluster 1: Patient value
Figure 1a. Combined results of all six cases for cluster ‘patient value’. (Abbreviations: O=Oncology,
20 Focus on patient value (improvement)
One element specific for applying VBHC is the focus on (improving the) value for the patient, which came forward in Oncology, Medicines, Cardiology and Several. This focus is important, as for example
Oncology – 2 states: “What is of value for us does not always mean that that is of value for the patient as well”. The main focus in healthcare should for example not solely lie in processes or cost reduction.
“If you just focus on [currency], efficiency goes down, complications go up, and the overall cost in care it goes up, and value goes down” (Cardiology – 2)
“Process indicators can be very important for your internal steering and you don’t want to lose sight on them. But […] if you think from the patient’s perspective, then just the outcomes are actually very important of course” (Several – 2)
Cluster 2: costs aspect
Figure 1b. Combined results of all six cases for cluster ‘costs aspect’). (Abbreviations: O=Oncology,
M=Medicines, C=Cardiology, S=Several.)
Cost control
21
“Evaluate whether it works […] if it […] has no effect […] it all doesn’t make sense to continue a certain medication that generates nothing for the patient and […] extra costs” (Medicines – 2)
Reimbursement for activities that add value for patient
In order to apply VBHC, reimbursements for value-adding activities are important. Although this element is not completely implemented yet, it is deemed important for applying VBHC based on the respondent’s experience. Currently, reimbursements are provided for activities performed in a hospital, so not for non-treatment and non-face-to-face contact. Providing customized care requires relatively longer conversations and this time should be appreciated in monetary terms as well.
22 Cluster 3: organization of care
Figure 1c. Combined results of all six cases for cluster ‘organization of care’. (Abbreviations:
23 Provide care within a care pathway
The provision of care within a care pathway is specifically related to VBHC by multiple respondents.
“We said every time: we make the care pathway from the patients perspective and that is value-based healthcare” (Oncology – 2)
“You should arrange care around [care pathways] to get it as value-based as possible” (Medicines – 4)
(Intensive) team collaboration
(Intensive) team collaboration is part of applying VBHC as well. Multidisciplinary discussions are performed by Palliative, Oncology and Medicines and specifically multidisciplinary collaboration by
Oncology and Medicines. One example of the need for intense collaboration is that one should
collaborate in order to be able to go to the patient if deemed necessary, to increase value for the patient.
“[Colleague] calls me: [the patient] sits here and you have a call appointment, but they want to see you shortly […] so that means you have to collaborate intensive and have to know each other for that. I mean, I trust his/her judgement, so I just know it is better to bend over backwards now to go see that patient shortly, while I don’t have office hours. Because the patient has cancer, of course you want to start as soon as possible” (Oncology – 4)
Patient involvement
One can involve patients during and after implementation of VBHC. Patients can be involved in setting up the care pathway, but as well in deciding on the outcomes to steer on. This is important, as doctors do not always know what patients want. The following example underlines this:
“What I think the patient wants and what they say they actually want or how do they respond at what we’re doing is, we’re so far of base” (Cardiology – 2)
Customized care
24
whether the patient is able to handle the care. Therefore, one should take the personal factors of the patient into account, like age.
“We have a conversation with a patient and we look at what he/she suffers most from, but then either on a physical, mental, spiritual, social level. And especially what his/her wishes and goals are. And then we look at what fits that person” (Palliative – 1).
Shared decision-making is applied by Oncology and Medicines as part of customized care within VBHC. This means the patient and clinician decide together on the treatment plan for the patient. So the patient can choose from options provided by the clinician based on what is of value for him. This is important, as the individual patient might have different wishes than the general patient. Therefore, non-treatment can be an option as well if the patient does not wish to receive a certain treatment.
“You can come up with a plan, but perhaps that plan is not a good idea for that patient or not feasible” (Oncology – 6).
“You notice that if you take time for it, people dare to decide to say: I think I will not start the that treatment. […] So for me that is value-based healthcare” (Oncology – 4)
Advanced care planning is another way to provide customized care, applied in Palliative. It involves a long conversation with a patient to know his/her wishes and goals. Thereafter, a plan with scenarios is made with the corresponding wishes of the patient. This is regarded as important for cost savings and as valuable for the patient as this enhances his/her quality of life.
Evidence-based practice
In Oncology and Cardiology, the link between applying VBHC and evidence-based practice is explicitly stated. In Cardiology they try to implement evidence-based practice at the best value, so the cheapest possible. An example of the need to apply evidence-based practice is as follows: when evidence-based practice is not taken into account during shared decision making, all kind of treatments could be provided based on the wishes of the patient, while these do not make sense from a medical point of view.
25 Systematic choices
‘Systematic choices’ is an element highlighted in Cardiology. An example of this element is the use of an electronic discharge program. By performing things the same way, variation goes down. Furthermore, for example when clinicians make use of a decision tree, costs can be reduced.
“I would say that we are a little bit further along the journey for value-based care and for us, where it starts is in assuring the proper, the right test in the right setting at the right time. […] Oftentimes […] they [physicians] just don't know which is the right test. And so, we design our order sets, our electronic medical record, to help guide them based on current assessment of a patient. It's a decision tree, that leads them down to the proper test, and when we do that, we save a lot of money […] So we have what we call the order set […] when you login to the medical record […] you have to answer some questions based on the clinical presentation of the patient in front of you, and it will guide you to the right test” (Cardiology – 1)
Process improvement
Process improvement is part of applying VBHC as well. When value is the main focus, everything that does not add value for the patient, hence waste should be eliminated. Variation reduction or elimination is regarded as important as well in Cardiology. For example, one specific way to increase value is by variation reduction as this reduces costs.
“When we improve our care, we see decreased costs just by simply decreasing variation” (Cardiology – 3)
There are other ways as well to add value by means of process improvements, namely by more efficient care for the patient, reduction of time investment for patients and by aligning processes:
“Should someone necessarily go to the hospital? […] there are also patients that indicate: call me [regarding results]. I do that often” (Oncology – 4)
26 Cluster 4: Steering (information)
Figure 1d. Combined results of all six cases for cluster ‘steering (information)’). (Abbreviations:
27 Plan, Check and Act
These three are discussed together, as these elements belong to the same cycle, namely the PDCA cycle, mainly regarding patient outcomes, patient experiences and costs. Indicators that are of value for patients should be established (‘Plan’) when one applies VBHC. Within Several the importance of standardization is highlighted, as it is necessary that indicators, parameters and questionnaires that will be send to the patient are standardized to enable comparison.
“We [all involved medical specialists) have devised indicators with each other of which we think and have heard from patients, that they think are important” (Oncology – 2).
“We just said: What do we think are the most important [indicators]? […] You can better put some energy in following some things and do that in a systematic way” (Oncology – 4)
In order to evaluate (‘Check’) the performance when care is provided (the provision of care resembles ‘Do’ in the PDCA cycle), mainly costs, patient experiences and patient outcomes should be monitored and measured. One specific way to evaluate is mentioned in Oncology, Cardiology, Several and
Hematology, namely by means of benchmarking. One can improve based on, and steer in response to
the obtained information (hence ‘Act’).
“On the basis of PROMs and PREMs you will look: where do I have to make adjustments” (Oncology – 4)
Use of dashboard
A dashboard is used as a tool when applying VBHC and especially for the PDCA cycle. The dashboard is used for indicators and for costs. It can visualize the results and helps to evaluate where to act upon.
“And from the dashboard you take into account: what do we see in the outcomes, where should we act upon, and the next time […] we can see already if there are improvements. […] such a dashboard showing much more real-time results, helps in the improvement cycle, in the PDCA” (Several – 2)
“We can track over time [with a dashboard] whether the costs increase, decrease, and whether the interventions we carry out have influence on the costs” (Medicines – 8).
Support by support departments
28
Oncology and Hematology. Quality employees should be involved in establishing indicators, and data
support is necessary to get insight and entrance to data as well as for receiving the necessary programs.
“What do we want to have […] in the dashboard. But therefore you need a program […] that supports that” (Several – 2)
Cross-case analysis
In the cross-case analysis, the cases with relatively more experience regarding applying VBHC (Category A: Cardiology and Several) were compared to the cases with relatively less experience with applying VBHC (Category B: Palliative, Oncology, Medicines and Hematology). Table 2 shows the main elements mentioned per case. The results including the 1st order concepts per respondent can be found in
Appendix D.
Cluster Main element Category A:
relatively more experience with applying VBHC
Category B: relatively less experience with applying VBHC
C S P O M H
patient value focus on patient value (improvement)
x x x x
costs aspect cost control x x
reimbursement for activities that add value for the patient
x x
organization of care
provide care in a care pathway
x x
(intensive) team collaboration x x x x
patient involvement x x x x x customized care x x x evidence-based practice x x systematic choices x process improvement x x x steering (information)
plan & check & act (from PDCA cycle)* x x x x x use of dashboard x x x support by support departments x x
Table 2. Cross-case comparison. (Abbreviations: C=Cardiology, S=Several, Palliative=P, O=Oncology,
M=Medicines, H=Hematology.)
* mainly regarding patient outcome and patient experience indicators, but also regarding costs.
29
main elements were highlighted by both cases of category A and, although to a smaller extent, were prominently highlighted by the category B cases as well. Main elements of the ‘costs aspects’ cluster were less highlighted in both categories.
‘Reimbursement for activities that add value for the patient’ is only mentioned by two respondents within category B. This element was deemed important although it is not yet implemented in the Netherlands and in the United States. “Right now, hospitals and doctors get paid to do things to people
[…] if you don’t operate, if you don’t do an X-ray: You don’t get paid” (Cardiology – 1). However, the need for these reimbursements is recognized by health insurers as: “they are looking: what are
different ways of funding that stimulate that you apply the principles of value-based healthcare” (Medicines – 6).
‘Provide care in a care pathway’ and ‘customized care’ were only explicitly mentioned in the cases of category B. Furthermore, ‘systematic choices’ was only clearly highlighted in Cardiology, a category A case. So in the United States, the element ‘systematic choices’ (implying variation reduction) is part of the team’s current practice to improve value, while in the five Dutch cases none of these elements came forward.
Focus group
Categorization of elements
The clinicians constructed five categories with VBHC elements used during the focus group session and attributed the following names to these categories: ‘process / care pathway’, ‘patient perspective’, ‘cost perspective’, ‘care provider perspective’ and ‘data (management)’. The non-clinicians constructed seven categories: ‘process efficiency’, ‘chain-oriented work according to guidelines and standards’, ‘information for steering and improving’, ‘patient value central’, ‘outcome funding and financial incentives’, ‘personalized care’ and ‘clinical governance culture’. The tables including both the category names and their corresponding elements of applying VBHC can be found in Appendix E.
Selection of most important elements
30
31
DISCUSSION
Summary
The results show thirteen main elements of how VBHC could be applied in practice. These elements are based on the perceptions of practitioners with experience regarding applying VBHC. The main elements of applying VBHC are categorized in four clusters. Table 3 shows these four clusters with their corresponding elements.
Cluster Main elements
Patient value focus on patient value (improvement)
Costs aspect cost control
reimbursement for activities that add value for patient
Organization of care provide care in a care pathway
(intensive) team collaboration patient involvement
customized care evidence-based practice systematic choices process improvement
Steering (information) plan & check & act (especially regarding outcome and/or experience indicators and costs) use of dashboard
support by support departments
Table 3. Summary of results on main elements of applying VBHC.
Furthermore, a clinician group and non-clinician group considering to implement VBHC reflected during a focus group session on the results from eight interviews regarding elements of applying VBHC (both 2nd order themes and aggregate dimensions) combined with elements retrieved from literature.
Both groups were asked to categorize these elements. The result of categorization by the non-clinicians was rather similar to the clusters constructed by the researchers, as will be discussed in the Interpretation section. Furthermore, the focus groups indicated the following elements as very important when applying VBHC: to choose and adapt ICHOM sets, to learn from relating outcome data to costs data, and to evaluate - and steer in response to patient experiences, as well as to patient outcomes. Customized care came forward as a very important element as well.
Interpretation
32 Interpretation of the thirteen main elements
The complete framework (see Figure 1a-1d) shows the overall results regarding how VBHC could be applied in practice. All thirteen main elements of applying VBHC are discussed below. Thereafter, the findings of the cross-case analysis are discussed.
Cluster patient value and cluster costs aspect
First of all, ‘focus on patient value (improvement)’ is completely in line with VBHC in general, as for example Porter (2010) and Porter & Lee (2013) describe the main focus in healthcare delivery should be the value for the patient. ‘Cost control’ clearly came forward, as for example Medicines – 5 mentioned: “there is critically monitored: does the therapy work, yes or no? Because, when it does not
work […], you have to quit it, because it is not effective and much too expensive to continue”. So, during the interviews came forward that costs should be controlled by for example cost reduction when this does not affect outcomes, but the focus did not lie on cost reduction without regard to outcomes. This is important, as Porter (2010) warns for the danger and the self-defeating effect when outcomes are not regarded when one decides to reduce costs. Cardiology – 1 adds: “if we're ever going to control
costs, we […] have to do things the same way, as long as it's the best way […] When you allow this massive variation throughout the organisation, you never know what's working and what's not”.
‘Reimbursement for activities that add value for patient’ is an important element that helps to focus
on value. Porter (2009) acknowledges this need when stating such a reimbursement system should align the interests of every stakeholder that is involved in improving patient value, so bundled payments should cover a medical condition’s full care cycle. Regarding these bundled payments,
Oncology – 5 added: “in the ideal world, because it is not yet that far”. Reimbursements should come
as well for non-treatment and for non-face-to-face contact when this is of value for patients, as
Oncology – 4 noted for example that he/she does not profit when he/she calls patients “because it is no face-to-face contact […] but it is good care”.
Cluster organization of care
33
make choices regarding for example the setup of the care pathway or outcome indicators, while customized care refers to care provided at the individual level in consultation with the individual patient. Although these elements are in line with the need to get consumers more involved in healthcare as stated by Porter (2009), the 2nd order themes of patient involvement and customized
care, which are for example advanced care planning and shared decision making, show more specifically how this can be performed. However, Barry & Edgman-Levitan (2012) remark that the role patient preferences play in decision-making depends on whether or not there is a clear superior path to go. When there are multiple possible paths (including non-treatment as an option), value is added by patient involvement in the decision-making process.
‘Evidence-based practice’ is regarded as important, as for example Oncology – 6 noted: “I think it’s something the patients find important where we have to steer on. But also on what is proved scientifically. So you should not only say: ‘patients think this is important’ […], you also have to just look at what works and what makes sense”. Porter & Teisberg (2006) noted that competition should be
based on results and not just practice in line with accepted care standards. ‘Systematic choices’ is in line with the first habit of healthcare organizations providing high value defined by Bohmer (2011), namely planning and specification whereby core clinical decisions are based on established explicit criteria. ‘Process improvement’ is according to Porter (2010) an important tactic, but no substitute for cost and outcome measurement. However, in the cases where process improvement came forward, the emphasis did not lie on process improvement, but patient value and/or outcomes.
Cluster steering (information)
‘Plan’, ‘Check’ and ‘Act’, mainly regarding patient outcomes and patient experiences, but also regarding costs, are part of the PDCA cycle1. Measuring and evaluating outcome indicators (for
example PROMS) as well as costs are key items in the literature by Porter & Teisberg (2006), Porter (2008; 2009; 2010), Porter & Lee (2013) and Porter et al. (2016), whereas experience indicators do not clearly come forward as an important element of VBHC in these articles. For example, when Porter (2009) and Porter & Teisberg (2006) describe the need of measuring experience, they refer to experience of physicians regarding the treatments they perform. Furthermore, Porter & Teisberg (2006) explicitly mention that health outcomes, instead of service experiences, should determine the choice for a health plan. Also, in the article by Porter et al. (2016) a distinction is made between indicators regarding patient experience and regarding outcomes, and argued the measurement of outcomes that matter to patients is limited while these are essential performance measures. So the
34
patient experience indicators are not specified clearly as important when applying VBHC. Therefore, the retrieved PDCA cycle regarding patient experience indicators is an interesting finding for and addition to how VBHC could be applied. The addition of PREMs is also mentioned by others, as for example Elf, Flink, Nilsson, Tistad, von Koch & Ytterberg (2017) argue that, when complex as well as long-term conditions are managed, PREMs are important to incorporate as an outcome measure. The
‘use of dashboard’ came forward as tool for applying VBHC. Therefore, the necessary information
technology should be in place, which is in line with the sixth agenda item by Porter & Lee (2013), as they mention the need for an information technology platform for applying VBHC. This ‘support by
support departments’ regarding information technology came forward as well during the interviews.
Concluding, none of the elements are in conflict with existing literature on VBHC. However, the framework showing thirteen main elements and their corresponding more explicit examples and practical tools of how VBHC could be applied (the 2nd order themes) provides an important overview
of how VBHC could be applied in practice (see Figure 1a-1d).
Cross-case
Of the thirteen main elements, three patient-centered elements, namely ‘focus on patient value’, ‘patient involvement’ and ‘Plan & Check & Act (from the PDCA-cycle)’ mainly regarding patient outcome indicators and patient experience indicators, were highlighted by both cases with relatively more experience with applying VBHC. These elements also came forward in at least half of the cases that have less experience with applying VBHC. Therefore, we see a broad confirmation on these elements, while the elements in the cluster ‘costs aspects’ were less highlighted in both categories. This suggests that the respondents regard VBHC especially as a patient-centered approach instead of a cost-centered approach.
Some elements were not broadly confirmed, although they were emphasized in some cases. For example, ‘customized care’ was emphasized by many respondents in three out of four cases with less experience on applying VBHC (see Appendix D), but not explicitly highlighted by the cases with relatively more experience. However, the fact that an element is not highlighted by some cases, does not imply that the element would not be of added value. For example, it could be that an element is not applicable for the specific team or that it is regarded as self-evident and therefore not specifically mentioned.
35
elements did not came forward in the Dutch cases (see Table 2). This raises the question whether there are differences in how VBHC is applied between nations.
Interpretation categorization process
Regarding the categorization of the elements, we see both focus groups, clinicians and non-clinicians, defined similar category names. In Table 4, the similarities between the names provided by the focus groups and the cluster names, including the resulting common categories, are presented. Two category names, namely ‘care provider perspective’ and ‘clinical governance culture’, show relatively less similarities with other category names and are therefore not included in Table 4.
Cluster interview Clinicians Non-clinicians Resulting common categories
Patient value Patient perspective Patient value central Patient
Costs aspect Cost perspective Outcome funding and
financial incentives
Costs Organization of care Process / care
pathway
Process efficiency Chain-oriented work according to guidelines and standards
Personalized care
Organization of care
Steering (information)
Data (management) Information for steering and
improving
Data/information (for steering)
Table 4. Similarities between category names.
This comparison resulting in four common category names provides the possibility to compare the division of elements by both focus groups with our division of elements in the results section. The comparison of the way elements were divided in categories can be found in Table 5.
Similarities in division Amount of cards
Interviews, clinician group and non-clinician group 11
Interviews and clinician group 4
Interviews and non-clinician group 17
Clinician group and non-clinician group 5
No similarities between groups 5
Table 5. Comparison of division of elements over the common categories.2
2 During the focus group, fifty-one cards were used. Of these fifty-one cards, nine cards concerned elements
36
This comparison shows some notable findings. First, our categorization process during the analysis of interview data shows more overlap with the categorization process of the non-clinicians than with the categorization process of the clinicians. This shows that the categorization process might be influenced by the background of the participants, as the categorization process from the non-clinicians shows more similarities with our way of categorizing during the analysis of the interview data. Second, when comparing the categorization process of the non-clinicians with our categorization process during the analysis as a validity-check for the internal analysis of this research, we see an overlap of 74% in the categorization process of the cards3. Concluding, there is a remarkable consensus on our performed
analysis when comparing this analysis with the element categorizations of the non-clinicians.
Interpretation of indicated important elements
Both focus groups were asked to indicate the importance of elements of applying VBHC. The following elements were indicated as very important by both focus groups: 1.) ‘the choice for and adaptation of ICHOM sets: sets with standardized indicators for different diseases’, 2.) ‘evaluation of - and steering in response to patient experiences (for example Patient Reported Experience Measures / PREMs)’, 3.) ‘evaluation of - and steering in response to patient outcomes (for example Patient Reported Outcome Measures / PROMs)’ and 4.) ‘learning by relating data on outcomes to data on costs of care’.Striking is that these elements represent the PDCA cycle mainly regarding patient outcome indicators, patient experience indicators, but costs as well, as the first element refers to ‘Plan’ regarding outcome indicators, the second and third element to ‘Check’ and ‘Act’ regarding outcome and experience indicators, and the fourth to ‘Check’ regarding the balance between outcomes and costs. Although these elements do not explicitly represent each cycle step, ‘Check’ and ‘Act’ cannot take place without first planning what to measure (‘Plan’) and providing the care (‘Do’).
Even so notable is that the first, second and third element of the previous paragraph (the choice for and adaptation of ICHOM sets, evaluation of - and steering in response to patient experiences, and to patient outcomes) represent three of the five elements that were similarly divided by the clinicians and non-clinicians, namely in the common category: ‘patient’, while we placed it in the common category ‘data/information (for steering)’. Both classifications are arguable, as these three elements
3 Of the fifty-one used cards during the focus group, nine cards concerned elements exclusively retrieved from
37
relate to steering information on what is of value for the patient. So, the main perspective when applying the PDCA cycle mainly regarding patient outcome and patient experience indicators, but costs as well, should be the patient’s perspective, which is in line with the core focus of VBHC, namely high value realization for patients (Porter, 2010). When analyzing the results more into depth, another element was highlighted as very important by both subgroups, namely ‘customized care’. The clinicians indicated ‘customized care with care pathway’ as one of the most important elements and the non-clinicians indicated ‘shared decision making’ and ‘advanced care planning’, both part of the main element ‘customized care’, as very important elements.
Concluding, when combining both the interview and focus group results, we find a broad confirmation for perceiving the element ‘PDCA cycle’ regarding patient outcome and patient experience indicators, but costs as well, as an important element of applying VBHC. ‘Customized care’ came forward as important as well. Furthermore, we found a broad confirmation for the importance of focusing on patient value and/or performing activities that are of value from the patient’s perspective. Therefore, the elements should be applied especially from a patient-centered point of view.
Interpretation focus group evaluation
When summarizing the participants’ responses (see Appendix F for the focus group evaluation), the focus group discussion is perceived as useful in the thinking process about whether or not to apply VBHC, and the Digitable is perceived as a nice tool for the reflection by a focus group. So especially the process was appreciated (with making use of the Digitable) whereby the content (including elements of our framework) seemed helpful. Therefore, the set-up of the focus group discussion is regarded as a valuable method for (future) VBHC practitioners. The complete schedule for the focus group discussion can be found in Appendix C.
Theoretical implications
38
to literature, because as far as the authors know, no distinction in importance of VBHC elements has taken place in literature. Of these indicated most important elements, the PDCA cycle regarding patient experience indicators adds to literature by Porter & Teisberg (2006), Porter (2008; 2009; 2010), Porter & Lee (2013) and Porter et al. (2016), as they do not specifically highlight patient experience indicators as of great importance for applying VBHC. Even so, the more specific examples of customized care (among others shared decision making) add to existing literature on applying VBHC, as they show how customized care could be applied in practice.
Managerial implications
The interview results regarding elements of applying VBHC provide an overview that can be used by (potential) VBHC practitioners in order to know which elements are suggested by other practitioners as part of applying VBHC. In this way, they can reflect on whether or not they would like to apply certain elements. Furthermore, the focus group assignments with making use of the backgrounds and VBHC elements on the Digitable can be performed by (potential) VBHC practitioners as this process is perceived successful by clinicians and non-clinicians in the thinking process of whether or not to apply VBHC. Enough time for the (plenary) discussion at the end of the focus group session was suggested as desirable as well. Therefore, this research does not solely provide insight into the content of the perceptions on VBHC by providing an overall framework of VBHC elements, but does also provide a method that gives (potential) VBHC practitioners a shared awareness of VBHC and is of added value in their thinking process of whether or not to apply VBHC.
Limitations
The first limitation regards the case selection. As no organization fully transformed to VBHC yet (Porter & Lee, 2013), none of the selected cases represents a multidisciplinary team completely applying VBHC as defined in the value agenda proposed by Porter & Lee (2013). This may have influenced the perceptions on applying VBHC by its practitioners, as they only have experience with applying certain elements of VBHC instead of VBHC in its entirety. Therefore, the variety of cases we selected for this research provided a broader overview of VBHC elements than if we would not have selected varying cases.
39
to go more into depth and to improve the validity, the focus groups reflected on the findings regarding applying VBHC.
Because for certain cases only one interview could be conducted, less variety in perceptions on VBHC might have come up in these cases than if more interviews were conducted. The lack of variety in functions represented in these cases might have limited the completeness of perceptions on the application of VBHC for the relevant cases.
Although the cases were selected because the teams applied VBHC to a certain extent (note: some did not explicitly name their approach as such), not every interview participant was aware of the concept VBHC at the beginning of the interview. This might have influenced the results to a limited extent. In order to decrease this limitation and to ensure the concept VBHC was represented well during the interview, before the data collection started a short summary of VBHC was included later in the interview-protocol, on which the interview participant could elaborate.
Regarding the focus group results, the limitations are threefold. First, the categorization of elements and the indicated most important elements for applying VBHC can be influenced by the limited time frame for the activity and the missing interview context, which might have affected the reliability of the focus group results to a limited extent. However, although the timeframe was ambitious, it was doable. Second, both the clinicians and non-clinicians groups were not able to categorize all fifty-one elements. This might have affected the categories they constructed and the list of indicated most important elements. Third, not all elements of applying VBHC stemming from the twenty interviews were included in the digital ‘cards’. Therefore, it could be that other elements from the overall framework we provided might have been deemed important as well by both focus groups, if these were included during the focus group discussion.
Future research
For this research, clinicians and non-clinicians working in hospitals were interviewed, but future research can investigate the perceptions held by other stakeholders, among others patients and health insurers. Even more, perceptions might differ per country. An example for this possibility is that ‘systematic choices’ was highlighted in the American case as element of applying VBHC, but in none of the Dutch cases. Therefore, it seems interesting to investigate the different perceptions across countries. This might improve the interpretation of literature on this topic.
40
possible conflict between societal value and individual value (as the value for the individual patient might increase by increasing costs that much relative to outcomes, that the value in the value equation decreases). Therefore, a suggestion for future research is to investigate the issues and conflicts that arise when applying VBHC. Even so, future research can investigate facilitators and barriers for implementing VBHC, which came forward as well during the interviews. When the issues and conflicts, as well as facilitators and barriers are studied more into depth, we can learn even more from the experiences of practitioners, hence local experiments, regarding implementing and applying VBHC.
Conclusion
41
REFERENCES
Barry, M. J., & Edgman-Levitan, S. (2012). Shared decision making—the pinnacle of patient-centered care. New England Journal of Medicine, 366(9), 780-781.
Bohmer, R. M. (2011). The four habits of high-value health care organizations. New England Journal
of Medicine, 365(22), 2045-2047.
Bohmer, R. M. (2016). The hard work of health care transformation. New England Journal of
Medicine, 375(8), 709-711.
Charmaz, K. (2011). Grounded theory methods in social justice research. The Sage handbook of
qualitative research, 4, 359-380.
Elf, M., Flink, M., Nilsson, M., Tistad, M., von Koch, L., & Ytterberg, C. (2017). The case of value-based healthcare for people living with complex long-term conditions. BMC health services research, 17(1), 24.
Fern, E. F. (2001). Advanced Focus Group Research. Thousand Oaks: Sage Publications.
Fisher, E. S., Wennberg, D. E., Stukel, T. A., Gottlieb, D. J., Lucas, F. L., & Pinder, E. L. (2003). The implications of regional variations in Medicare spending. Part 2: health outcomes and satisfaction with care. Annals of Internal Medicine, 138(4), 288-298.
Gioia, D. A., Corley, K. G., & Hamilton, A. L. (2013). Seeking qualitative rigor in inductive research: Notes on the Gioia methodology. Organizational Research Methods, 16(1), 15-31.
Government of the Netherlands. (n.d.). Monitoring the quality of healthcare. Retrieved from https://www.government.nl/topics/quality-of-healthcare/monitoring-the-quality-of-healthcare Independer. (n.d.). Bij welke zorgverleners kun je terecht in 2017?. Retrieved from https://www.independer.nl/zorgverzekering/fi/vergoeding-zorgaanbieders
International Consortium for Health Outcomes Measurement. (n.d.) Our standard sets. Retrieved from http://www.ichom.org/medical-conditions
Karlsson, C. (2016). Research in operations management. In Research Methods for Operations
