MSc Thesis
Healthcare reform: Identifying important
elements in moving towards a value based
healthcare system
!
Author: R. A. Groote (s 2341298)
First supervisor: prof. dr. ir. C.T.B. Ahaus
Assessors: prof. dr. ir. C.T.B. Ahaus, prof. dr. J.T. van der Vaart, A.C.
Noort, MSc
Date of submission: Monday February 19
th, 2018
Abstract: Although the concept of value based healthcare has recently
started to gain traction in academics as well as in practice, it still
remains unclear as to how a value based system can be implemented
and maintained. Due to a lack of historical precedent in academic
literature, we aim to identify the concepts and practices that are most
useful in moving towards a value based healthcare system by
1. INTRODUCTION
In recent years, the principle of value based healthcare has started to gain some
traction amongst healthcare providers, as well as research institutions. However, to this day, a clear definition of the concept and a guide to achieving a functional value based health care system have proven to be difficult to construct. So far, value based health care has probably been best defined by Porter (2009), who stated that, for a healthcare system to be sustainable, it must mainly focus on optimizing the ratio of health outcomes achieved per dollar spent. A positive outcome of a healthcare procedure (e.g. a successful treatment to a disease or an injury) would be the ‘value’, which is in this case experienced by the patient. Porter (2010) later emphasized that, in optimizing this ratio, the value of the patient should be the most important, as opposed to focusing solely on cost reduction, which appears to have been the case over the last years. Although reducing costs ultimately contributes to a higher value as defined by Porter (2009), a sole focus on cost reduction often leads to neglect of improving patient outcomes, which can ultimately lead to a decrease in overall value (Porter, 2010).
Therefore, it is necessary to address this issue, and investigate how value can play a central role in healthcare, whilst balancing cost reduction with improvement of outcomes. Thus far, several researches have been conducted in regards to concepts such as quality of care (e.g. Popovich, 1998), performance measurement in a healthcare context (Meyer, Nelson, Pryor, James, Swensen, ... & Hunt, 2012), and other measures related to improving
performance in healthcare organizations. Many of these concepts can be categorized under improving health outcomes, and therefore (potentially) improving value. However, no clear guidelines of how to organize a system that delivers value based healthcare have yet been researched. We therefore believe that there is a critical need to address this, since defining value in healthcare is one thing, but this does not automatically provide a clear way to move towards a healthcare system that is value based at large.
professionals on value based healthcare, in order to gain a greater understanding of the concept, and how it can best be implemented in practice. Porter & Teisberg (2006) have already made several suggestions regarding the move to a value based system. These include proposed actions for healthcare providers and professionals, as well as suggestions focused on the outcome for patients. We therefore want to examine if these (and other) suggestions and practices are used in practice, and if experts in the healthcare sector perceive these
suggestions to be as relevant as researchers suggest they are, or if they feel that priorities should lie elsewhere. The main question this research aims to answer is therefore: which
concepts and practices are most important in moving to a value based healthcare system?
We will attempt to answer this question by consulting a panel of healthcare experts in a Delphi study, in order to identify which concepts and practices they deem most important. The list of concepts and practices will be constructed through a literature review of existing researches, which have been attributed to value in healthcare delivery. These concepts will be tested with a panel of healthcare experts that have at least some experience in an organization that has recently started to adopt some form of value based healthcare, or in consultancy, which other healthcare providers (can) use to enquire about practices related to value based healthcare. This way, we hope to gain understanding of how healthcare experts perceive value based healthcare, and what they perceive to be the best way to implement it. Since historical data is not exactly in abundance, we believe a Delphi study is a valid way to ‘set the tone’ in value based healthcare research
In this paper, we will first provide a theoretical background of the topic of value based healthcare, as well as topics related to value based healthcare. Out of this theoretical
2. THEORETICAL BACKGROUND
In this section of the report, we will assess the current state of literature related to value based healthcare. From this, we will eventually extract our constructs and practices that will be used in the Delphi study. The specifics of the Delphi study will be addressed in the methodology section.
2.1 Value based healthcare
Although the concept of value in healthcare has been mentioned before (Porter & Teisberg, 2006), the specific concept of value based healthcare was first ‘materialized’ by Porter (2008). He argued that the need to address this issue came from the fact that the current healthcare system (in the USA) is not sustainable, and in order for it to be sustainable, its focus should shift into several areas. As stated before, his first work regarding value based healthcare, and therefore the recommendations in it, was drawn up after analyzing the current state of the USA’s healthcare system. However, despite being aimed at the USA’s current system, many proposed changes are universally applicable. Since then, Porter has published several more works on the topic of value based healthcare, in which he has made several recommendations. Some of his earlier recommendations, such as attempting to make
individual health insurance more affordable, and attempting to eliminate the threat of people not being able to afford insurance because their employer does not provide any, are aimed at achieving universal healthcare insurance coverage. (Porter, 2009)
in large part drawn from an earlier report (Porter & Teisberg, 2006), in which the authors stress the fact that providing such value should be achieved mainly through improving quality, rather than solely trying to cut costs, since this is a more sustainable approach. However, several definitions still exist regarding what exactly is defined as high quality of care, and how this should be measured. Therefore, it is also still somewhat unclear what the best way to measure how well a healthcare organization is able to provide value based healthcare is. Porter (2009) mentions that it is important that we use the right measures, but not what these measures are, or how they are selected. These issues will be addressed in the upcoming sections.
2.2 Quality of care
The concept of quality of care has been around in literature for a long time. One of the first who defined it was Donabedian (1980), who stated that care is of high quality when it maximizes patient welfare, whilst taking into account the expected gains and losses of undergoing a certain healthcare procedure. This definition of high quality care concerns the welfare of the patient purely in terms of physical and mental fitness, which, in hindsight, seems to be somewhat incomplete.
One of the most recent definition of quality of care, and perhaps even more so one that has been widely adopted is that of the US national Institute of Medicine (IoM). In their book
Crossing the Quality Chasm (2003), they propose a framework in which quality of healthcare
is defined in six ‘dimensions’. The first dimension is safety. The authors argue that, first and foremost, the general safety of patients when undergoing a medical procedure should be guaranteed. The second dimension is effectiveness, which the authors argue mainly relates to avoiding excessive use of healthcare provision by those who do not need it, and the lack of use of healthcare provision by those who do need it. The third dimension is patient
centeredness, which contains the argument that the patient should be involved in the
healthcare delivery process as much as possible (IoM, 2003). Since Porter (2008; 2010) also argues that the patient should play a central role in healthcare delivery, this topic will be discussed more in depth at a later stage in this section of the paper. The fourth dimension is timeliness, which mainly aims for a reduction of waiting times for both patients and
caregivers. The authors argue that a long waiting time by itself constitutes poor quality of care, since a long waiting time may be harmful to patients, and may cause frustration and a loss of faith in the system in medical professionals. The fifth dimension is efficiency, by which the authors mean the reduction (or, ideally, elimination) of waste in healthcare delivery. The final dimension is equity or equality, which addresses the issue that quality of care should not differ between patients because of personal characteristics (e.g. race, age, gender) (IoM, 2003). Since this framework is rather widely adopted, and partially ties in with Porter’s (2008; 2010) work on value based healthcare, we are interested to see how healthcare experts value these six dimensions.
which is ultimately expected to lead to better outcomes (Feeley, Fly, Albright, Walters & Burke, 2010). If, alternatively, the focus would remain on individual procedures, the
outcomes of these procedures may improve, but the overall patient outcome may not see such improvements.
To be able to center care delivery around the full cycle, Porter & Teisberg (2006) suggest the use of integrated practice units (IPUs). IPUs are “co-located,multidisciplinary teams of clinical and nonclinical providers (e.g., casemanagers, social workers) to treat conditions over the full care cycle” (Keswani, Koenig & Bozic, 2016, p. 2100). The latter (Keswani, Koenig & Bozic, 2016) did research into the use of IPUs in a certain care cycle (the treatment of musculoskeletal disease). They found that the use of IPUs can even have its advantages in a non-value based healthcare system, stating that using these teams can lead to “offering more integrated care, engaging patients virtually, [and] addressing modifiable risk factors” (Keswani, Koenig, & Bozic, 2016, p. 2102).
Overall, it can be concluded that there is a general consensus that, in defining quality of care, the outcomes for patients play a central role. In more recent years, focus shifted to combining this with cost minimization. As Porter (2010) mentioned, it is important to realize that, even though both factors should be considered, the main force behind optimizing value should be the quality of care and its outcomes for patients. Solely focusing on cost reduction, with little or no attention being paid to improving outcomes for patient ultimately defeats its own purpose of trying to improve healthcare delivery, since it is unsustainable and ultimately leads to a less healthy population (Porter, 2009). Since Porter (2009) also suggested that, when reforming towards a value based system, it is important that the right outcome measures are used, we shall look into this issue next.
2.3 Measuring value
assessing quality of care is done mostly through process measures. Many healthcare
organizations are judged based on their scores on tests that measure how well they follow pre-constructed guidelines for the process of delivering the care, rather than reporting patient outcomes. The authors also address the fact that such process measures are less fit than, say, outcome measures, to be used as a driver for performance improvement in healthcare (Porter et al., 2016). They argue that this is the case because the outcome plays a central role in Porter’s (2008; 2009; 2010) definition of value in healthcare.
This last statement was backed up by an earlier research by Krumholz, Normand, Spertus, Shahian, & Bradley (2007), who found that using quality measures that relate to the process rather than the patient outcome are less likely to result in outcomes that are desirable for patients. However, a recent study also exists where the authors found that, when using process measures in a care pathway implemented for chronic obstructive pulmonary disease (COPD) care, it positively affected outcomes (more specifically, the 30-day readmission rate) (Vanhaecht et al, 2016). A study in the Journal of the American Medical Association (JAMA) also found that, although rather weak, the use of standardized process measures had a
negative relationship with short-term mortality rates in acute myocardial infarct (AMI, more commonly known as a heart attack) patients, meaning that short-term mortality was lower in cases of high compliance with process guidelines (Bradley, Herrin, Elbel, McNamara, Magid, Nallamothu, ... & Krumholz, 2006). However, they did find that that the use of these
measures was only responsible for 6% of the explained variation in these short-term mortality rates at a hospital level, and therefore suggest that using only these standardized process measures is somewhat unreliable (Bradley et al., 2006).
only 139 (7%) are actual outcomes and only 32 (<2%) are patient-reported outcomes” (Porter et al. 2016, p. 504). Overall, these somewhat conflicting findings make for an interesting discussion. We are therefore eager to find out how and which performance measures can be used, and how experts currently perceive them in practice.
An issue that exists in healthcare quality measurement in general, is the fact that an extremely wide range of measures exists, and that there are no clear guidelines for which guidelines should be used in what situations (Meyer et al., 2012). We therefore feel that this is a necessary issue to address in our study. Meyer et al. (2012) reported that, per 2011, there were more than 700 measures for quality of care that had been approved by the United States’ National Quality Forum. They also reported cases in which healthcare providers designed their delivery of care in such a way that it met all or most of the quality measures that were most important to certain stakeholders. This of course enabled these providers to report results that are desirable to these stakeholders, but often resulted in a failure to deliver desirable outcomes in terms of the needs of their patients. This is of course somewhat contradictory, since Porter argued that the patient should be the focal point of healthcare delivery (2008; 2010), which could lead one to argue that compliance to quality measures should result in desirable outcomes for patients rather than for other stakeholders.
Meyer et al. (2012) argue that this overload of quality measures that is present can cause confusion as to which quality measures are most fit to use in certain situations. As mentioned before, Porter (2009) also pointed out that in the current system, it occurs too often that it is unclear which measures are ‘the right ones’. At a later stage, Porter et al. (2016) propose using a smaller set of standardized outcome measures, that should be adjusted for different major diseases or conditions and their respective care pathways.
As a matter of fact, an institution that works on creating these standardized sets of outcome measures already exists. It is called the International Consortium for Health Outcome Measurements (ICHOM), and it was co-founded by Michael Porter (https://
relatively new, but show promise according to researchers (Porter et al., 2016), we are interested to investigate how they are perceived by healthcare experts.
2.4 Patient centeredness
The concept of patient centered care has been around since before the introduction of value based healthcare, but may play a part in creating a better understanding of the latter. As mentioned, patient centeredness is argued to be a key to providing high quality healthcare (IoM, 2003). A recent literature review on the topic of patient centered care (Kitson, Marshall, Bassett, & Zeitz, 2013) found that, while many different definitions have surfaced over the past several years, these definitions revolve around three common ‘themes’. The authors defined these themes as “patient participation and involvement, the relationship between the patient and the healthcare professional, and the context where the care is delivered” (Kitson et al., 2013, p. 4). Patient centeredness appears to be evident in designing value based healthcare systems, since we already addressed the need to design care delivery around cycles of care for specific medical conditions, and thus around patients.
this process is needed, in order to understand how they can be used in moving towards a value based system (Black, 2013), or how they can be optimally utilized in an attempt to improve quality of care (Marshall, Haywood & Fitzpatrick, 2006).
Another example of patient reported measuring is the patient reported experience measure (PREM). This practice ties in a bit more with the relationship between patient and caregiver dimension we discussed earlier (Kitson et al., 2013). These PREMs are more concerned about measuring patients’ experiences in receiving care (e.g. undergoing certain treatments) (Black, 2013). It assesses, for example, how pleased patients were with the delivered service, and how they were treated, rather than how they felt after treatment in terms of their general health (Black, 2013). An example of a PREM initiative that is currenlty quite prevalent is the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS). The use of HCAHPS (and related/similar initiatives) is not only suggested to expand the scope of quality measurement in healthcare, but is also linked to improving the quality of care (Giordano, Elliot, Goldstein, Lehrman & Spencer, 2010). Although some skepticism still exists around the use of PREMs, a study by Teale & Young (2015) in an elderly care facility found that the use of PREMs provided a method of analyzing the patients’ experience that could ultimately be used to improve its service. The use of PREMs runs the risk of creating a discussion of how important the experience during the delivery of care is, compared to the actual outcome. However, the use of more recent PREM initiatives such as HCAHPS is said to challenge “outdated assumptions that patient-reported data are less reliable and less valid than data obtained from abstraction, medical record review, and
administrative claims” (Giordano, Elliot, Goldstein, Lehrman & Spencer, 2010). Since we are exploring practices associated value based health care, and ranking them in order of
importance, we feel that the use of PREMs has a place in our study.
2.5 Paying for healthcare
structures, the proposed ‘solution’ could prove to be universally applicable, which makes it relevant for this research.
Porter & Kaplan (2014) argue for a system that uses bundled payments. A bundled payment is “a single payment for treating a patient with a specific medical condition across a full cycle of care” (Porter & Kaplan, 2014, p. 4). They also stress that, in order for bundled payments to help achieve a value based healthcare system, they must only apply in case of desirable outcomes. The burden of the cost of negative health outcomes should lie on the healthcare provider, they argue, since the patient cannot be held responsible for an undesirable outcome.
Porter & Kaplan (2014) also stress that, in order for bundled payments to be truly value based, the payments should accurately reflect the costs of care delivered over a full cycle of care. Many current forms of bundled payments also incorporate certain discounts of charges based on past occurrences, which the authors argue is ineffective and unnecessary. Finally, they also argue that providers should be exempt of responsibility of care that is unrelated to the medical condition for which the payment is meant. In essence, the authors argue that the bundled payment should not cover any healthcare delivery that is related to another condition, or possible catastrophic events (such as traffic accidents). Since the bundled payment should be based on the costs of the care delivery for a specific medical condition, Porter & Kaplan (2014) argue that any care delivery that does not contribute to curing that condition should not be covered by the bundled payment.
they will go above and beyond to provide desirable outcomes for their patients, since failing to do so will most likely result in financial losses. Since this is currently mostly still a working theory, we are interested to find out what the expert opinion of this proposition is, and if they believe it could work in practice.
2.6 Performance measurement in operations management
As we have learned from previous sections on this report, the use of performance measures in a healthcare context is still somewhat disputed, in terms of how it should be done exactly. We therefore decided to look into existing literature from a different field of research, to see if the healthcare sector can possibly benefit from existing beliefs practices in these other fields. We chose the field of operations management, because, in essence, the healthcare sector can be considered a part of the service industry. The service industry in general is not only related to operations management, but is very prevalent in existing operations
management literature. Of course, many existing researches in this field regard other
industries (the manufacturing industry for example), and may therefore be less applicable to the healthcare sector. However, some studies (and findings, accordingly) exist which could be ‘translated’ into a healthcare context.
One example of this is the balanced scorecard. The balanced scorecard is a comparison between (traditionally 4) different sets of performance measures (Kaplan & Norton, 2005). Although traditionally used in a business context, “the balanced scorecard lets executives see whether they have improved in one area at the expense of another” (Kaplan & Norton, 2005, p. 1). This approach “emphasizes a balance between the use of financial and non-financial measures” (Hudson, Smart & Bourne, 2001, p. 1096). This could prove to be useful in a value based healthcare context, since it provides an opportunity to analyze whether performance improvement and cost reduction are kept ‘in balance’. Since the goal of value based
The practice that currently exists in healthcare that most closely resembles a balanced scorecard type of approach is the use of clinical dashboards. A clinical dashboard is “a tool set of visual displays developed to provide clinicians with the relevant and timely information they need to inform daily decisions that improve the quality of patient care” (Daley,
Richardson, James, Chambers, & Corbett, 2013, p. 85). A dashboard somewhat resembles a balanced scorecard approach, since it takes multiple sorts of patient quality measurements into consideration, as well as internal performance measurements (Kroch, Vaughn, Koepke, Roman, Foster, Sinha & Levey, 2006). Although such an approach has already been adopted to some extent by some healthcare providers, recent researches found that it is often unclear how they should be used exactly in order to improve performance (Kroch et al., 2006), and that some of these providers have trouble in assessing the exact effects of the use of
dashboards (Dowding, Randell, Gardner, Fitzpatrick, Dykes, Favela, ... & Currie (2015). We therefore believe it can be valuable to consult our panel regarding the use of these dashboards, in order to assess how its members believe they should be used, and to estimate how
important they are in a value based healthcare context.
Another idea regarding performance measurement that is addressed in operations management literature, is that it is necessary to have a link between an organizations strategy, and the performance measures that an organization uses (Wouters & Sportel, 2005). This idea constitutes the “belief that what an organisation should measure depends on what it is trying to achieve” (Moxham, 2009, p.743). A different idea that ties in with this belief is that, after implementing a performance measurement system, the measures that are used should continuously be reviewed and updated in accordance to an organizations strategy (Bourne, Mills, Wilcox, Neely, & Platts, 2000). Although the need to update performance measures is addressed on multiple occasions (Kuwaiti, 2004), research in operations management shows that this idea is often not carried out in practice (Kennerley & Neely, 2002). We are therefore interested to find out if this is also perceived to be important, and if this idea is known and carried out in the healthcare sector.
In designing our study, several precedents regarding the Delphi method were consulted. First, we considered the website of the RAND corporation, who have initially developed the Delphi method as a way to conduct an academic study (https://www.rand.org/ topics/delphi-method.html). Consequently, we reviewed a paper regarding suggested
guidelines for a Delphi study in a healthcare context (Hasson, Keeney & McKenna, 2000), as well as a literary review of several existing Delphi studies in a healthcare context (Boulkedid, Abdoul, Loustau, Sibony, & Alberti, 2011). The latter also played a significant role in the decision to apply the Delphi method for this study, since it addressed the fact that Delphi studies have proven to be useful in developing quality indicators in a healthcare context before (Boulkedid et al, 2011). Finally, in finalizing the details of the design of our study, we considered a previous research that was similar to ours, in the sense that is goal was to identify which elements were deemed most important in developing a new concept in healthcare (Minkman, Ahaus, Fabbricotti, Nabitz, & Huijsman, 2008).
The research was conducted in three stages. First, a list of concepts and practices related to value based healthcare was extracted from our theoretical background, which can be found in the previous section of this paper. The initial list of concepts and practices (elements) is listed in appendix A. Consequently, this list was presented to a panel of experts in a two round Delphi study. The Delphi method was used in order to assess the experts’ opinions in regards to which concepts and practices are most important in moving towards a value based healthcare system. The specifics of the Delphi study will be addressed in the next section of this report. Afterwards, after the Delphi study was completed, the final list of remaining elements was analyzed. From this final list, we attempted to identify areas of focus that can help society move towards a value based healthcare system. Ultimately, we will discuss the practical and theoretical implications of our findings.
3.1 Delphi study
the process repeats itself” (https://www.rand.org/topics/delphi-method.html). We feel that a Delphi study is appropriate for our research because, as stated before, Delphi studies proved to be useful in previous research when identifying constructs and indicators related to
healthcare delivery and its performance (Boulkedid et al, 2011). Additionally, Franklin & Hart (2007) acknowledged that Delphi studies are appropriate (and are often used) when relatively little previous content exists on a certain topic. Martino (1993) also considered a lack of historical precedent a valid condition for a research to make use of expert opinion as a source of information. Since this is the case for value based healthcare, we believe the Delphi method is appropriate to use in our research.
Initial research before the study
The classical Delphi study, as developed by the RAND corporation, started with a so called ‘blank slate’, and made use of four rounds to construct and refine a list of events (which in our case is a list of concepts/practices) (Martino, 1993). However, more recent forms of Delphi studies have also made use of an initial list as a starting point, rather than going into a study without any established context. In these cases, a study of two or three rounds was also deemed appropriate (Martino, 1993). Since the topic of investigation is relatively broad in our case, and several contributions to the topic have already been made in recent research, we believed it was best to provide our panelists with an initial list, in order to put these existing findings ‘to the test’.
Panel size
Thus, we asked a panel of experts to rank our list of pre-formulated constructs and practices according to their opinion of how important they are in moving towards a value based healthcare. Experts were selected based on their experience and expertise in either working for a healthcare organization which has recently started adopting or developing initiatives based on the philosophy of value based healthcare, or selected from academics or consultancy in the healthcare sector.
However, Thangaratinam & Redman (2005) report that, in the past, “panel sizes have ranged from 4 to 3000” (Thangaratinam & Redman, 2005, p. 120), which makes it difficult to estimate an exact panel size for our study. In the end, in order to ensure that we reach the minimum recommended panel size of 7 (Linstone, 1978), we have asked 10 experts to cooperate in our Delphi study. In accordance with existing Delphi guidelines (Martino, 1993; Boulkedid et al., 2011; Hasson et al., 2000), we will ask our panelists to participate
anonymously, in an attempt to eliminate potential bias. All of the 10 potential participants that were approached were informed about the background of the study and about the Delphi methodology. All 10 approached participants either gave a verbal or written statement that indicated that they were willing to participate in the study. Additionally, all participants gave informed consent regarding their participation in the study, as well as the use of the data gathered throughout study, before the start of the first round of the Delphi study.
Round one
Our study consisted of two different rounds that were made up of a similar structure: a ranking section and a reformulating section. We asked our panelists to rank the elements on our initial list in order of importance. Using perceived importance as a means to reach expert consensus in finding quality indicators on a certain topic was acknowledged as a valid technique by a review of past Delphi studies (Boulkedid et al., 2011), and was also used in a previous research in a similar context (Minkman et al., 2008). The ranking system we used was a 1-4 Likert scale, with 1 meaning ‘Very important’, 2 meaning ‘Important’, 3 meaning ‘Moderately important’, and 4 meaning ‘Not important’. Following the example of Minkman et al. (2008), we chose to use a 4-point Likert scale rather than a more traditional 5-point scale in order to prevent panelist from tending to favor a middle ground option when unsure about an element. This way, the panelists were ‘forced’ to judge the value of each element, and were unable to exempt themselves from judgment by choosing the middle ground. We believe that this measure increases the likelihood of valuable elements being included, and decreases the likelihood of unjustly excluding elements.
study suggest that the contents of the list should be able to be developed over the course of the study (Martino, 1993). Therefore, after asking the question to rank a certain item, we asked the panelists for each concept or practice if they would suggest reformulating it. We asked them to only do so if they believed that the reformulated version of the item would be of greater importance than the initial version of that same item. Following the example of Minkman et al. (2008), we also provided panelists with the opportunity to add new items to the initial list, if they felt that we had left out crucial concepts/practices. The results were documented by providing all panelists with a digital survey, using the survey software Qualtrics, containing our initial list of concepts and practices. The survey consisted of a sequential listing of our elements. Each element was accompanied by fields in which panelists could provide their Likert score (1-4), as well as fields that provided space for panelists to reformulate the initial elements. The elements were displayed one at a time, and panelists were ‘forced’ to provide a Likert score before being able to move on to the next element. Evidently, the panelists were not forced to provide a reformulation of the initial element, but had the opportunity to do so. At the end of the survey, the panelists were granted the
opportunity to suggest new items to introduce into the study.
Round two
Before entering round two, our list of concepts and practices needed to be adjusted in accordance with the consensus, so that the panelists could provide feedback and/or further insights in a second round (Franklin & Hart, 2007). Currently, no clear guidelines exist regarding how an expert consensus is reached in a Delphi study (Boulkedid et al., 2011). We therefore consulted an earlier research using the Delphi technique, which was conducted in a similar setting, and with a similar goal (Minkman et al., 2008). Following their example, an element was included if, in the previous round, it was ranked by 80% or more of our panelists as ‘Important’ or ‘Very important’ (i.e., a score of 1 or 2). An element was automatically excluded if 50% of more of our panelists ranked it as ‘Not important’ or ‘Of little importance’ (i.e., a score of 3 or 4).
certain of keeping an element which have a high agreement on importance (>80%). Secondly, to be cautious about eliminating an element (>50%) so as not to miss a topic, and thirdly to make sufficient use of the option of reformulation.” (Minkman et al., 2008, p. 68).
For each element that scored anywhere in between inclusion and exclusion, an
individual analysis was conducted in order to assess the need for reformulation. Additionally, we conducted an analysis of the newly suggested elements from our panelists, to decide if new items should be included in round two’s list. Following the example of Minkman et al. (2008), the author of this paper conducted these analyses. Based on these analyses, elements were either reformulated, initial elements remained unchanged, or new elements were added to the list, if deemed necessary by the author. If certain reformulations strayed to far from the original items, and the formulation of a new item therefore seemed more appropriate, a new item was introduced.
The second round consisted of panelists scoring and reformulating our second list of items in the same way as in round one (i.e. rank items in order of importance on the same 1-4 Likert scale, and provide a reformulation if necessary). The second list of items consisted of the items that scored in between inclusion and exclusion, and new elements that were suggested by panelists during the first round. Based on the individual analyses, these items could either be reformulated, or remain unchanged. The panelists were also, again provided with the opportunity to make suggestions for new elements, as was the case in the first round of our study.
Analysis of results
4. RESULTS
The literature review that was conducted resulted in an initial list of 40 elements. As stated before, this list with initial elements can be found in appendix A. For the purpose of being able to identify underlying concepts later on in the analysis, these elements were grouped based on which ‘sections’ of work they are based upon. These sections coincide with the six different sections of our literature review: “the basic concept of value based health care” (VBHC), “quality of care” (QoC), “measuring value” (MV), “patient
centeredness” (PC), “paying for healthcare” (PFH), and “performance measurement in operations management” (PMiOM). Evidently, this categorization of elements was not made clear to our participants when partaking in the study.
The full list of initial elements was presented to our panelists during the first round, while the panelists were presented an adjusted list for the second round of the study. As stated before, we assembled a panel consisting of 10 experts, based on their experience and expertise in either practice in value based healthcare, or in academics or consultancy. For both the first and the second round, 9 out of 10 of the panelists responded, resulting in a 90% response rate. However, one of the responses of the second round appeared to be unusable, since it showed that a response was filled out for all questions, but no actual scores or results were visible. Since the software that was used to carry out the study (Qualtrics) did not allow a participant to participate twice, this resulted in 8 useful responses. Nevertheless, both of the rounds showed a response rate that exceeded the desired threshold of 7 participants (Linstone, 1978). The outcomes of the first and second round were analyzed, which ultimately resulted in a final list of “included” elements, which was used for further analysis and discussion. A summary of these results is displayed in Table 1 and Table 2, which can be found below
Table 1: summary of Delphi study results per category
VBHC QoC MV PC PFH PMiOM Total
Initial number
of elements: 8 9 7 5 4 6 39
Included after
*Percentages are rounded up/down to the nearest whole percentage
Table 2: Breakdown of Delphi study results as a whole for each round
*Standard deviation, rounded up to the nearest double decimal
Table 1 shows the amount total of elements for each category and the amount of elements included after each round, as well as the total outcome in terms of inclusion as well as exclusion. It also displays the amount of elements that was reformulated based on the suggestions from our panelists. Table 2 shows a more detailed breakdown of the specific results for each of the two rounds of the Delphi study. In some cases, panelists provided suggestions for reformulation for an element that was already included based on the scores it received in the first round. These suggestions were all rejected, since the score for the element in round one indicated that the initial element was relevant enough to be included in our final list. This is partially due to the fact that we aim for saturation, which means that we
eventually wish to end up with a final list that does not require the reformulation of existing elements, or the addition of new elements.
It also occurred that items scored in between inclusion and exclusion, but remained unchanged in assembling the list for the second round. This was the case if no suggestions for reformulation were made, or if the author deemed the suggestion for reformulation
unnecessary or irrelevant. Coincidentally, all “unchanged” elements were elements where no suggestion for reformulation was given. The panelists were also given the option to
reformulate elements during the second round. Suggestions for reformulation of elements that were included based on their score in the second round were all rejected, similar to the
method applied after the first round. Consequently, no elements were reformulated after the second round, since there were no (valid) suggestions for reformulation for the remaining elements.
New elements: 3 0
Suggestions for reformulation:
62 4
Lastly, the panelists were also provided with the option to suggest new elements to add to the list. After the first round, we found 5 suggestions for new elements. Two of these
suggestions were rejected, resulting in 3 new elements being introduced to the study during round two. One of these rejected suggestions coincided too much with an element that was already on the initial list. The other suggestion that was rejected involved an upcoming change in Dutch legislature regarding patient privacy. Since this was deemed too specifically tailored to the Dutch healthcare climate, it was excluded from this study, since we are
attempting the make the ultimate outcome as universally applicable as possible. The other suggestions were taken into consideration and, after some slight rephrasing, were entered into the second round of the study. Although participants also had the opportunity to provide suggestions for new elements during the second round, none were given.
To provide some perspective, Table 1 also displays the amount of elements on the final list as a percentage of the total amount of elements entered into the study. This percentage is based on the amount of elements included on the final list compared to the cumulative total of all initial elements and the new elements entered into the study (i.e., for the total amount of elements included on the final list, the 67% is obtained by dividing the 29 elements included by 43, since there were 40 initial elements, and 3 newly introduced elements). The full final list of elements that resulted from our study can be found in Appendix B. Elements that were reformulated are ‘marked’, and can be recognized by “(R)” being written behind them. Newly introduced items that weren’t on are initial list are also marked, having written “(NEW)” behind them.
5. DISCUSSION
In this section, we will first provide an analysis and discussion of the overall results in terms of the categories identified in our literature review. Afterwards, we will go into a more in-depth analysis and analyze the elements on an individual level. Consequently, we will discuss the implications of our findings, and conclude with a section that addresses and discusses the limitations of this study, along with suggestions for future research.
When taking a glance at the overall results, one of the first things that comes to our attention is the fact that most of the elements extracted from the category regarding the basic underlying foundation of value based health care (VBHC), which are predominantly based on Porter’s (2006; 2008; 2009; 2010; 2014; 2016) ideas, are deemed important by experts. Since 7 out of a total of 8 elements were included after two rounds, we believe that there is a clear indication that value based healthcare as a whole is a very promising concept, and that it is therefore valid to grant it more attention, in research as well as in practice.
A second aspect that stands out is the somewhat underwhelming ‘performance’ of elements related to the “performance management in operations management” (PMiOM) and the “paying for healthcare” (PFH) categories. This indicates that our experts suggest to, initially, mostly steer away from focusing on payment structures and the cost of healthcare. This coincides with earlier findings from Hubley & Miller (2016), who suggest that the restructuring of how we pay for healthcare follows healthcare delivery reforms. It also suggests that, although there may be some useful aspects of PMiOM that may be applied in the implementation of value based healthcare, this is a less relevant area than, for example, quality of care. This can be partially attributed to the fact that performance measures in operations management are related to an organization’s strategy, meaning that the
performance measures that are used should be in line with an organization’s strategy (Mills, Wilcox, Neely & Platts, 2000). Since strategy is less prevalent healthcare organizations, this fact may have contributed to the rejection of several PMiOM elements in our study.
Furthermore, it stands out that all elements related to “patient centeredness” (PC) were eventually included in our final list of elements. We believe we can not only learn from this that, according to our results, this seems to be (one of) the most crucial aspect(s) of value based healthcare, but also find that there are still insights to be gained in this area, since one of the elements needed to be reformulated (Appendix B, element 29), after which it was included unanimously during the second round.
Upon further inspection of our final list of elements, one of the first things is becomes apparent is that only one of the elements was unanimously rated with a score of 1 (Appendix B, element 26: “Involving patients in the shared decision making process (regarding
treatment options etc.) as much as possible”. The element is therefore marked with “(+)”),
indicating that it is “Very important”. The fact that this element was extracted from PC coincides with our earlier finding and suggestion that PC, according to our results, appears to be a crucial factor in understanding and moving towards value based healthcare.
Accordingly, we also found that several included elements regarding other categories could also be partially categorized under PC. An example of such an element can be found in the “Quality of Care” (QoC) category (Appendix B, element 12), which regards involving the patient’s perspective in assessing the quality of delivered care. We believe that this further adds to the ‘claim’ that PC appears to be a key concept in value based healthcare.
Subsequently, we found that it was agreed upon that what we considered to be some of the core elements of the VBHC section, were all included in our final list. These elements (Appendix B, elements 1 and 4) regard some of Porter’s initial ideas, concerning the fact that health care systems should strive to provide universal healthcare insurance coverage (Porter, 2009), as well as the proposition that care delivery should be organized around complete care cycles rather than individual treatments (Porter, 2009). This does not only coincide with Porter’s propositions, but also with earlier findings that suggest that using integrated care pathways is likely to improve quality of care (Campbell, Hotchkiss, Bradshaw & Porteous, 1998, Vanhaecht et al., 2016). These two are related in a sense that one could argue that a higher quality of care should lead to better outcomes, which would in turn improve the ratio of health outcomes per dollar spent.
basis of value based healthcare and its initial core ideas are valid, and are supported by experts in the field of healthcare.
5.3 Theoretical and practical implications
Firstly, as mentioned before, we believe that our results show that PC appears to be a vital component of (moving towards) value based healthcare. Porter’s (2009) propositions about designing care ‘around’ patients by organizing care delivery in full care cycles, rather than managing individual treatments, which is mostly convenient for the care provider, are supported by expert opinion. Additionally, our panel also supported the use of patient reported measures. Besides indicating that this may be a valuable area in the transition towards a value based healthcare system, this also ties in to earlier findings that suggest that the use of patient centered practices can lead to better health outcomes (Robinson, Callister, Berry & Dearing, 2008). Some earlier research even goes so far as to say that “a good outcome must be defined in terms of what is meaningful and valuable to the individual patient” (Epstein & Street, 2011, p. 100).
However, one aspects of PC, namely the use of PREMs, was only supported and included after reformulation. This element (Appendix B, element 29) was only included after its reformulated version addressed the fact that PREMs should only be used to evaluate performance in terms of the care delivered to the particular patient that provided the reported experience measure. Since this is in contrast with our initial finding in existing literature, which proposes to use PREMs for performance evaluation in general, we believe this is an area that deserves additional research.
results indicate that this area may prove to be a valuable addition to the existing literature, and suggest that healthcare providers that aim for value based healthcare delivery start practicing shared decision maker to a higher extent.
Furthermore, we believe that our final list of elements can provide a broad basis of future research. For example, we suggest that a multiple case study could be conducted between several healthcare institutions, where one (or more) of the elements on our final list is more prevalent in one (or more) cases that in the other case(s). Since Porter’s (2008) value concept, the health outcomes per dollar spent, can be measured, such a study can possibly provide theoretical evidence that the use of a certain element is indeed useful in providing ‘valuable’ care. If certain elements are identified as being strongly related to delivering high value in healthcare, this provides an even broader basis for the use of these elements in moving towards a value based system, or even for their use as a general means to improve quality of care.
In terms of practical implications in its most simple form, healthcare organizations could simply seek to apply one or more concepts and/or practices from our final list
(Appendix B) in practice. Since a panel of experts with great experience in healthcare, who were enthusiastic about the concept of value based health care and showed support for these concepts and practices, we believe that they provide a good starting point for healthcare providers who are seeking to move towards value based healthcare delivery. However, since we ultimately aspire to move towards a value based system, we believe the best approach would be to start with the ‘bigger picture’. Since healthcare is often organized at a national level, we suggest that larger institutions (government, for example) should lead the way in moving towards a value based healthcare system. This suggestion is, in a way, backed up by existing theory in healthcare academics, namely in literature related to the Chronic Care Model (CCM). The CCM is a framework that is laid out such that it identifies several elements in six different areas at the ‘system-level’, that should ultimately lead to improved health outcomes (Wagner, 1998). Examples of these elements include the design and organization of healthcare at large, which includes delivery system design as well as
that they should be in place to improve outcomes in chronic disease care (Wagner, 1998). The use of the CCM has already been linked to actually achieving better health outcomes and a higher quality of care (Nutting, Dickinson, Dickinson, Nelson, King, Crabtree & Glasgow, 2007). Although the CCM is tailored towards chronic care, we believe that its overall notion that designing a healthcare system ‘top-down’ (i.e. to start at a larger scale) could carry over into the area of value based healthcare.
In terms of our lists of concepts and practices, this boils down to two major
recommendations, the first of which relates to insurance coverage. We suggest governments should strive to provide health insurance coverage to its entire population, and to make health insurance as affordable as possible. This relates to two of the elements (Appendix B, elements 1 and 2) of this study. We believe that this should be the first step in moving towards a value based system because ultimately, high quality care with a high degree of patient centeredness will be less effective in optimizing value for patients if not everyone in the population is granted access to this care. The second suggestion we make is that the amount of performance measures used should be reduced as much as possible (Appendix B, element 20). Although our study supports the use of performance measures in general, we believe that it is important to understand and acknowledge that they should be used in appropriate amounts. As
mentioned before, many hospitals currently have to answer to a very wide range of performance measures (Meyer et al, 2012), which may cause the use of this measures to exceed its initial aim, since the use of more measures may cause confusion rather than accurate measurement (Meyer et al, 2012). Because accurately measuring performance is a key step in the improvement of health outcomes (Porter, 2010), we feel that working towards the use of accurate, appropriate performance measures is a good first step in moving towards a value based healthcare system. Since the choice of which performance measures are used to assess the performance of healthcare providers is usually organized on a larger scale, we believe that it is critical to address this issue before moving on to making changes at a smaller scale. This is in line with our previous proposition that the change towards a value based healthcare system should be organized starting at a large scale.
This study does have its limitations. The first limitation that comes to mind is that our study was carried out among Dutch experts only. Since the study is predominantly based on these experts and their opinion(s), the results may be tailored somewhat more to the elements in relation to the current Dutch healthcare climate, and may therefore not be as universally applicable as one would like. A suggestion for future research could therefore be to carry out a similar or related study in a more multicultural environment, in an attempt to gather data and information that may prove to be of higher generalizability.
Another limitation lies in the specific setup of our Delphi study. Since we did not conduct a ‘true’ Delphi study, where the panel starts with a blank slate, and initial elements are only developed after the first round, we cannot be sure that we have not missed any important elements regarding the move towards a value based healthcare system. We argued that a literature review is a sufficient way to compose an initial list of elements, which can in turn be used to start a Delphi study. This method may also prevent elements being overlooked by the experts, and therefore being unjustly excluded from our study. However, it may also occur that an element that did not categorize under one of the six categories identified in the literature review ‘slipped through the cracks’. Another suggestion for future research could therefore be to carry out a ‘true’ Delphi study on value based healthcare, without using a list of pre-constructed elements, letting the elements emerge and evolve more naturally. Although we believe we addressed the issue of possible exclusion of important elements by granting participants the option to suggest new elements, a ‘true’ Delphi study may motivate
participants more to introduce elements themselves. Coincidentally, one of the panelists wrote in the ‘suggestions for new elements’ section that we should consider executing such a Delphi study. The risk and downside of such an approach would be that results are based solely on experts opinion, rather than having at least some theoretical basis for the initial elements entered into the study. Perhaps the results of such a study and this study could be compared and/or combined.
6. CONCLUSION
value based healthcare system. A literature review was carried out in order to identify concepts and practices that exist in what little literature currently relates to value based healthcare, and its core principles as proposed by Porter (2006; 2008; 2009). During the literature review, six different categories were distinguished, and concepts and practices were identified accordingly.
A list of elements consisting of concepts and practices that were extracted from our theoretical background was constructed consequently. By subjecting this list of elements to a Delphi study, we were able to distinguish what concepts and practices are most important in the transition to a value based healthcare system, by using expert opinion. Results indicate that Porter’s basis for the concept, along with a high degree of patient centeredness during care delivery are necessary in ensuring a value based healthcare system can successfully be implemented. Our resulting list of elements that was obtained after two rounds of the Delphi study could be used to identify elements to put into practice, or in order to identify areas that deserve more attention in terms of research. More specifically, we suggest that the move towards a value based system should start at a large/national level. We advise that
REFERENCES
American Medical Association (1986), Council of Medical Service. Quality of care. JAMA;
Journal of the American Medical Association, 256, 1032-1034.
Black, N. (2013). Patient reported outcome measures could help transform healthcare. BMJ:
British Medical Journal (Online), 346:f167.
Boulkedid, R., Abdoul, H., Loustau, M., Sibony, O., & Alberti, C. (2011). Using and reporting the Delphi method for selecting healthcare quality indicators: a systematic review. PloS one, 6(6), e20476.
Bourne, M., Mills, J., Wilcox, M., Neely, A., & Platts, K. (2000). Designing, implementing and updating performance measurement systems. International Journal of Operations &
Production Management, 20(7), 754-771.
Bradley, E. H., Herrin, J., Elbel, B., McNamara, R. L., Magid, D. J., Nallamothu, B. K., ... & Krumholz, H. M. (2006). Hospital quality for acute myocardial infarction: correlation among process measures and relationship with short-term mortality. JAMA; Journal of the American
Medical Association, 296(1), 72-78.
Campbell, H., Hotchkiss, R., Bradshaw, N., & Porteous, M. (1998). Integrated care pathways.
BMJ: British Medical Journal, 316(7125), 133-137.
Dawson, J., Doll, H., Fitzpatrick, R., Jenkinson, C., & Carr, A. J. (2010). The routine use of patient reported outcome measures in healthcare settings. BMJ; British Medical Journal, 340, c186.
Donabedian A. (1980). Explorations in quality assessment and monitoring. Vol. 1. The definition of quality and approaches to its assessment. Ann Arbor, Mich.: Health
Administration Press, 1980.
Dowding, D., Randell, R., Gardner, P., Fitzpatrick, G., Dykes, P., Favela, J., Hamer, S., Whitewood-Mores, Z., Hardiker, N., Borycki, E., & Currie, L. (2015). Dashboards for improving patient care: Review of the literature. International Journal of Medical
Informatics, 84 (2), 87-100.
Epstein, R. M., & Street, R. L. (2011). The values and value of patient-centered care. Annals
of Family Medicine, 9(2), 100-103.
Feeley, T.W., Fly, H.S., Albright, H., Walters, R., & Burke, T.W. (2010). A method for
defining value in healthcare using cancer care as a model. Journal of Healthcare Management
/ American College of Healthcare Executives, 55(6), 399-411.
Franklin, K. K., & Hart, J. K. (2007). Idea generation and exploration: Benefits and limitations of the policy Delphi research method. Innovative Higher Education, 31(4), 237-246.
Giordano, L. A., Elliott, M. N., Goldstein, E., Lehrman, W. G., & Spencer, P. A. (2010). Development, implementation, and public reporting of the HCAHPS survey. Medical Care
Research and Review, 67(1), 27-37.
Hasson, F., Keeney, S., & McKenna, H. (2000). Research guidelines for the Delphi survey technique. Journal of Advanced Nursing, 32(4), 1008-1015
Hubley, S., & Miller, B. (2016). Implications of healthcare payment reform for clinical psychologists in medical settings. Journal of Clinical Psychology in Medical Settings, 23(1), 3-10
Hudson, M., Smart, A., & Bourne, M. (2001). Theory and practice in SME performance measurement systems. International Journal of Operations & Production Management, 21(8), 1096-1115.
Kaplan, R. S., & Norton, D. P. (2005). The balanced scorecard: measures that drive
performance. Harvard Business School Publishing.
Kennerley, M., & Neely, A. (2002). A framework of the factors affecting the evolution of performance measurement systems. International Journal of Operations & Production
Keswani, A., Koenig, K. M., & Bozic, K. J. (2016). Value-based Healthcare: Part 1—
Designing and implementing integrated practice units for the management of musculoskeletal disease. Clinical Orthopaedics and Related Research, 474(10), 2100-2103.
Kitson, A., Marshall, A., Bassett, K., & Zeitz, K. (2013). What are the core elements of patient-centred care? A narrative review and synthesis of the literature from health policy, medicine and nursing. Journal of Advanced Nursing, 69(1), 4-15.
Kroch, E., Vaughn, T., Koepke, M., Roman, S., Foster, D., Sinha, S., & Levey, S. (2006). Hospital boards and quality dashboards. Journal of Patient Safety, 2(1), 10-19.
Krumholz, H. M., Normand, S. T., Spertus, J. A., Shahian, D. M., & Bradley, E. H. (2007). Measuring performance for treating heart attacks and heart failure: the case for outcomes measurement. Health Affairs, 26 (1), 75-85.
Kuwaiti, M. E. (2004). Performance measurement process: definition and ownership.
International Journal of Operations & Production Management, 24(1), 55-78.
Linstone, H.A. (1978). The Delphi technique. Handbook of Futures
Research. Westport, CT: Greenwood, 271–300.
Lohr, K. N., ed. (1990). Medicare: a strategy for quality assurance. Washington, D.C.:
National Academy Press, 1990.
Marshall, S., Haywood, K., & Fitzpatrick, R. (2006). Impact of patient-reported outcome measures on routine practice: A structured review. Journal of Evaluation in Clinical Practice, 12(5), 559-568.
Martino, J. P. (1993). Technological forecasting for decision making. McGraw-Hill, Inc..
Meyer, G. S., Nelson, E. C., Pryor, D. B., James, B., Swensen, S. J., Kaplan, G. S., Weissberg, J. I., Bisognano, M., Yates, G. R., & Hunt, G. C. (2012). More quality measures versus
measuring what matters: a call for balance and parsimony. BMJ Quality and Safety, bmjqs-2012.
Minkman, M., Ahaus, K., Fabbricotti, I., Nabitz, U., & Huijsman, R. (2008). A quality management model for integrated care: results of a Delphi and Concept Mapping study.
International Journal for Quality in Health Care, 21(1), 66-75.
Moxham, C. (2009). Performance measurement: Examining the applicability of the existing body of knowledge to nonprofit organisations. International Journal of Operations &
Production Management, 29(7), pp.740-763
Porter, M. E., & Teisberg, E. O. (2006). Redefining health care: creating value-based
competition on results. Harvard Business Press.
Porter, M. E. (2008). Value-based health care delivery. Annals of Surgery, 248(4), 503-509.
Porter, M. E. (2009). A strategy for health care reform—toward a value-based system. New
England Journal of Medicine, 361(2), 109-112.
Porter, M. E. (2010). What is value in health care? New England Journal of Medicine, 363(26), 2477-2481.
Porter, M. E., & Kaplan, R. S. (2014). How should we pay for health care?. Harvard Business
School, working paper.
Porter, M. E., Larsson, S., & Lee, T. H. (2016). Standardizing patient outcomes measurement.
New England Journal of Medicine, 374(6), 504-506.
Robinson, J. H., Callister, L. C., Berry, J. A., & Dearing, K. A. (2008). Patient-centered care and adherence: Definitions and applications to improve outcomes. Journal of the American
Association of Nurse Practitioners, 20(12), 600-607.
Stiggelbout, A. M., Van der Weijden, T., De Wit, M. P., Frosch, D., Légaré, F., Montori, V. M., ... & Elwyn, G. (2012). Shared decision making: really putting patients at the centre of
healthcare. BMJ; British Medical Journal, 344, e256.
Teale, E.A., & Young, J.B. (2015). A patient reported experience measure (PREM) for use by older people in community services. Age and ageing, 44(4), 667-672
Thangaratinam, S., & Redman, C. W. (2005). The delphi technique. The Obstetrician &
Gynaecologist, 7(2), 120-125.
Vanhaecht, K., Lodewijckx, C., Sermeus, W., Decramer, M., Deneckere, S., Leigheb, F., ... & Panella, M. (2016). Impact of a care pathway for COPD on adherence to guidelines and hospital readmission: a cluster randomized trial. International Journal of Chronic Obstructive
Pulmonary Disease, 11, 2897.
Wagner, E.H. (1998). Chronic disease management: What will it take to improve care for chronic illness? Effective Clinical Practice : Ecp, 1(1), 2-4.
Wouters, M., & Sportel, M. (2005). The role of existing measures in developing and implementing performance measurement systems. International Journal of Operations &
APPENDIX A: INITIAL LIST OF ITEMS ENTERED INTO THE
DELPHI STUDY
Please note that the purpose of the terms between parentheses is to indicate from which section of the theoretical background the corresponding element was extracted. The list we distributed to our panelists does not include these terms. Also, since the study was carried out among a panel of Dutch experts, all elements were translated into Dutch. For full
transparency, the Dutch translation that was used while carrying out the research is listed in parentheses following “NL:”. The items remained numbered during the study, for the purpose of convenience for both the author and the panelists.
1. Providing (or aiming to provide) universal healthcare insurance coverage (VBHC) (NL: Het bieden van een zorgklimaat waarin het streven is om de gehele bevolking te voorzien van een zorgverzekering)
2. Focusing on making individual health insurance as affordable as possible (VBHC) (NL: Streven om zorgverzekeringen voor individuen zo betaalbaar mogelijk te maken)
3. Reducing costs of operations and materials as a driver for performance improvement (VBHC) (NL: Het minimaliseren van operationele kosten gebruiken als drijver van
prestatieverbetering voor zorgverleners)
4. Organizing delivery of care around full treatment cycles of medical conditions, rather than around individual procedures (VBHC) (NL: Zorgverlening zo organiseren dat het gehele behandeltraject centraal staat, in plaats van de individuele behandelingen (bijvoorbeeld de DBC-methode))
5. Developing a technological platform that enables data sharing related to improving patient care (VBHC) (NL: Het ontwikkelen van een technologisch/digitaal platform dat kan worden gebruikt om data in te zien en te delen met anderen, met als doeleinde het verbeteren van de verleende zorg)
6. Facilitating by a data or business intelligence manager or team that retrospectively collects and analyzes existing data from patient records. (VBHC) (NL: Het aanstellen van een data of business intelligence manager (of team van personen) die zich richt op het verzamelen en analyseren van bestaande data uit patiëntendossiers)
7. Developing road maps that guide participating healthcare providers through the process of moving towards value based healthcare delivery (VBHC) (NL: Het ontwikkelen van een gestandaardiseerd stappenplan (“road map”) die zorgaanbieders kunnen gebruiken om te transformeren in een "value based" zorgaanbieder)
8. Facilitating by a change manager or team to implement value based health care (VBHC) (NL: Het aanstellen van een "change manager" (een expert op het gebied van value based zorgverlening) die zorgaanbieders helpt om te transformeren naar een "value based" zorgaanbieder”)
9. Facilitating by including a patient representative in the improvement team to ensure patient input (PC) (NL: Het aanstellen van een patiënt als ervaringsdeskundige, met als doeleinde het het perspectief van de patiënt in het verbeterproces zeker te stellen)
11. Using a patient’s mental well-being in assessing the outcome of healthcare delivery (QoC) (NL: Het mentale welzijn van de patiënt gebruiken als uitkomstindicator bij het evalueren van de zorgverlening)
12. Focusing on attempting to deliver a desired and sustainable outcome from a patient’s perspective (QoC) (NL: Het leveren van een gewenste, duurzame uitkomst van zorg, gezien vanuit het perspectief van de patiënt)
13. Ensuring the general safety of patients when undergoing treatment (QoC) (NL: De veiligheid van de patiënt voorop stellen tijdens het ondergaan van een behandeling)
14. Avoiding over- and under use of healthcare services (QoC) (NL: Het overmatig of ondermatig gebruik van zorgverlening door patiënten voorkomen)
15. Minimizing waiting or throughput times for medical procedures (QoC) (NL: Het minimaliseren van wachttijden en behandeltijden)
16. Reducing waste, meaning e.g. the waste of time and/or staff capacity (QoC) (NL: Het reduceren van verspilling ("waste"), zoals bijvoorbeeld de verspilling van tijd, materialen, of capaciteit)
17. Creating integrated practice units (IPUs) (QoC) (NL: Het gebruik van zogeheten integrated practice units (IPUs). Een IPU is een team of afdeling binnen een zorginstelling die als het ware om een patiënt heen is georganiseerd, en de verantwoordelijkheid voor alle behandelingen binnen een zorgtraject op zich neemt, in plaats van de patiënt van de ene naar de andere behandeling te sturen.)
18. Revising and reformulating treatment protocols iteratively in improving patient care (Qoc) (NL: Het iteratief updaten en herformuleren van protocollen en voorschriften, om zo de kwaliteit van de zorg te verbeteren)
19. Using process measures to affect health outcomes (MV) (NL: Het gebruik van procesindicatoren als determinanten van gezondheidsuitkomsten)
20. Reducing the amount of performance measures used (MV) (NL: Het reduceren van het aantal prestatie-indicatoren dat wordt gebruikt in het evalueren van de verleende zorg)
21. Standardizing performance measures for treatment cycles of medical conditions, rather than for individual treatments/procedures (MV) (NL: Prestatie-indicatoren standaardiseren voor volledige behandeltrajecten, in plaats van indicatoren voor afzonderlijke behandelingen te gebruiken)
22. Assessing the success of a treatment cycle by measuring the achieved health status (MV) (NL: De kwaliteit van de geleverde zorg evalueren op basis van de
