Insights in how surgical teams can improve value by learning from data Deriving value from measuring outcome and cost data within healthcare:

(1)

Deriving value from measuring outcome and cost data

within healthcare:

Insights in how surgical teams can improve value by learning

from data

July 9, 2018 By: Wout Essink Student number: s2377314 E-mail: w.essink@student.rug.nl

University of Groningen

Faculty of Economics and Business

MSc Supply Chain Management and MSc Technology and Operations

Management

(2)

1

Preface

(3)

2

Abstract

In 2006, value-based healthcare (VBHC) was introduced as an approach in order to maximize value for the patient. Value is maximized, by accomplishing the best outcomes for the patient at the lowest possible costs. One of the principles of VBHC is that outcomes and costs should be measured in order to improve value. This study aims to provide a better understanding of how outcome and cost data can be measured, analyzed, represented and used in improvement activities to increase value for surgical teams. Furthermore, this research also explores how surgical teams actually learn from outcome and cost data. This is done by analyzing outcome and cost data of five different surgical teams who participated in a 1-year VBHC project and by performing eleven interviews with surgeons and managers from the same five surgical teams. This study demonstrates a method for measuring outcome as well as cost data and shows how this data can be represented and used. Additionally, the results of this study show that although outcome and cost data for two diseases were measured and compared with other surgical teams for a period of one-year, surgical teams did not show to have learned much from the benchmarking data, which is very surprising. Besides, the results show that surgical teams are not following prescribed steps for successful benchmarking, which is also surprising. Due to the unexpected results, additional research is done and eight contextual conditions are identified which should be in place in order to learn from outcome and cost data. It was found that a lack of presence of these contextual conditions results in limited learning outcomes

Keywords: Measuring VBHC, costs VBHC, Learning VBHC, data VBHC, Learning conditions

(4)

3

1. Introduction

In the beginning of the 21th century, healthcare systems performed poorly. Costs were increasing year after year while the quality was unsatisfactory (Miller, 2009; Porter & Teisberg, 2004). As an answer to the failing healthcare system, Porter and Teisberg wrote a book in 2006 called “Redefining Healthcare: Creating Value-based competition on results” and introduced the concept of value-based healthcare (VBHC). Value can be defined as: “health outcomes achieved per dollar spent”, and achieving the highest possible value should the overarching goal of healthcare delivery system (Porter, 2010, p.2477). Moreover, they stressed the importance of measuring costs and results, because without measuring it is hard to tell what improves value or not. Since the introduction of value-based healthcare, the concept is further explored and in some countries, we already see examples of a shift from volume to more value-based healthcare systems (Porter & Lee, 2013; Mkanta et al., 2016). However, although slightly improvements are made since the introduction of the concept of VBHC, progress has been slow and there is still a growing pressure for hospitals to cut down costs and improve quality (Radnor et al., 2012; Porter & Lee, 2013; Ryan et al., 2017). The stagnation of progress can be partly explained by the challenges and difficulties encountered regarding the measurement of outcome and costs in order to improve value. Even with the most advanced systems, organizations lack the critical capability to create and extract cost and outcome data (Porter & Lee, 2013).

(7)

6 of value-based healthcare depends heavily on the achievements made in measuring value (Mkanta et al., 2016).

So, although VBHC seems to be a very promising concept in order to improve quality and reduce costs, it still seems to be difficult to realize the full potential of the concept. This is not strange, since there is still a lot of ambiguity regarding the meaning, definition and measurement of value (Mkanta et al., 2016). Furthermore, there is a lack of rigorous scientific research which proves that VBHC indeed improves quality and decreases costs (van Deen et al., 2017). Additionally, up to now, far too little attention has been paid to the question how providers could actually learn from outcome and cost data in order to increase value. The “learning” part is almost completely neglected by Porter, while this is of great importance, since plainly collecting, analyzing, displaying and disseminating data is not sufficient in order to realize change within healthcare (Mannion & Davies, 2002). Therefore, the most important goal of this research is to explore and shed new light on how surgical teams can actually learn from outcome and cost data. Additionally, this study aims to provide more insights in how surgical teams can measure, analyze, represent and use outcome and cost data in improvement activities to increase value for surgical teams. Lastly, this study aims to find out how surgical teams can use outcome and cost data for benchmarking purposes in order to increase value. In order to address the previously described gaps, the following research questions should be answered:

1. How can outcome and cost data be measured, analyzed, represented and used in

improvement activities to increase value for surgical teams?

2. How can surgical teams actually learn from outcome and cost data?

(8)

7 The main contribution of this research is to provide new insights into how surgical teams can actually learn from outcome and cost data. This is highly relevant, since surgical teams are increasingly gathering data due to developments in IT and therefore need to know how to learn from that data in order to improve value. Another contribution of this research is that it sheds new light on the third principle of VBHC, namely the measurement of value. This is done by demonstrating the value and possibilities of Textbook Outcome in practice and by proposing a method for measuring costs which is based on standard cost prices and widely used by surgical teams for benchmarking purposes within The Netherlands. Additionally, this research shows whether surgical teams are currently following prescribed guidelines of successful benchmarking from literature. Lastly, this research shows which contextual conditions should be in place in order to learn from outcome and cost data.

(9)

8

2. Theoretical background

First, section 2.1 will discuss the current state of the concept VBHC and highlights areas where up to now, far too little attention has been paid to. Next, section 2.2 will discuss one of these areas, namely measuring value in more detail including the shortcomings of current the literature. Lastly, section 2.3 discusses relevant benchmarking literature. Benchmarking can be seen as a learning program, which is based on performance comparison (Camp, 1989). Therefore, benchmarking theory will be used as a framework to explore how surgical teams can actually learn from outcome, process and cost data.

2.1 The concept of value-based healthcare

According to Porter and Teisberg (2006), in order to solve the cost crisis, a shift in focus is needed from volume to a more value-based healthcare system, where the central focus is on increasing the value for patients and not on volume and profitability (Porter, 2010). The concept of VBHC, which is based on three principles, should guide this shift. In this section, the concept of VBHC and the three principles of VBHC will be further explained.

(10)

9 However, although there is already a shift from volume towards more value-based approaches, a search of literature about VBHC revealed that much of the research up to now has been descriptive in nature. Examples are the papers of Kaplan & Porter (2011), Porter & Lee (2013) and Mkanta et al. (2016). This might be partly explained by the fact that the concept is relatively new and changing a healthcare strategy takes a lot of time. Still, definitional issues, measurements and the relationship between cost and outcome should be more clarified, since it is not yet understood by all stakeholders (Mkanta et al., 2016). This can be partly explained by the fact that the concept itself provides no detailed explanation about how to measure outcomes and costs. Furthermore, the choice for only using outcome indicators is arguable, since the determinants of outcomes are also relevant (Krumholz et al., 2007). Additionally, no attention has been paid to how organizations can actually learn from outcome and cost data, while simply analyzing and displaying data will not result in change within healthcare (Mannion & Davies, 2002). So, although the concept of VBHC seems promising, more empirical research is needed in order to clarify the concept and to prove its promises.

The focus of this research will be on the third principle of VBHC (measuring value) and on learning from VBHC, which are until now relatively unexplored fields of research. The next section will discuss the literature concerning the measurement of value in more detail.

2.2 Measuring value

2.2.1 Measuring outcome

(11)

10 sustainability of recovery. Furthermore, the measures should include short-term as well as long-term outcomes and outcome measurements should also include sufficient measurement for risk factors in order to allow risk adjustment (Porter, 2010). Taken together, Porter et al. (2010) argue for a minimum but sufficient set of outcome measures for major medical conditions, which should be adjusted for risk. Moreover, these measures should be standardized nationally and globally.

(12)

11 consuming to collect data concerning health outcomes. Additionaly, this approach makes it possible to get a simple comprehensive summary of clinical care, give valuable clinical insights into the daily practice per patient group and is especially appropriate for clinical interventions such as surgery and invasive diagnostics (Kolfschoten, et al., 2013; Salet et al., 2018). This research will demonstrate the learning and benchmarking possibilities which are possible with this method. Lastly, this research proposes a way of representing outcome, process and cost data.

Another component which should be measured in order to determine value is costs. Therefore, the next section will discuss the current literature regarding the measurement of costs within healthcare.

2.2.2 Measuring costs

An important aspect of VBHC is measuring costs. However, different researchers (Porter & Kaplan, 2016; Teckie et al., 2014; Kaplan et al., 2014; Keel et al., 2017) acknowledge that it is still difficult for healthcare organizations to accurately measure costs. According to Teckie et al. (2014), the inability to assess costs is the main challenge in determining value within healthcare. Currently, most of the providers have no clarity on what the costs of various components are. Kaplan et al. (2014) further argue that currently used cost systems impede cost reductions and improvement, since they rely on inaccurate cost allocations. Besides the accuracy problems, hospitals measure costs mostly by department, while they should measure costs over the full cycle of care (Porter, 2010).

(13)

12 condition. Therefore, possibilities to improve value are limited as a consequence of only implementing TDABC in one department (Berwick et al., 2008; Porter & Lee, 2013). One reason for the limited application of TDABC in healthcare is the initial effort and money which is needed to implement and maintain the tool (Demeere et al., 2009; Hennrikus et al., 2012). Another reason forwarded is that it takes a lot of time to analyze all the gathered data. Finally, providers may resist implementing a cost methodology that is completely transparent (Martin et al., 2018)

Overall, based on literature it can be concluded that accurately measuring and assigning costs is still very challenging for hospitals. Although Porter (2009) stresses the importance of accurately measuring costs, a detailed explanation of how to appropriately assign all types of costs is missing. TDABC may have the potential to solve this problem, but we should be careful to conclude that TDABC can actually solve the problems within healthcare. Until now empirical evidence of a successful integration of TDABC along the full cycle of care is still lacking (Keel et al., 2017). Besides, when implementing TDABC, different hospitals in- and excluded different cost components and used different methods to obtain time estimates (Keel et al., 2017). This reduces the possibilities to compare costs between hospitals. However, reaching consensus about the validity of data used for organizational comparison is needed for successful benchmarking (Ettorchi-Tardy et al, 2012). Therefore, the data used in this research is based on a method which makes use of standardized cost prices. The standardized price of an activity or resource is calculated by taking the average of the real price for an activity or resource which were handed in by ten surgical teams. For every surgical team, the same costs are assigned to the same activity and use of materials. This makes it possible to compare the costs of surgical teams based on their activities and resource usage. Consequently, this research shows that the argument of Porter and Kaplan (2011) that benchmarking without TDABC is more difficult is not necessarily true. This is done by demonstrating the method of standardized cost prices, which is relatively easy to implement.

(14)

13

2.3 Learning from data

As is previously mentioned, Porter and Lee (2013) emphasize that value will be improved if providers start measuring and comparing outcome and cost data. However, a detailed explanation about how data can be used and represented and how providers could actually learn from outcome and cost data is not given, while this is important because “the way how data are acquired, analyzed, interpreted, communicated and acted upon has an impact on business unit performance” (Bourne et al., 2005, p.374). Besides, simply analyzing and displaying data will not result in change within healthcare (Mannion & Davies, 2002). The way which would help providers to learn best is almost completely neglected in the articles about VBHC. Other previous studies of VBHC have also not dealt with this topic. Therefore, this study is trying to explore this topic in order to identify and clarify how surgical teams can actually learn from measuring value.

A management approach from industry which may facilitate healthcare providers to actually learn from measuring value is benchmarking. Benchmarking highlights the importance of undertaking learning in a structured formal approach (Cox et al., 1997). Consequently, it can be a very useful approach in supporting the development of clinical practices because of its structured assessment and reflection (Ellershaw et al., 2008). However, a misconception concerning benchmarking is that it is the simply comparison of indicators (Ettorchi-Tardy et al, 2012; Thonon et al., 2015). Benchmarking within a healthcare context can be defined as:

“a process of comparative evaluation and identification of the underlying causes leading to high levels of performance. It involves a sustained effort to measure outcomes, compare these outcomes against those of other organizations to learn how those outcomes were achieved, and apply the lessons learned in order to improve” (Ellis, 2006, p.1).

(15)

14 be universally applied, which is also the reason that the model is very extensive. This was another reason why this framework is chosen, since this framework makes it possible to study how surgical teams learn from outcome and cost data in detail. However, not all phases and steps of the framework are relevant for this research, because this research is only interested in the phases which could have influenced how surgical teams have learned from the data during the VBHC project. Therefore, some phases and steps within phases are excluded, such as partner selection, because this step was executed by an external institution. Therefore, the surgical teams had no influence on that. The phases included are: team formation, subject identification, management validation, self-analysis, benchmarking, gap analysis, action plans, implementation and continuous improvement. An overview of the phases, the related steps and descriptions included for this research can be found in appendix B.

(16)

15

3. Methodology

In this section, the choice for conducting a case study will be elaborated. Next, the research setting and the case selection will be discussed. Third, the operationalization of the concepts and the way how the data is collected will be addressed. Lastly, the methods which are used for data analysis will be discussed.

3.1 Choice for case study

The case study method is primarily chosen due to the aim of the research. The aim of this study is to gain new insights in how outcome and cost data can be measured, analyzed, represented and used in improvement activities to increase value for surgical teams and to explore how surgical teams can actually learn from outcome and cost data. Until now, very little is known about this subject and thus can this research be seen as explorative. Therefore, an in-depth case study will be conducted, since a case study is widely recognized as a method to adequately answer how and why questions with an explorative nature (Meredith, 1998; R. Yin, 1994). Additionally, a multiple case study approach is chosen, since this allows for comparison between cases and augment external validity and guard against observer bias (Eisenhardt & Graebner, 2007; Voss et al., 2002). Furthermore, two methods are used for collecting case data, namely historical data analysis and interviews. This enables triangulation, which is the use of different methods in order to increase the validity and reliability of the research (Karlsson, 2016).

3.2 Setting

(17)

16 basis of outcome and cost. Costs were measured by using standardized cost prices, since they encountered too many difficulties and problems with comparing the real cost prices, which were handed in by the surgical teams. For the outcome measurement, Textbook Outcomes are created and used for the gall bladder as well as for the inguinal hernia. A schematic overview of the general patient flow for a patient with an inguinal hernia or gall bladder disease can be found in appendix D.

3.3 Case selection

A clearly defined unit of analysis is a prerequisite for a well-structured case study (Yin, 2009). In this research, the unit of analysis are surgical teams. The surgical teams on which this research focuses are teams which are treating patients with an inguinal hernia or gall bladder disease. These diseases are chosen out of the five diseases, because these diseases are high volume low complex interventions. The pro of this is that the outcome measure is less influenced by case mix factors compared to a complex low volume intervention. For instance, with rectal carcinoma, the place of the tumor or whether there are two tumors instead of one will have a big impact on the outcomes and cost. Therefore, causes of differences in value are harder to identify compared to a low complex disease. Furthermore, a TO is developed for the inguinal hernia and gall bladder. By choosing these diseases, this research can demonstrate the value of TO in practice by showing which insights can be obtained from TO measurement. Not all ten surgical teams who participated in the Wvz project wanted to participate in this research. Therefore, this research includes five surgical teams, which is in adherence to the guidelines for a case study according to Eisenhardt (1989).

3.4 Data collection

(18)

17 which is based on Textbook Outcome (TO) and cost data. A TO is a composed measure of clinical process indicators. An outcome can be labeled as TO if all desired short-term health indicators are met (Kolfschoten et al., 2013). Additionally, the use of TO is especially suitable for clinical interventions such as surgery and invasive diagnostics (Salet et al., 2018). The use of textbook outcome measurement for inguinal hernia and gall bladder is unique, since until now, there does not exist a textbook outcome for these interventions in literature. For the inguinal hernia and gall bladder, a patient is considered TO if all six indicators are met. These are shown in table 1 and 2.

Table 1: Definition TO for inguinal hernia

Table 2: Definition TO for gall bladder

The data makes it also possible to analyze whether surgical teams made progression on the indicators individually and to assess the impact of clinical indicators where TO was not met. The outcome of the data analysis is also used as input for the interviews and to check whether improvement activities forwarded in the interviews also resulted in desired value improvement. Costs are measured by making use of standardized cost prices. For every surgical team, the same costs are assigned to the same activity and use of materials. To give an example, the costs for one nursing day for every surgical team is set on €387,50. This price is based on the real costs of a nursing day which were handed in by the surgical teams at the beginning of the Wvz project. Costs are assigned to an activity, and an activity is part of an activity group. Therefore, costs can be compared on two different levels of detail. By summing up all the activities and resources used by a given patient, the total costs of that patient can be calculated. By comparing the average costs per patient per activity with other surgical teams, cost differences and therefore room for improvement can be identified. The numbers shown in the results are the average costs per patient per activity. To give an example, the standard cost price for a CT scan

Criteria which has to be met in order to consider a patient TO

1. No re-admission 2. No IC admission 3. Inpatient stay < 2 days

4. Operation room hours < 1 hour

5. No outpatient consult within 1 week after the operation 6. No lab request within 2 weeks after the operation

Criteria which has to be met in order to consider a patient TO

1. No re-admission 2. No IC admission 3. Inpatient stay < 3 days

4. Operation room hours < 2 hour

(19)

18 is set on €131. If ten out of 100 patients made use of a CT scan, the average price for the activity CT scan per patient for that surgical team would be ((10 x €131)/100) = €13,1. An overview of some activities and the related standard cost prices can be found in appendix E. The outcome and cost data of all five surgical teams is provided by Value2Health (Deventer, The Netherlands). However, Value2health received cost data from two different companies (Performation provided cost data for surgical team 1&2, and LOGEX for surgical team 3-5) which both have their own standardized cost prices. Both companies use approximately the same cost prices, except for the surgery costs. The surgery costs calculated by Performation are way lower (see appendix I). Therefore, the costs for surgical team 1&2 cannot be directly compared with the costs of surgical team 3-5. In order to overcome this problem, it was tried to apply the LOGEX method on the data which was provided by Performation and vice versa, but this turned out to be impossible due to the different formats used by the different companies. Furthermore, not every surgical team registered the OR time or ASA score and for some surgical teams the data concerning the lab requests were not valid and could therefore not be used.

(20)

19

ST Role Date

1 Department manager surgery 08-05-18

1 Gastrointestinal surgeon and chairman surgical department 09-05-18

2 Department manager surgery 16-5-18

2 Gastroenterological and oncological surgeon 16-5-18

3 Cluster manager 30-5-18

3 Vascular and general surgeon 31-5-18

4 General and vascular surgeon 16-4-18

5 Operational manager 6-6-18

6 Oncological surgeon 11-6-18

Table 3: Overview of performed interviews

(21)

20 The semi-structured interviews are used in order to explore how surgical teams actually learned from data. The interview questions are related to the framework of Anand and Kodali (2008), since this framework is based on best-practices benchmarking models. Therefore, this framework can function as a solid foundation for exploring learning practices, also because benchmarking emphasizes the importance of learning in a methodical way (Cox et al., 1997). By framing the research by existing theory, external validity will be increased (Karlsson, 2016).

3.5 Data analysis

3.5.1 Quantitative data analysis

In order to identify whether the surgical teams have learned, outcome and cost data before the beginning of the Wvz project are compared with outcome and cost data after the Wvz project. The analysis is performed by using Excel. For the comparison, two groups are made. Group 1 includes patients who have been treated between January 2016 and September 2016. In the results, they will be referred to as “2016”. The second group includes patients who have been treated after May 2017 until now and will be referred to as “2017/18”. In order to ensure a fair comparison between the groups, it is decided to only include patients with the same DBC (diagnose treatment combination). Therefore, for patients with a gall bladder disease, only patients with cholecystitis (DBC 323) are compared. In the dataset, patients with cholecystitis are accountable for 92% of the total patients with a gall bladder disease. For the inguinal hernia, it is decided to compare only patients with hernia femoralis / inguinalis (DBC 121). In the dataset, patients with hernia femoralis / inguinalis are accountable for 95% of the total patients. Furthermore, two control variables are used in order to control if the two groups are in fact comparable. One control variable is the ASA score. The ASA score is used by anesthesiologists in order to assess the patient’s overall health based on five classes before the surgery (Daabiss, 2011). The classification scheme can be found in appendix G. The second control variable which is used is age, since age seems to be a driver for Textbook Outcome and costs for the gall bladder and inguinal hernia. Table 4 and 5 show this for the inguinal hernia and gall bladder respectively.

Age category <50 50-59 60-69 70-79 80+

Textbook outcome 62% 54% 50% 41% 38%

Average costs per patient €3400,- €3450,- €3500,- €3600,- €3950,-

Table 4: Overview of TO and average costs per patient per age category for inguinal hernia

Age category <50 50-59 60-69 70-79 80+

Textbook outcome 63% 58% 54% 41% 24%

Average costs per patient €4900,- €5250 €5400,- €6350,- €7700,-

(22)

21 3.5.2 Qualitative data analysis

(23)

22

4. Results

The following section discusses the findings of the empirical study. First, section 4.1 will discuss the results concerning the measurement of value. Second, section 4.2 discusses the results of the within-case analysis of each surgical team. Third, section 4.3 will discuss the results of the cross-case analysis in order to examine the differences and similarities in learning outcomes. Lastly, the results of the additional research concerning the contextual conditions which should be in place in order to learn will be discussed in section 4.4. The outcomes of the coding process provided the information for the analysis. Besides, quotes are also used to clarify the results. Respondent S-1 refers to a surgeon of surgical team 1, whereas respondent M-2 refers to a manager from surgical team 2. Furthermore, a detailed cost comparison of each surgical team for the inguinal hernia as well as for the gall bladder can be found in appendix I.

4.1 Measuring value

Textbook Outcome

(24)

23 useful measure which have certainly potential, but also needs to be further developed and validated.

Standardized cost price

Costs were measured by making use of a standardized cost price. Based on the interviews, a few things became clear. First of all, some interviewees mentioned that it is currently simply not possible to measure and compare real cost prices. This is reflected by the following quotes: “In my surgical team, costs are not patient bound, therefore it is very difficult to calculate and compare real cost prices.” (S-2) and “No, that [comparing real costs] is not going to happen. That is impossible” (S-6). One surgeon believes that it may be possible to implement a system which can more accurately measure real cost prices, however: “That would cost a lot of time and money, and that’s exactly what we both don’t have” (S-2). Second, because of that, most interviewees forwarded that using a standardized cost price is the only way to currently compare prices and that it is therefore the second-best option. This is illustrated by the following quotes: “If you want to compare prices anyway, this is the most obvious method” (S-4-2) and “I believe that it does tell us something on the cost aspects which are included. This is the best method we currently have” (S-4-1). Lastly, they also emphasized that using a standard cost price limits the learning opportunities. For example, S-1 mentioned: “You can discuss about it, but it also has flattened out a lot”. The standardized cost price did for instance not include expensive material, used for surgeries. This was also seen as a con: “For some surgeries, an expensive stent is needed. The price of these stents however differs a lot. These costs are not included but are very relevant” (S-4-1). If you want to compare prices, it was also mentioned that you should be aware of the case mix. Summarized, the standardized cost price was not seen as the ideal method in order to compare costs and maximize learning outcomes, but as the best method which is currently available due to difficulties in measuring and comparing the real cost price.

4.2 Within-case analysis

Surgical team 1

(25)

24 that they had the intention to learn from the project. “We wanted to get a better insight in our own performance in comparison to others. Additionally, we hoped to learn something from others.” (M-1). Furthermore, the goal of the project was in line with the goal of the surgical department: “The goal of this project was totally in line with our own department goals, namely paying attention to the relation of cost and quality in order to improve the value of care” (S-1). The ST did not provide any resources for this project. The analysis of the data is shown in table 6 and 7. Table 6 reveals that ST 1 was not able to decrease the costs for patients with an inguinal hernia. Looking at the TO, we see a huge increase of 55,24%. Looking at the underlying indicators, it becomes clear that this increase resulted from an increase of 61,48% on indicator five. However, there was no real change in the process of ST 1. “This is exactly why you should watch very critical to things like this” (S-1). The low score on indicator five in 2016 was because of a nurse called the patient after the surgery, to ask how he/she was doing. However, this was not a real consult and therefore it was not labeled as consult anymore in 2017. Table 6 also shows that ST 1 reduced the re-admissions by 4,19%. However, they could not clarify this.

Inguinal Hernia 2016 2017 Difference

Number of patients 316 115

Average ASA score 1,54 1,63 0,08

Average age 57,07 56,17 -0,90

Average costs per patient € 2381,- € 2426,- € 45,-

% of patients TO 22,15% 77,39% 55,24%

1. No re-admission 94,94% 99,13% 4,19%

2. No IC admission 100,00% 100,00% 0,00%

3. Inpatient stay <2 days 100,00% 97,39% -2,61%

4. OR time < 1 hours 96,84% 96,52% -0,31%

5. No policonsult within 1 week after OR 30,70% 92,17% 61,48% 6. No lab request within 2 weeks after OR 86,08% 85,22% -0,86%

Table 6: Cost and outcome comparison inguinal hernia ST 1

(26)

25 in 2016 and 2017 (average costs per patient is around 20% lower compared to the average costs of inpatient treatment).

Gall bladder 2016 2017 Difference

ASA score 1,68 1,65 0,03

Average age 51,77 51,73 0,03

Average costs per patient € 4.041,- € 3.604,- € 437,-

%TO 72,05% 70,42% -1,63%

1. No re-admission 96,89% 95,77% -1,12%

2. No IC admission 98,76% 98,59% -0,17%

3. Inpatient stay <3 days 88,20% 95,77% 7,58%

4. OR time <2 hours 97,20% 97,18% -0,02%

5. No policonsult within 1 week after OR 84,78% 84,51% -0,28% 6. No lab or echo request within 2 weeks after OR 91,30% 90,14% -1,16%

Table 7: Cost and outcome comparison gall bladder ST 1

(27)

26 which was provided during the project. However, the manager saw the tool a good starting point for improving value.

Surgical team 2

For ST 2 an interview is performed with an oncological / gastroenterological surgeon and the department manager of surgery. “For such projects, a mix of people with different backgrounds is needed. Surgeons for the medical side and a manager for the process side” (M-2). Besides, a data manager participated in the project. Before the project, there was no arrangement regarding the division of tasks nor was someone specifically assigned to focus on the inguinal hernia or gall bladder. From the interviews, it became clear that ST 2 had the intention to learn from the project and that the project was fully in line with the goals of the department: “The intention was to improve care. On the cost or quality side, but preferably both” (S-2) and “The project goal was in line with our department goal, namely optimizing quality and costs” (M-2). The surgeon had to execute this project in his own time, but for the manager and data manager, resources were provided by the department. Table 8 and 9 (see next page) show the outcomes of the data analysis for the inguinal hernia and gall bladder respectively. Table 8 shows that the costs and TO score both marginally increased. The detailed cost analysis shows that the small increase in costs is mainly caused by a cost increase in the activity “nursing days”. The average nursing days increased from 0,2 to 0,286 and might be partly explained by the higher average ASA score in 2017. Looking at the TO, one indicator that stands out is indicator four. The interviews revealed that this indicator was not reliable, because within the hospital of

Inguinal Hernia 2016 2017 Difference

ASA score 1,66 1,78 0,12

Average age 59,81 60,55 0,74

Average cost per patient € 1983,- € 2008,- € 25,-

%TO 77,88% 78,74% 0,86%

1. No re-admission 96,94% 100,00% 3,06%

2. No IC admission 100,00% 100,00% 0,00%

4. OR time <1 hours 22,94% 28,46% 5,52%

5. No policonsult within 1 week after OR 91,44% 93,50% 2,06% 6. No lab request within 2 weeks after OR 85,32% 84,55% -0,77%

Table 8: Cost and outcome comparison inguinal hernia ST 21

1_{Indicator 4 is not included in the calculation of the TO score, since this indicator is not valid for the inguinal}

(28)

27 ST 2, they also measured the time before and after the OR, while indicator 4 is only about the OR time. Therefore, this indicator should be neglected for the inguinal hernia. During the Wvz project, ST 2 decided to use less different surgeons for inguinal hernia surgeries. However, this was not initiated due to the insights from the Wvz project. Besides this change, ST 2 had no clarification for the increase of 3,06% on indicator 1, since they did not change anything else related to the inguinal hernia during the Wvz project. Table 9 reveals that for the gall bladder,

Gall bladder 2016 2017 Difference

ASA score 1,69 1,77 0,09

Average age 53,67 55,68 2,01

Average cost per patient € 2.875,- € 3.123,- € 249,-

%TO 54,90% 42,03% -12,87%

1. No re-admission 82,17% 71,74% -10,43%

2. No IC admission 100,00% 100,00% 0,00%

4. OR time <2 hours 85,31% 87,68% 2,37%

5. No policonsult within 1 week after OR 81,82% 73,19% -8,63% 6. No lab request within 2 weeks after OR 87,76% 81,88% -5,88%

(29)

28 research on the re-admission rate for the gall bladder, ST 2 did not collect any additional data concerning the inguinal hernia or gall bladder. Furthermore, the insights of the data are only communicated orally within the surgical department. They did not set any goals. Additionally, they did not search for root causes (besides the re-admissions), did not make an action plan and did not change or implement something. Within ST 2, there is no reward for improving value, while they believe that this should be changed: “Although the intrinsic motivation is way more important, but there is a line” (S-2). Besides the argument about the transition, other reasons are forwarded why they were not able to learn from the data. “Despite the idea of mixing financial and quality data is very charming and the fact that it led to insights that there were differences in value, it did not result into insights in where the differences came from. [..] The complexity lies in finding the hook. We simply could not find that hook” (M-2). According to the surgeon, this is partly caused by a lack of openness: “If everyone had said:“Let’s get everything out on the table”, we would probably have learned more”. Another reason forwarded was: “The TO is still in development, therefore it is hard to tell if this is the right way to do such a project” (S-2). The manager did not come up with that reason. Although the disappointing results, the manager still believes that there is much to improve and save concerning the gall bladder and inguinal hernia: “There is still a lot to earn. It is about lower margins, however it is also about a high volume. Price times quantity and you can do the math” (M-2).

Surgical team 3

(30)

29 the Wvz project. Table 10 and 11 show the outcomes of the data analysis for the inguinal hernia and the gall bladder. The average costs per patient shows a decrease of 8,5%. The detailed cost analysis shows that this saving is the result of less surgery costs (probably due to less re-admissions), less IC costs and less costs in consults. Furthermore, the re-admissions decreased by 4,57%, but surgical team 3 had no specific clarification for this improvement.

Inguinal Hernia 2016 2017/18 Difference

Number of patients 357 229 -

ASA score - - -

Average age 61,01 59,87 -1,14

Average costs per patient € 3.754,- € 3.436,- -€ 318,- %TO (only based on indicators 1,2,3,5) 78,43% 77,29% -1,14%

1. No re-admission 94,12% 98,69% 4,57 %

2. No IC admission 98,32% 100,00% 1,68%

4. OR time <1 hours - - -

5. No policonsult within 1 week after OR 93,84% 87,11% -6,73%

6. No lab request within 2 weeks after OR - - -

Looking at the results of the gall bladder (see table 11), the data analysis shows that ST 3 improved the scores on all TO indicators and also realized a cost reduction. The cost reduction is the result of a reduction in costs concerning nursing days and consults. ST 3 changed one thing related to the inguinal hernia and gall bladder: “We started to use more standardized disposable instruments which saves money” (S-3). However, this change was not

Gall bladder 2016 2017/18 Difference

ASA score - - -

Average age 55,82 53,71 2,11

Average costs per patient € 5.299,- € 4.951,- -€ 347,- %TO (only based on indicators 1,2,3,5) 61,29% 69,72% 8,43%

1. No re-admission 94,93% 95,27% 0,34%

2. No IC admission 97,24% 98,74% 1,50%

4. OR time <2 hours - - -

5. No policonsult within 1 week after OR 76,96% 84,54% 7,58% 6. No lab or echo request within 2 weeks after OR - - -

(31)

30 initiated due to insights of the Wvz project and the costs of instruments are not included in this research. However, what might clarify the improved outcome and decreased costs is that the surgeon mentioned that the Wvz project created more “awareness” concerning the inguinal hernia and gall bladder. Additionally, “The feedback about the performance worked very stimulating. We started to steer more actively on the performance” (M-3). Furthermore, it might be the case that their value improved because they became more experienced with their new way of working (their new “streets”). On the question whether they have learned anything with regard to the inguinal hernia or gall bladder the manager answered: “No, it only confirmed that we are in the right direction. Besides, we got insights in what is possible with such data. The next step is to see how we can use these insights and to eventually make some modifications.” Furthermore, ST 3 did not perform a self-analysis and did not collect any additional information: “No we didn’t, since have already a very sharp care pathway” (M-3). ST 3 only searched for root causes concerning the IC admissions for the gall bladder, but this did not lead to new insights or changes. The insights of the Wvz project were orally communicated within the department of surgery: “They are surgeons, they often don’t write a lot” (M-3). They did not make an action plan and did not set any goals: “We did not find something to improve” (S-3). ST 3 does not reward continuous improvement. “I think it’s absurd to reward improvement. That should come from intrinsic motivation” (M-3). However, the surgeon had a different view, since he thought that rewarding improvement is something to think about. Besides the reason that ST 3 was best practice for the gall bladder and that they could not find anything to improve, they forwarded other reasons why they did not learn. “We paid too little attention to the inguinal hernia and gall bladder” (S-3) and “We did not contact other best practice surgical teams, probably due to time constraints “(M-3). During the Wvz project, ST 3 developed also “streets” for the carotid and vascular surgery, which might partly explain why they paid less attention to the inguinal hernia and gall bladder. Additionally, “The learning possibilities were also obstructed by the fact that there was no open data. […] I also think that there were too little opportunities to share knowledge. More thematic focused meetings are probably better” (S-3).

Surgical team 4

(32)

31 medical and managerial people would be the best” (S-4-1). There was beforehand no division of responsibilities and there was no one specifically designated to be concerned with the gall bladder or inguinal hernia. From the interviews, it became clear that ST 4 participated in the Wvz mainly because of three reasons: “To see if we could improve the quality of care or reduce the costs” (S-4-1), “To get more insights in the relation of quality and costs” (S-4-2) and “To see what is possible with outcome and cost data” (S-4-1). The goal of the project was also in line with the goals of the department. The employees executed the project mostly in their own time. Table 12 shows the results of the data analysis for the inguinal hernia. For the inguinal hernia, the average costs per patient decreased marginally. This decrease is the result of small savings concerning nursing days and consults. Besides, table 12 shows an improvement of performance on 4 indicators, which in turn resulted in an increase of TO score by 13,28%. These improvements might be partly explained by the following quote: “We decided to start working in a more standardized way concerning the inguinal hernia. Besides, we decided to make some more experienced surgeons lead” (S-4-3). However, according to S-4-3, these changes were not implemented due to the Wvz project.

ASA score - - -

Average age 56,91 54,57 -2,34

Average costs per patient € 3.561,- € 3.514,- -€ 47,-

%TO (only based on indicators 1,2,3,5) 60,67% 73,95% 13,28%

1. No re-admission 96,63% 99,07% 2,44%

2. No IC admission 100,00% 100,00% 0,00%

4. OR time <1 hours - - -

5. No policonsult within 1 week after OR 88,76% 93,49% 4,73% 6. No lab request within 2 weeks after OR 86,52% 93,02% 6,50%

(33)

32

Number of patients treated by day care 0 7 -

ASA score - - -

Average age 55,42 54,16 -

Average costs per patient € 6.088,- € 5.372,- -€ 716,-

Average costs per patient (day care) - € 3755,- -

1. No re-admission 96,43% 98,13% 1,70%

2. No IC admission 97,86% 100,00% 2,14%

4. OR time <2 hours 95,54% 92,01% -3,53%

5. No policonsult within 1 week after OR 85,71% 90,64% 4,93% 6. No lab or echo request within 2 weeks after OR - - -

Gall bladder Percentage of patients Average nursing days

Cost categorie 2016 2018 2016 2018

<€5.000,- 62,50% 67,42% 3,02 2,94

€5.000-€10.000,- 26,79% 27,34% 5,84 5,92

<€10.000,- 10,71% 5,24% 15,67 11,21

Table 14: Comparison of patient groups concerning cost categories and average nursing days

(34)

33 Currently, we are also busy with others projects. We have to make time to put it all together and to see what is possible” (S-4-1). ST 4 did not search for root causes and communicated the insights of the Wvz project orally within the surgical department and with the business manager. Additionally, ST 4 did not set any future goals and did not draw up an action plan. “No, I have to be very honest that drawing up an action plan is a point for improvement for our department and also a target for ourselves” (S-4-2) and “The problem with such projects is that, they are nice projects, it shows the painful areas. However, we lack time to pick up such projects, we need the help of others” (S-4-1). ST 4 does not reward continuous improvement, while both surgeons believe that rewarding improvement could be beneficial: “Rewarding improvement is not common within our hospital. However, people will go one step further if you share the savings” (S-4-1). The main reasons forwarded by ST 4 why they have not learned (yet) is because of time constraints. Besides, ST 4 also experienced a lack of openness and that the meetings were a little bit impersonal. The surgeons saw the Wvz project as a mirror with information, which shows the strong and weak points and therefore provides good starting points for improvement. However, it must be acknowledged that: “In order to improve value, much more effort and more detailed additional research is needed [besides looking at differences in the underlying indicators of the TO]” (S-4-1).

Surgical team 5

For ST 5, an interview is performed with an operational manager of the surgical department (M-5) and some additional questions were asked to an oncological / gastrointestinal surgeon (S-5). Furthermore, other surgeons, an accountant and a project manager also participated in the project. Also, another business manager and someone of the managing board were also involved but did not participate in the meetings with the other surgical teams. There was no division of tasks or responsibilities and there was no one specifically designated to the inguinal hernia or gall bladder. ST 5 clearly had the intention to learn: “We were very curious about our performance and we were wondering if we could learn something from other surgical teams (M-5). The goal of the project was also in line with the goal of the department: “Our goal is to maximize the quality of care and to do that as inexpensive as possible” (M-5). ST 5 is the only

(35)

34 surgeries laparoscopic. It might be the case that surgeons became more trained. However, besides that, I have no clarification for the difference, since we did not change anything during the Wvz project”.

ASA score - - -

Average age - - -

Average costs per patient € 3.402,- € 3.231,- -€ 171,- %TO (only based on indicators 1,2,3,5) 67,47% 70,87% 3,40%

1. No re-admission 95,18% 98,03% 2,85%

2. No IC admission 100,00% 100% 0,00%

4. OR time <1 hours - - -

5. No policonsult within 1 week after OR 80,72% 79,13%% -2,02%

6. No lab request within 2 weeks after OR - - -

Looking at table 16, we see that the costs for the gall bladder increased and that the TO score stayed the same. On the question whether ST 5 changed something related to the gall bladder the answer was: “No, however we learned from other teams that you can save costs by performing one day treatment for a part of the patients. Therefore, we are also considering to do that” (M-5) and “Performing day treatment is the fastest way to reduce costs. However, we are too busy with others things so we have not implemented this yet” (S-5).

ASA score - - -

Average age - - -

Average costs per patient € 4.996,- € 5.117,- € 121,-

1. No re-admission 99,54% 96,79% -2,75%

2. No IC admission 99,08% 100,00% 0,92%

4. OR time <2 hours - - -

5. No policonsult within 1 week after OR 90,83% 88,53% -2,30% 6. No lab or echo request within 2 weeks after OR - - -

(36)

35 a self-analysis since they did not have a measure yet. ST 5 did not collect any additional information and besides did not search for root causes. They communicated the insights of the Wvz project only orally. ST 5 set one goal, namely to perform day treatment for gall bladder patients. Besides, ST 5 did not write an action plan and did not implement an intervention. ST 5 does also not reward continuous improvement: “No, we don’t. I think that intrinsic motivation is most the important. However, a combination of incentives would be the best” (M-5). ST 5 was already happy with their delivered value, which was also one of the main reasons that they have not learned that much. This is illustrated by the following quotes: “We are already happy with our value of care. In such a situation, you will be less eager to take action” (M-5). Another reason forwarded was: “In my opinion, such a project should not be anonymized. I know this is something very sensitive, but I believe that you learn a lot more when everything is transparent” (M-5). Furthermore, ST 5 experienced the meetings as a little bit impersonal, which obstructed the information sharing. Lastly, “There was no real shared goal within the group of hospitals. It might be better if you define beforehand what you want to achieve together” (S-5). The manager of ST 5 however believes that the project provided enough possibilities to learn something from each other and added: “If you see a performance gap, it is your own responsibility to actively deal with it.”

4.3 Cross-case analysis

In this section the differences in learning outcomes will be discussed. Table 17 and 18 on the next page give an overview of the learning outcomes. Looking at the costs2, we see that ST 3 realized the biggest saving and that ST 2 and 5 have the lowest costs per patient. Looking at the TO scores3, it becomes clear that ST 4 realized the biggest increase (neglecting ST 1, because that difference is based on wrong data). They were also the only ST who changed something (less lab request and a more standardized way of working). ST 1, 2 and 3 almost have the same TO score, while ST 4 and 5 score a little bit lower on TO. Looking at the costs2, we see some variation between the surgical teams, but the differences overall are not that big, except for the difference between ST 1 and 2. In general, the surgical teams mentioned that they did not

2_{For ST 1 and 2, the cost data is provided by a different company which uses different standard cost prices,}

compared to the company which provided the data for surgical team 3-5. Therefore, the prices of ST 1 and 2 cannot be compared directly with ST 3-5.

3_{ST 3 and 5 did not provide data concerning the OR time and lab request. Therefore, it should be noted that for}

(37)

36 change anything during the Wvz project, or at least not because of the insights of the Wvz project as became clear in the within-case analysis.

Ing. hernia Average costs per patient Percentage of patients with TO

ST 2016 2017/18 Difference 2016 2017/18 Difference 1 € 2381,- € 2426,- 1,89% 22,15% 77,39% 55,24% 2 € 1983,- € 2008,- 1,26% 77,88% 78,74% 0,86% 3 € 3.754,- € 3.436,- -8,46% 78,43% 77,29% -1,14% 4 € 3.560,- € 3.513,- -1,32% 60,37% 73,95% 13,28% 5 € 3.401,- € 3.230,- -5,03% 67,47% 70,87% 3,40%

Table 17: Cost and TO comparisons inguinal hernia23

For the gall bladder, table 18 shows that ST 1 and 4 realized the biggest savings. This is not strange, since they were both relatively expensive compared to the other surgical teams at the beginning of the project. Besides, the within-case analysis shows that ST 4 was the only ST who implemented some interventions. Additionally, ST 4 also improved their TO score by 3,76%. ST 3 realized a cost reduction and also realized the biggest increase in the TO score of all ST. As mentioned before, in the year before the beginning of the Wvz project ST 3 implemented the so-called gall bladder and inguinal hernia “streets”. That they were able to improve the value of care during the Wvz project might be explained by more motivation/awareness and the fact that they become more experienced with their new way of working.

Gall bladder Average costs per patient Percentage of patients with TO

ST 2016 2017/18 Difference 2016 2017/18 Difference 1 € 4.042,- € 3.604,- -10,84% 72% 70% -1,63% 2 € 2876,- € 3.038,- 5,65% 55% 42% -12,87% 3 € 5.299,- € 4.951,- -6,55% 61% 69% 8,43% 4 € 6.088,- € 5.371,- -11,76% 54% 58% 3,76% 5 € 4.996,- € 5.116,- 2,42% 66% 66% 0,00%

Table 18: Cost and TO comparisons gall bladder45

Furthermore, ST 5 learned that gall bladder patients can be treated by day care. However, they are not performing day care yet. Looking at the costs, we see some serious differences,

4_{ST 3 and 5 did not provide data concerning the OR time and lab request. Therefore, it should be noted that for}

hospital 3 and 5 indicators 4 and 6 are not included in the TO score. As a result, the TO scores shown in table 18 are higher than in reality. ST 4 had no reliable data concerning indicator 6. Therefore, indicator 6 is excluded in the TO score for ST 4. For surgical team 2, indicator 2 (re-admissions) was not 100% reliable, therefore the TO score is in reality higher than is shown in table 18.

5_{For ST 1 and 2, the cost data is provided by a different company which uses different standard cost prices,}

(38)

37 especially at the beginning of the project. Also, for the TO score there is some variation, despite the fact that this is not a 100% fair comparison4. Furthermore, ST 2 and 3 are the only surgical teams who are treating a serious number of patients by day care. They also have the lowest average costs per patient. Summarized, based on the results of the interviews, it seems to be the case that the surgical teams did not learn much, although the data shows some value improvement for some surgical teams.

4.4 Learning conditions

In contrast to what was expected, the results of the within-case analysis show that the surgeons and managers mentioned that the learning outcomes were very limited. This raised another question, namely: “Which conditions should be met in order to learn from outcome and cost

data”? During the interviews, questions were asked why their surgical team did not learn much

from the outcome and cost data and under which condition they believe they would have learned more. By critically reflecting on their one-year experience of participating in a VBHC project, they were able to forward several aspects and conditions which should be met in order to learn from outcome and cost data. The results are discussed in this section.

A culture which emphasizes self-improvement instead of competition

(39)

38 don’t like the outcomes, you are going to search for reasons to hide behind” (S-4-1). Therefore, a culture is needed where people are not ashamed about their numbers, feel free to ask each other questions and where the focus is on becoming better instead of performing better than others. This will enhance information sharing and learning possibilities.

Transparency and trust

Full transparency is a condition which is seen as very important in order to learn from outcome and cost data by most of the interviewees. This is illustrated by the following quote: “When everything is fully transparent, only then you can learn from each other” (S-5). Some of the interviewees even forwarded that they would have learned more if everything would have been fully transparent. Additionally, trust was also highlighted as an essential condition by a few interviewees. This is explained by the following quote: “Everyone can learn something, but only if there is mutual trust” (S-2).

Time and commitment

In order to reap the benefits and actually learn from insights provided by outcome and cost data, commitment and enough time are needed. From the interviews it became clear that these conditions seem to be partly the cause of the limited learning outcomes. This is illustrated by the following quotes: “A lot of people executing such a project on the side. They don’t get time for it, which limits the learning possibilities” (S-6) and “The problem is that we lack time to really pick up such nice projects” (S-4-2). This is also confirmed by the fact that almost every manager and surgeon had to perform this project in his/her own time, while they are already very busy with their daily tasks. Additionally, from the interviews it became clear that learning from data takes a lot of time: “It takes a lot of time to go in-depth in the data and to find out what you can actually improve” (S-6). Furthermore, full commitment is needed: “If you really want to learn something, it is very important that you beforehand decide to be fully committed to the project” (M-1). This also includes that you take insights of the data analysis serious and that you actively deal with that, even when the outcomes are not the desired outcomes.

Quick access to clearly represented, reliable and valid data

(40)

39 digestible way: “A surgeon is not the type of person who reads very well. Therefore, the data should be represented in a clear way” (M-1).

Data expert with medical knowledge

Measuring outcome and cost data will result in a lot of data which has to be analyzed and interpreted in order to learn from it. From the interviews it became clear that surgeons may have difficulties in interpreting this data. This is especially the case with cost data: “In the base, a surgeon is not educated to be concerned with costs. We have other people for that (M-5). Therefore, there is a strong need for people who can interpret and explain the measured data. This is illustrated by the following quote: “If you really want improvement, you need someone on the table who can interpret and explain the data” (S-4-2). Additionally, it is also very important that the person who interpret the data also has knowledge concerning the disease where it is about. A manager stated the following about that: “For surgeons, it is really important that the person on the other side of the table have a good understanding of things. If they don’t have that impression, it won’t work (M-1)”

Clear (shared) project goal

A few interviewees forwarded that focus is important and that a clear and mutually agreed project goal will enhance the learning outcomes. This is illustrated by the following quotes: “I would define beforehand what you want to achieve with such a project. Not just gathering data in the hope you find something. That is a shot in the dark” (S-1) and “I believe that it is better to jointly set goals before the beginning of the project about what you want to learn and achieve as a group” (S-5).

Consensus needed between surgical teams about the execution of the project

Prior to the start of the project, consensus should be reached about the methodology of the project and the extent to which the surgical teams are comparable. From the interviews, it became clear that some surgical teams just kept arguing about the methodology and therefore limited their learning outcomes. This is supported by the following quote: “There was some discussion about the methodology of the project. At some point, you have to stop with that. However, some hospitals continued to argue about it and therefore did not come any further” (M-2).

Theme-oriented meetings

(41)

40 organize theme or disease-oriented meetings preferably with not too many people if you want to learn from cost and outcome data when you want to include multiple diseases. According to some interviewees, this will enhance the knowledge sharing which in turn enhances the learning outcomes. This is explained by the following quote: “I believe that surgeons share more knowledge when you organize more disease-oriented meetings. If you have vascular and intestinal surgeons in the same room, that does not talk easily” (S-6). Furthermore, prior to the project, an arrangement should be made that the best performing surgical team should be linked to a less performing surgical team. Another option forwarded is that the best practice explains their process in detail from the beginning to the end. This should help to identify the root causes of the performance gaps. Some interviewees explained that they had difficulties to find out what they could improve. This is illustrated by the following quote: “The data showed the differences, however, it did not lead to insights were the differences came from” (M-2).

Lagging performance

(42)

41

5. Discussion

In order to answer the main research questions, this section discusses the empirical findings of this study in the context of prior literature.

5.1 Measuring value

The first question of this research is: How can outcome and cost data be measured, analyzed,

represented and used in improvement activities to increase value for surgical teams? As for the

measurement of outcome, this research shows that TO is a suitable method for evaluating and comparing outcomes and that TO provides useful starting points for improvement initiatives. Additionally, that the use of this method is not time-consuming was seen as a big advantage. These findings are in line with Kolfschoten et al. (2013) and Salet et al. (2018), who also concluded that without much effort, TO can provide a useful and comprehensive summary of clinical care which can be used for improving the quality of care. Besides, one impediment for reaching the goal of improving value by insights of data within healthcare is the underuse of existing uniform data (Schneeweiss, 2014). This is exactly why the TO has a lot of potential for the concept VBHC, because this method makes use of existing uniform data, which is something revolutionary. Therefore, this method should make it easier to measure the outcome side of value. However, despite that TO has a huge potential, it should not be forgotten that TO still has to prove its promises as a method for improving value. Therefore, more development and validation are needed.

Looking at the cost side, it was found that it is still not possible to accurately measure and compare costs which is in line with extant literature (Keel et al., 2017; Porter & Kaplan, 2016; Teckie et al., 2014). Nevertheless, this research shows a new method for measuring and comparing costs, which is relatively easy to implement and is seen as second-best despite its limitations. Since the implementation and use of TDABC is resource intensive (Hennrikus et al., 2012), it is not expected that this method will be implemented in the near future. Therefore, learning from measuring and comparing costs remains a challenge for the future. When outcome and cost data have been collected, it can be analyzed by using Excel and represented in clear tables as is shown in the results section of this research.

5.2 Learning from outcome and cost data

The second question of this research is: How can surgical teams actually learn from outcome

and cost data? Unfortunately, the within-case analysis shows that in contrast to what was

Insights in how surgical teams can improve value by learning from data Deriving value from measuring outcome and cost data within healthcare:

Deriving value from measuring outcome and cost data

within healthcare: