Chapter 4 Results diagnosis change readiness (sub-question 1)

(1)

CREATING A CONTEXT FOR CHANGE IN VALUE BASED HEALTH CARE IMPLEMENTATION – PROBLEM SOLVING STUDY OF PROM INTEGRATION

Master Thesis, MSc Change Management & MSc Health, Faculty of Economics and Business, University of Groningen

February 26, 2020

Loes Marie Sietske Visser S2520397

L.M.S.visser@student.rug.nl Folkingestraat 46-2 9711 JZ, Groningen

Thesis supervisors

Dr. M.A.G. van Offenbeek & Dr. N. Renting

(2)

L.M.S. Visser – S2520397 Master Thesis BA Change Management & Health

2

TABLE OF CONTENT

1 INTRODUCTION 5

2 LITERATURE 7

2.1 Towards Value Based Health Care 7

2.1.1 What - Patient Reported Outcome Measures 8

2.1.2 How – process 10

2.2 Change readiness 11

2.3 What to Change 14

3 METHODS 15

3.1 Research Context and Case Selection 16

3.2 Research Design 17

3.3 Data Collection 18

3.3.1 Data Collection: diagnosing readiness 19

3.3.2 Participants 21

3.3 3 Data Collection: designing interventions 21

3.3.4 Data Saturation 22

3.4 Data Analysis 22

3.4.1 Data analysis: diagnosing readiness 22

3.4.2 Data Analysis: Designing interventions 23

4 RESULTS DIAGNOSIS CHANGE READINESS 23

4.1 Change Content: Use of PROMs in palliative care 23

4.2 Change Process: PROM integration pulmonary unit 27

4.3 Change Context: PORM integration pulmonary unit 31

4.4 Individual Attributes 33

4.5 Facilitating and Constraining Forces other than Readiness Issues 34

4.6 Conclusion Readiness Diagnosis 35

5 DESIGNING CHANGE INTERVENTIONS 37

6 DISCUSSION 44

6.1 Summary Results Case Study 44

6.2 Practical Implications 45

Transferability PROM integration 45

Recommendations 46

(3)

3

6.3 Theoretical Reflection 48

6.4 Limitations and Future Research 49

6.5 Conclusion 50

7 REFERENCES 51

8 APPENDICES 58

A: Table 1: Contextual Factors influencing PROMs 58

B: Organizational Structure CU in MCL 59

C: Interview protocol First Version 60

D: Interview protocol Second Version 63

E: Data Structure 65

F: Codebook overview 66

G: Overview PROM aims 69

H: Factors determining the outcome of care (Mainz, 2003) 70

(4)

4

Abstract

Patient Reported Outcome Measurements (PROMs) are popular instruments used to measure health outcomes and often to increase patient value whilst striving toward a value based health care strategy.

Studies found justification for the use of PROMs as part of standard care, however, more studies on their usability in different contexts is needed. This research investigates how PROMs on the pulmonary oncology unit within the Medical Center Leeuwarden can be successfully integrated in daily care. To do so, the level of change readiness of the unit was assessed by looking at the factors content, process, context and individual attributes. This was done by one-on-one interviews and direct observation. The level of readiness was found to be medium to high. Based on literature and emerged key drivers found were: (1) driven, capable change initiators that function as champions; (2) high internal motivation of unit staff members; (3) high commitment and general openness to change; and (4) the availability of a digital infrastructure on ‘Mijn MCL’ and EPIC that offer opportunities for PROM data collection.

Observed barriers were: (1) uncertainties regarding changes in routines and work tasks due to PROMs (depends on chosen aim); (2) non-effective or lack of communication; (3) low participation and involvement of management and staff members in the change; (4) few financial resources due to the financial challenge of the organization; and (5) high work load of staff members and especially the nurse consultants. Based on these forces, interventions were designed to enhance unit- and organizational change readiness. Results regarding change readiness are transferable when the perceived urgency is low. Interventions regarding PROM integration appear to be less applicable for all contexts, as PROMs need a high level of customization.

Keywords: Patient Reported Outcome Measures; Integration; Quality of life; Value Based Health Care

(5)

5

Chapter 1 Introduction

The rising healthcare costs due to demographic changes and improved technology are becoming a problem (Porter, Pabo, & Lee, 2013). To manage these costs Porter (2013) argues from focusing solely on the volume and profitability of services provided to focusing on the core business of health care delivery system: delivering and improving care (Porter, Larsson, & Lee, 2016). Nowadays, an increasing number of health care delivery organizations describe enhancement of value for patients as a fundamental goal (Thomas & Lee, 2015). This value can be considered patient’s health outcomes achieved per dollar spent (Porter, 2009; Tseng & Hicks, 2016). The idea to redefine healthcare in terms of patient value and, hence, decrease costs lie at the basis of the concept value-based health care (VBHC). Porter and Teisberg (2006) claim that VBHC is the strategy that will fix healthcare.

However, Walshe (2009) addressed that the ongoing pattern of new management concepts, such as VBHC, is ‘pseudo-innovation’. Pseudo-innovation is the repetition and reinvention of essentially similar ideas and methods using different terminologies every 3–5 years. On top, also Fredriksson, Ebbevi, and Savage (2015) warned for the danger of the management hype VBHC and argued it may be undergoing a process of dilution rather than diffusion.

To guide organizations Porter and Teisberg (2006) designed the ‘value agenda’ which provides six steps that that are key for organizations aiming for a VBHC strategy. The second step focuses on measuring patients’ outcomes and costs. Patient reported Outcomes (PROs) represent the opinion and appreciation of the patients’ own health. PROs can include symptoms, physical functioning or quality of life (Bouazza, Chiairi, Kharbouchi et al., 2017). The increasing need of insights in the quality and efficiency of delivered care results in a growing attention for patient experiences and a higher demand for measuring treatment effects (Werken aan zorgkwaliteit, 2017). Questionnaires to measure these outcomes are the Patient Reported Outcome Measures (PROMs).

This academic problem-solving study was requested by MCL and performed at the pulmonary oncology unit. The Medical Center Leeuwarden, one of the 28 Dutch top clinical hospitals, is located in the province of Friesland in the Northern part of the Netherlands. MCL shows interest in a VBHC strategy, however, as this is a broad umbrella concept one focus was chosen. As PROMs are a The pulmonary oncology unit already started an experimental project with PROMs, this research focuses at PROM integration. The research answers the following research question:

“How can MCL pulmonary unit successfully integrate PROM indicators in the daily care for non-operable stadium III and stadium IV lung cancer patients?”

MCL is part of the association of Samenwerkende Topklinkische opleidings Ziekenhuizen (STZ) hospitals that exchange knowledge and have a focus on training, education and scientific research.

(6)

6 According to the ‘Strategienota STZ’ (2015) the STZ hospitals strive to the implement the principle of VBHC broadly across all aspects of care. Moreover, MCL is part of the ‘Friese Zorglandschap’ that aims to preserve medical specialist care and secure future accessibility and high quality of care for citizens. In 2018, MCL started the ‘Financieel Gezond’ program aiming to bring down and control its entire hospital budget. This program emphasizes the need to regain the status of a financially healthy organization in the year 2025. Within this context, this research aims to investigate how Patient Reported Outcome Measures (PROMs) can be successfully integrated within the pulmonary oncology unit. To research this, two questions are created:

1. To what extent is the pulmonary cancer unit ready to integrate PROM indicators in daily care for non-operable stadium III and stadium IV lung cancer patients?

2. Which interventions can enhance organizational readiness for the unit's use of PROMs?

In order to answer the research questions, this thesis is outlined as follows. First, literature on VBHC, Patient Reported Outcomes Measures and change readiness is assessed. After that the used methodology is explained. The research follows the regulative cycle by van Strien (van Strien, 1997) to describe the case, diagnose the readiness of the pulmonary unit towards PROM integration and design interventions. To assess readiness semi-structured interviews are held at the pulmonary unit.

Moreover, the results of the diagnosis are set out and analyzed. Based on the results, researcher provides interventions for the unit and the hospital as an organization to enhance organizational readiness toward PROM integration and, moreover, help to integrate and actively use the PROM questionnaires. Furthermore, the transferability of the findings will discuss the extent to which these findings can be applicable for other contexts and which prerequisites must be taken into account in PROM integration.

(7)

7

Chapter 2 Literature

In the following chapter the theoretical lens of this study will be further explained in order to fully comprehend the concept of value based health care and, moreover, how it is defined as an organizational change within the organizational context of MCL. When implementing any change three aspects are considered important (Cawsey, 2016). For determining a future direction it is a prerequisite to first diagnose where an organization currently stands. The answer to the question “why change?”

then defines the desired future state. The second aspect, the “what?”, should determine the content that has to change. Lastly, the “how?” aspect of change is important to lead the process of organizational change. This section will, firstly, elaborate on the ‘what’ aspect of this change by emphasizing the four quadrant framework of value based health care (Ahaus, 2018). Thereafter, the section will elaborate further on the ‘how’ to change by emphasizing five interactive elements that are critical in the transformation process of improving patient care (Lukas, 2007). We use the model of Lukas and colleagues (2007) as a guiding model for conceptualizing the ‘how’. Finally, the concept of change readiness is introduced as of its importance before any change are implemented. The readiness model by Holt et al. (2007) is used to assess change readiness on team level at the pulmonary oncology unit as well as to explore readiness for change on the organizational level.

2.1 Towards Value Based Health Care

According to Porter and Lee (2013) a transformation towards Value Based Health Care is not one single step, but an overarching strategy. The transformation process, which Porter and Teisberg (2006) call ‘the value agenda’, entails a restructuring of the way health care delivery is organized, measured, and reimbursed (porter, 2013). This means that shifting towards Value Based Health Care would entail an organizational change for MCL as an organization. Organizational change is referred to as ‘planned alterations of organizational components to improve the effectiveness of the

organization’ (Cawsey et al., 2016). As mentioned earlier, ‘what’ to change should be defined, before the ‘how’ process starts.

In line with Porter’s (2013) ideas, Ahaus (2018) introduced the four quadrant framework of value based health care. In this framework he addresses the importance of four different areas in successfully performing value based health care: The first quadrant involves (1) the value for the patient. Value can be defined either by objective measures that provide information about clinical status (i.e. 30-day survival rates) or indicators that provide information about patients’ functional status (Patient Reported Outcome measures, PROMs). Latter provide unique information about the wellbeing of the patient (Ahaus, 2018). Ahaus (2018) addresses the importance of PROMs to be part of the conversation between physician and patient, as it facilitates shared decision making between

(8)

8 patient and physician and incites honest expectations on the patients’ side. The second quadrant of the framework is (2) costs (Ahaus, 2018). Ahaus (2018) claims the importance of cost reduction combined with increasing or equal outcomes. To obtain cost reduction he mentions that care payments should introduce financial incentives that stimulate value instead of volume. The next quadrant (3) ‘organization of care’ argues that multidisciplinary teams of different care professionals should intensively collaborate (Ahaus, 2018). The team, mostly led by a driven physician, must collectively take responsibility for costs and outcomes of the full cycle of care and improve care based on collected data (Porter & Lee, 2013). The end goal is personalized care in order to fit the needs of the patients. The last and fourth quadrant of Ahaus’ (2018) framework is (4) steering.

Patients do not always receive patient care as described according to evidence based practices described in guidelines (Seys, 2017). Therefore, next to outcome- and cost indicators, process indicators could be a valuable addition to value based health care (Ahaus, 2018). Process indicators measure care activities instead of outcomes, for example determining whether a COPD-patient still smokes tobacco (Seys, 2017). Steering towards better care quality holds measuring, monitoring, benchmarking and improving based on standardized indicators (Ahaus, 2018) and eventually learn and improve. To visualize performance Ahaus (2018) mentions that a dashboard could be effective.

All four quadrants should be taken into account in the shift towards VBHC. However, due to both time restraints and the fact that the pulmonary oncology unit started an experimental project with PROMs, this study lies focus on the first quadrant ‘Value for the patient’ of Ahaus’ (2018) framework. The research involves the integration of Patient Reported Outcome Measurements (PROMs) at the pulmonary unit. The experimental project (PROTECT, see Box 1) measures the quality of life in the treatment process of non-operable lung cancer patients. Currently, 33 patients

voluntarily agreed to participate in PROTECT and fill in the PROM questionnaires. In the following subsection PROMs are described and elaborated on in terms of their importance, usability and accompanying challenges.

2.1.1 What: Patient Reported Outcome Measurements in palliative care

PROMs are instruments, designed for patients, in the form of questionnaires that measure value from a patient perspective and, hence, give insight in the efficiency of the delivered care (Werken aan zorgkwaliteit, 2017). Over the past few years, PROMs gained popularity in the research field and still interest for routine use of PROMs in daily care practice is increasing (van de Glind, Bakker-Jacobs, Triemstra, de Boer, & van der Wees, 2018). PROMs might even contribute to the paradigm shift towards patient-centered care and improve patients’ objective outcomes (Bouazza, Chiairi, Kharbouchi, De Backer, Van houtte, Janssens & Van Meerbeeck, 2017). In the Netherlands, many health related institutes already measure these outcomes, but the actual use of PROMs varies (van

(9)

9 de Glind et al., 2018). It is argued that more research regarding their benefits and relevance should be performed. Former is important to avoid proliferation and collect useful and valid information for the intended purposes. Especially in treatment settings with a generally poor prognosis or rapid course of disease the use of PROMs gain more importance, as other factors than ( progression-free) survival become essential (Schulkens, Nguyen, van den Bos, Hamaker, & van Elden, 2016). An

example is lung cancer, which is often diagnosed in an advanced stage (Bouazza, 2017) and has a low survival rate. Moreover, its symptoms impact quality of life massively (Corner, Hopkinson,

Fitzsimmons, Barclay & Muers, 2005). This makes PROMs relevant for lung cancer treatment (Temel, Greer, Muzikansky, Gallagher, Admane, Jackson & Billings, 2010).

In the literature four main aims of PROMs are distinguished. PROMs can be used:

(1) in the individual patient care. For example used in the consultation room for screenings and diagnostics, for shared-decision making regarding treatment goals, for multidisciplinary meetings, monitoring health outcomes, and evaluating care and treatment plan;

(2) for quality improvements on organizational- and unit- level;

(3) for public information and transparency to increase information regarding treatment options.

(4) in scientific research and as policy information.

It is found that PROMs are often used for multiple aims, even on different levels (van de Glind et al., 2018). Striving toward a combination of multiple aims is, however, increasing the complexity, as each aim has different requirements regarding the form and type of the questionnaire, measuring moments, logistic process and analysis of the outcomes. In their article Greenhalgh, Dalkin, Gooding, Gibbons, Wright, Meads, et al. (2017), elaborate on different types of PROM questionnaires. The first type are standardized PROMs which are fixed questionnaires that are either general (for diverse populations) or specific (disease focused). The second type are individual PROMs. These function to provide an overview of individual problems that are experienced, and defined, by the patient.

Individual PROMs offer more opportunities to discuss individual problems. In oncological care, standardized PROMs can obstruct the patient physician conversation, as it is too nonspecific. On the other side, a disadvantage of individualized PROMs is that they are less suitable for monitoring change and, hence, less applicable for quality improvements and transparency (Greenhalgh et al., 2017). Moreover, they emphasize that contextual factors in which PROMs are used play a role in the potential impact of PROMs. This makes it difficult to draw any conclusions about causal relationships.

Furthermore, perceived value of PROMs on aggregated level aiming for quality improvements and transparency is dependent on the trust in the data and its presentation. Care professionals are more willing to undertake action regarding quality improvements when (1) PROM data is reliable, (2) PROM-feedback is given timely and continuously and indicates potential improvements (Greenhalgh et al., 2017). Professionals value PROMs when they are useful within the process of clinical decision

(10)

10 making. Table 1 (see Appendix A) presents examples of contextual factors which can influence the working mechanism of PROMs.

Hutchings and Alrubaiy (2017) found seven practical challenges that may limit widespread PROM integration in clinical practice, namely, they: (1) are not robustly validated; (2) may have limited generalizability; (3) may not be sensitive enough to detect changes in a patient’s condition;

(4) are long and require excessive time to complete; (5) use terms that are difficult to understand for patients; (6) have detailed scoring systems complicating application to short clinical consultation and (7) collecting information through PROMs presupposes that patients can complete these

questionnaires. Hence, completing the questionnaires is not always feasible for patients, as they could be frail, have limited intellectual understanding, or might be too sick to complete the questionnaires (Hutchings & Alrubaiy, 2017). Lavallee, Chenok, Love, Petersen, Holve, Segal, &

Franklin, (2016) argue for a more user-centered design. For example, including patient’s impairments, such as low literacy, in the design phase.

PROM integration is challenging and complex. As such, the integration process of ‘how’ to integrate these indicators is important. The next section will describe the organizational model of Lukas and colleagues (2007) that guides the move of organizations in the transformation of improvements in patient care.

2.1.2 How – the process

Organizational change is a complex phenomenon. According to Mintzberg (1979) health care organizations are considered professional bureaucratic organizations whose structures and cultures have often been highly resistant to change, especially changes that derive from policy or managerial directions (Baker & Denis, 2011). IOM (2001) claimed that intensive efforts at all organizational levels are needed to fundamentally redesign systems of care. Though, few organizations have succeeded in making a substantial transformation (Corrigan, 2005). Many quality improvement efforts did not lead to sustained system change, due to lack of support in the organizational culture and the structure of the organization (Repenning & Sterman, 2001; Rondeau & Wagar, 2002). Transformational change involves not only structures and processes, but also the organizational culture and values of the health care organization (Lukas, Holmes, Cohen, Restuccia, Cramer, Shwartz, & Charns, 2007). To guide health care organizations Lukas and colleagues (2007) developed a conceptual model to guide healthcare organizations towards sustained, evidence-based improvements which will eventually lead to patient care transformation across the organization. Five interactive elements are identified that appear critical to this successful transformation: (1) Impetus to transform;

(2) Leadership commitment to quality; (3) Improvement initiatives that actively engage staff in meaningful problem solving; (4) Alignment to achieve consistency of organization-wide goals with

(11)

11 resource allocation and actions at all levels of the organization; and (5) Integration to bridge

traditional intra-organizational boundaries between individual components (Lukas et al., 2007).

These five elements drive change by influencing the following components of a health care organization: (1) mission, vision and strategies; (2) organizational culture that reflects its informal values and norms; (3) operational functions and processes; (4) infrastructure which supports the delivery of care (Lukas et al., 2007). It was found that impetus to transform can differ in terms of source or nature. However, the impetus has to be sustained within the organization to motivate and engage staff members in the change.

According to Lewin (1947), organizational change is happening through people working in teams or work groups. Hence, the functioning of an organization depends on action and behavior changes of its employees (Goodman & Dean, 1982). This means that an effective change effort is dependent of the organizations’ ability to influence behavior of individual employees (Robertson et al., 1993). During an individual’s progression through change three stages of unfreezing, moving and refreezing respectively are experienced (Lewin, 1951; Holt, Armenakis, Feild & Harris, 2007). As noted by Armenakis and colleagues (1993) the ‘unfreezing’ stage of Lewin (1951) coincides with readiness for change. They argued that readiness was a precursor of resistance and adoption behaviors Armenakis, Harris, and Mossholder (1993). Furthermore, it was found that readiness for change may act to pre-empt the likelihood of resistance to change, increasing the potential for change efforts to be more effective (Armenakis et al., 1993). Despite awareness of the likability of resistance to change, managers often neglect opportunities to take time and determine whether, and for what reasons, staff members might resist (Kotter & Schlesinger, 1979). Therefore, the next section will elaborate on the importance of change readiness in a change context.

2.2 Change readiness

Change readiness is argued to be one of the most crucial factors involved in employees’ initial

support for change initiatives (Armenakis et al., 1993; Armenakis, Harris, & Feild, 1999). High levels of change readiness enhance positive effects of both effective leadership and support for the new strategy (Caldwell et al., 2008). Moreover, evidence suggests a positive relationship between people’s support for a new strategy and the effectiveness of the implementation (Caldwell et al., 2008). Hence, it is an essential factor for change success in healthcare settings (Hardison, 1998;

Weiner, 2009). Change readiness is defined as “organizational members’ beliefs, attitudes, and intentions regarding the extent to which changes are needed and the organization’s capacity to successfully make those changes” (Armenakis, Harris and Mossholder, 1993, p. 681). However, Bernerth (2004) claims that few conceptualizations of change readiness take into account all

interrelated aspects that influence change readiness. Bernerth (2004) argues that besides beliefs and

(12)

12 attitudes, also the social environment may shape change readiness. Therefore, he emphasizes the importance of modifying this social environment, as individuals can be influenced by each other. Also Amenakis and colleagues (1993) argue that existing organizational conditions affect staff members’

loyalty, commitment and other feelings toward the organization and management and, in turn, influence the level of readiness (Armenakis, 1993). Employee participation was found to play an important role in reducing resistance and readying employees for change (Coch & French, 1948; Holt et al., 2007). However, other evidence was found that units may vary in terms of their capabilities to implement changes, regardless of whether these changes are widely supported or not (Caldwell et al., 2008). This assumes that units may differ in their overall orientation toward change or in the general skills and competencies they need to implement new processes (Caldwell, Chatman, O'Reilly III, Ormiston, & Lapiz, 2008). Next to the differences in overall orientation toward change and capabilities to implement changes, Anderson and West (1998) found that work groups sharing a norm of valuing innovation were more likely to change than those that did not. Furthermore, Caldwell and O’Reilly (2003) found specific norms for change readiness that affect the ability of a group to implement change. These specific norms are developed around the areas of support for risk taking, tolerance of mistakes, teamwork, and speed of action. Similar studies of change readiness corroborated these findings (e.g., Armenakis & Harris, 1993; Eby, Adams, Russell, & Gaby, 2000;

Oreg, 2003).

Unit-managers may influence the speed and effectiveness with which new processes can be implemented (House & Aditya, 1997). When they are involved in the process and committed to the new strategy, success of this strategy is more likely (Wooldridge & Floyd, 1990). Also Lukas and colleagues (2007) found that improvement was greater when middle and frontline managers were committed to quality, being actively involved in supporting process redesign. However, managers may also delay the implementation process (Guth & MacMillan, 1986). Therefore, middle-managers have an important role in the implementation of strategic changes.

To assess change readiness several instruments have been developed (Holt et al., 2007; Jones, Jimmieson, & Griffiths, 2005). Holt and colleagues (2007) suggested a readiness model in which readiness for change is influenced by four factors namely: change process; change content; change context and; individual attributes (see Figure 1). These four factors guide the development of a readiness measure, suggesting a general set of beliefs shape readiness for change and, thus, provide the foundation for resistance or adoptive behaviors. They defined change readiness as an attitude that is being influenced simultaneously by what is changed (i.e. the content), how the change is implemented (i.e. the process), the circumstances under which the change is occurring (i.e. the context), and the characteristics of those being asked to change (i.e. the individuals). On top, Holt and colleagues (2007) argue that readiness collectively reflects the extent to which individuals are

(13)

13 cognitively and emotionally inclined to accept, embrace and adopt a particular plan. These four factors are aligned with issues that also appear to affect individual readiness. The first is (1) appropriateness, (i.e. is the proposed change appropriate for the organization); (2) management support regarding the change initiative (i.e. are leaders committed to the proposed change); (3) self- efficacy of the employees (i.e. believes of employees whether they are capable of implementing proposed change); and (4) personal valence (i.e.is the proposed change perceived as beneficial to the employees) (Holt et al., 2007). Wanberg and Banas (2000) suggested that staff members’ (5) general attitudes toward change and the two facets of an organization’s culture: (7) perceptions of the communication climate (i.e.how individuals at the unit interact with each other within their relationships); and (8) perceptions of management’s ability (i.e. managements’ skills) could be expected to correlate with readiness for change factors. Additionally, (9) employee participation (i.e.

involvement of employees in the change) is believed to increase change acceptance (Coch & French, 1948; Holt et al., 2007). Direct participation, for example through planning and implementing, tends to create affectively committed employees. Moreover, it increases the access to information which creates a better understanding of the justification for change and its objectives. Figure 5 presents the factors of the conceptual model of Holt and colleagues (2007) integrated with the aligned issues which are found to affect change readiness.

(14)

14 2.3 What to change for PROM integration

After determining the ‘what’ and the ‘how’ it is fundamental to understand what type of change, firstly, the shift towards Value Based Health Care and, secondly, the integration of PROM indicators entails.

Burnes (2017) created a framework (Figure 2) with two change continuums which distinguish four quadrants. According to this framework, the shift towards VBHC needs changes in all quadrants, as a VBHC strategy entails a fundamental redesign of the system of care that needs intensive efforts at all organizational levels (Corrigan, 2005). First of all, this multiple level change needs change on organizational level in structures and processes, as they need to be redesigned towards more patient centered, multidisciplinary teams (Q2). Moreover, staff members’ tasks and procedures are changing when shifting to VBHC, as process-, cost-, and outcome indicators have to be implemented (Q3). This, in turn, as an effect on the job tasks of care professionals. Next to structures and tasks,

transformational change involves organizational culture and values of the health care organization (Corrigan, 2005). For example, shared decision making and a continuous learning culture are two important aspects of VBHC that involve both culture and individual attitudes/behavior. To sum, a shift towards VBHC fits all quadrants of Burnes’ (2017) framework which also implies the complexity of change approaches while shifting to VBHC.

As this research focuses on the quadrant ‘Value for the patient’ of the framework of Ahaus (2018) we now place PROM integration as a change in the change framework of Burnes (2017). First, the change concerning the integration of PROMs on the pulmonary oncology unit only involves the individual- and unit level. First, focus will lie at attitudes and behaviors of the units’ staff members (Q4). These attitudes should be open and positive towards PROM integration. To enhance positive attitudes and readiness toward PROMs staff members should be involved in the process of

integration and, moreover, the change initiators should provide insight into reasons to collect data through PROMs (Greenhalgh et al., 2017). This will heighten the need for change and address the appropriateness. Furthermore, the confidence of the staff members can be bolstered so they believe they can accomplish this change (Cawsey, 2016). Second, PROM integration affects the individual procedures and job tasks of staff members at the unit. For example, the measuring and guiding of patients will lead to changes or additional job tasks for the staff members that will have to perform these extra tasks. Also a staff member needs to analyze the PROM outcomes in order to make it use the outcomes. When PROM are used on individual patient level in the consultation room nurses and physicians have to adjust the procedure of their patient conversations. Lastly, next to the attitudes and job tasks, processes need to be adjusted, as integration of PROMs in the Electronic Health Record (EHR) is marked as an important prerequisite. Therefore, the PROM outcomes have to be

(15)

15 integrated in EPIC, the EHR system which the MCL is using for four years now.

This research aims to provide the pulmonary oncology unit with advise to integrate PROM indicators in the daily care path way of non-operable lung cancer patients. The study diagnosis the readiness for change for the integration and use of PROMs on unit level. After, the researcher designs

interventions that the unit can undertake to enhance readiness and successfully integrate PROMs, thereby, contributing to the transformation towards VBHC. The following questions are developed to answer the research question appropriately:

“How can MCL pulmonary unit successfully integrate PROM indicators in the daily care for non- operable stadium III and stadium IV lung cancer patients?”

1. To what extent is the pulmonary cancer unit ready to integrate PROM indicators in daily care for non-operable stadium III and stadium IV lung cancer patients?

2. Which interventions can enhance organizational readiness for the unit's use of PROMs?

Chapter 3 Method

This chapter describes the research design and explains the steps which were taken to answer the research questions. This study had an academic problem solving approach to answer the question

“How can MCL pulmonary unit successfully integrate PROM indicators in daily care for non-operable stadium III and stadium IV lung cancer patients?” In the first sub question the readiness to integrate the PROM indicators in the daily care at the MCL pulmonary oncology unit was assessed. The second research sub-question involved the design of interventions on unit- and organizational level required to enhance the process of successful integration of PROMs. The regulative cycle of Van Strien (1997) was used as a guiding framework to diagnose, design and evaluate the case unit. The following section firstly explains the research context and case selection, second, the research design and case selection

(16)

16 which is followed by the data collection and the data analysis. In the subsections of data collection and data analysis the methods used will be discussed per question.

3.1 Research Context and Case Selection

This research was requested by and conducted at the Medical Center Leeuwarden (MCL), one of the 28 Dutch top clinical hospitals. MCL is located in the province of Friesland in the Northern part of the Netherlands. Being a top clinical hospital, MCL is part of the association of Samenwerkende

Topklinkische opleidings Ziekenhuizen (STZ) hospitals and focuses on training, education and

scientific research. Currently, the MCL is under strict financial supervision, as the hospital must bend approximately 21 million on its annual budget by 2021. To cut costs, a program ‘Financieel Gezond’

started in 2018 in which seven focus areas are selected. In line with the cost saving aim of ‘Financieel Gezond’, MCL put through an organizational restructuring in October 2019. The existing 27

independent units merged into six larger care-units (CU’s). Figure 3 (see Appendix B) shows the new organizational structure. The current six CU’s are: acute care, oncological care, surgical care,

diagnostic care, cardio-vascular care and woman-mother-child care. The newly designed CU’s lead to less waste due to shorter lines and less ambiguity between different disciplines. Former is achieved, as more care professionals are involved in the patients care process directly from the start.

Therefore, this new structure facilitates a quicker and more intense collaboration between different disciplines (Albers, 2019). Each CU is led by two managers, a Medical m and less ambiguity Manager (MM) and a Care manager (OM). The pulmonary unit is for example placed under CU diagnostic care but also partly operates under CU oncological care. This organizational restructuring is a start toward VBHC, as the care path is more centered around the patient. It increases value for patients and simultaneously decreases waste and, hence, costs. The new structure is an important context factor for this study. The restructuring caused a lot of insecurities and uncertainties. Some CU’s, for example, missed steering for the past months which led to uncertainties among some departments.

Moreover, the management layer of the departments have yet to be reorganized. Hence, in answering the research question these context factors are taken into account.

The pulmonary oncology unit was chosen as case unit for multiple reasons. First of all, the pulmonary oncology unit started an experimental project (PROTECT) regarding PROM indicators (Box 1). This made the unit a case of interest (Stake, 1995), as the project could be used to asses change readiness and function as an example for designing interventions for enhancing change readiness on unit- and organization level. Moreover, the pulmonary unit is an innovative unit. Multiple lung oncologists are considered innovators and are responsible for successful innovations within the hospital. The high percentage of innovators at the unit will facilitate in the change of integrating the PROMs. Furthermore, the pulmonary oncology has a high level of performance which is confirmed by

(17)

17 the highest adoption level of MCL’s new Electronic Patient Database system (EPIC) in comparison to all other units. This high level of performance is also a factor that facilitates this change. Moreover, when this high performing unit can successfully integrate PROMs, the unit can be compared to other units within the hospital and function as an example. In conclusion, the PROTECT project, the

innovative nature and high performance of the pulmonary oncology unit made it a suitable case unit for this study.

3.2 Research design

To answer the research question a design oriented, qualitative case study was conducted. A qualitative case study methodology is preferred when ‘why’ or ‘how’ questions are being posed (Yin, 1989). It is often used to study real-life complex phenomena in-depth within a certain context and it is a valuable method for developing interventions (Baxter & Jack, 2008). In this study at the pulmonary oncology unit the change readiness of staff members regarding the integration process of PROM indicators is assessed. This is followed by interventions designed by the researcher to enhance organizational readiness for PROM integration. The regulative cycle of Van Strien (1997) was used to describe, diagnose, design, plan and evaluate the case (Figure 4).

(18)

18 The regulative research cycle lends itself perfectly for design-oriented research, as the cycle is a structured organizational problem solving process (Heusinkveld and Reijers, 2009). Moreover, the process provides context specific output which is important within this study (Heusinkveld and Reijers, 2009). The iterative regulative cycle (1997) consists of five steps: (1) problem definition; (2) diagnosis; (3) design; (4) implementation (change); (5) evaluation. Due to time constraints within this thesis, we focused on the problem description, diagnosis and design of interventions. The

interventions designed were not implemented, but involve a proposal for the PROTECT initiators at the pulmonary unit and for management of the organization. In the fifth evaluation step the transferability of PROMs was evaluated based on literature and emerged issues. The following subsections will describe the data collection and analysis.

3.3 Data collection

The use of diverse sources of data is critical in case studies. It ensures that a thorough description of the case, its origins, its development and changes over time can be made (Small, 1995). To collect information for this study the researcher used three types of qualitative sources which included archival documents, semi-structured interviews and direct observations. First, archival documents and informal conversations contributed to a better understanding of the organization and its context. After the case selection was made, direct observations and specific documents contributed to the researchers’ knowledge about the research site’s context (see Table 1). To elaborate further on the data collection the researcher will distinguish between the sub-questions.

(19)

19 3.3.1 Data collection: Diagnosing readiness

This subsection describes the data collection of the first sub-question: “To what extent is the pulmonary cancer unit ready to integrate PROM indicators in daily care for non-operable stadium III and stadium IV lung cancer patients?”

First, general documents concerned the STZ strategy document, MCL’s framework paper of 2019 which included its vision and projects on the agenda, a document with improvement projects, evaluation documents and information documents regarding ‘Financieel Gezond’. These documents were provided by a program manager in the orientation phase of the research. Through these documents the researcher gained knowledge about MCL as an organization. Moreover, newspaper and online articles about MCL published in the Leeuwarder Courant were gathered and read. Next to this a PowerPoint with information about the CU restructuring was looked at. On top, direct

observations improved understanding of the unit’s context. The researcher was allowed direct observations at three outpatient consultations with two pulmonary oncologists. Beforehand, a confidentiality statement had to be signed in order to ensure anonymity of the patients. In total, the researcher observed for 12 hours; each consultation lasted 3 hours. Field notes of the observations were taken. These observations were performed to contribute to the researchers knowledge of (1) the organizational culture at the pulmonary oncology unit, (2) the communication between physician and patients and (3) to see if and how quality of life is discussed.

Semi-structured interviews were used as a data collection method within this study aiming to capture the perceptions and experiences of these staff members regarding PROM integration (Crouch & McKenzie, 2006). According to Cooper and Schindler (2008) semi-structured interviews start with a few specific questions and then follow the individual’s tangents of thought with

(20)

20 Figure 5 Resulting data structure for sub-question 1

interviewer probes. A total of 13 interviews were conducted. Beforehand an interview protocol was designed and structured based on relevant sub-topics that covered the aim of this study (see appendix C and D). Afore the interviews started, participants were required to fill in and sign the Informed Consent, which was translated into a Dutch version, to formally agree upon participation.

The interviews were recorded on an IPad (digital audio device) and transcribed by the researcher.

The transcriptions served for the analysis of the interviews. Afterwards, the interviewees were given the opportunity to review their transcript. The first interview protocol was used among the first two interviewees, from the third interview on some questions were redesigned, as their formulation lacked clarity (Gioia, Corley, & Hamilton, 2013). Moreover, some questions were added or removed, as more specific information was needed that the original protocol did not cover.

Based on organizational change literature and the model of Holt et al. (2007), a data structure was created (Figure 5, Appendix E). This structure was enriched with issues that emerged from the interviews.

Change Readiness

Change Content Change Process Change Context Individual Attributes

Appropriateness

Staff member participation Management

support

General attitude toward change

Organizational restructuring

Innovative nature organization

Motivation

Usability of PROM outcomes

Communication

Resource availability

Self-efficacy

Lack of clarity task division

Previous change experience

Commitment

Personal- valence Perceived ability

management

Content questionnaires

High workload

Note: red boxes are change specific factors for PROM integration within the unit Level of analysis: hospital unit

Need for change

Digital integration

Financial challenge Bottom-up or

top-down

(21)

21 3.3.2 Participants

All 13 interviews were conducted in person. The initiator of the PROTECT project helped the researcher with identification of participants. After identification the researcher contacted all participants and planned the interviews. The selection was made based on the logic that different positions within and directly linked to the pulmonary oncology unit enrich the collected data. Hence, this contributes to more insights and a completer overview of the problem. The following staff members were interviewed: a CU manager, a team-leader, 5 pulmonologists, 3 pulmonary nurse consultants (VLCO’s), a secretary and 2 research nurses. In this thesis the staff members were referred to by their function and an added number based on the order of the interviews. Table 4 shows the conducted interviews including gender of interviewees, function and used abbreviation.

3.3.3 Data Collection: Designing change Interventions

This subsection describes data collection of the second sub question: “What interventions can enhance organizational readiness for the unit's use of PROMs?”

Data from both interviews, as described earlier, and literature were used. The middle part of the interview (questions 5 until 12) focused on the interviewees’ knowledge about the PROTECT including their ideas about its integration and usability of PROM outcomes. Moreover, questions about participants’ experiences regarding success and fail factors of change within the organization provided useful data for designing the interventions. Especially (former) managers provided specific ideas which researcher used in stage two of the regulative cycle: designing interventions. Besides data obtained from the interviews, literature about readying an organization and creating

organizational change, such as Cawsey (2016) and Armenakis and colleagues (1993), were used to Table 4 Overview and details of conducted interviews

Function No. (N) Sex Abbreviation

Pulmonologist N=5 M N=4 F N=1

PM

Research nurse N=2 M N=0 F N=2

RN

Pulmonary nurse consultant

N=3 M N=0 F N=3

VCLO

Manager N=2 M N=1

F N=1

Manager

Secretary N=1 M N=0

F N=1

SEC

(22)

22 gain knowledge about change readiness interventions. More specific literature about PROM

integration helped to gain knowledge about prerequisites and contextual factors that influence this integration.

3.3.4 Data saturation

In the diagnostic phase, data saturation was reached, when the pre-determined factors to assess change readiness were adequately represented in the data (Fusch and Ness, 2015). Moreover, additional interview data did not lead to any new emergent themes or codes regarding the PROM integration at the unit (Given, 2015). Each interview obtained rich information and the total number of 13 interviewees represented all key staff members and positions of the pulmonary oncology regarding this change. However, albeit similar readiness themes are seen over and over again in the staff members’ interviews the diversity of the data could be stretched by interviewing more

managers outside the pulmonary unit. This could have provided richer information about the change.

Due to time constraints we did not interview more managers or a hospital board member.

3.4 Data Analysis

3.4.1 Analysis: Readiness

The analysis of sub question one “To what extent is the pulmonary cancer unit ready to integrate PROM indicators in daily care for non-operable stadium III and stadium IV lung cancer patients?”

started with reading documents, field notes taken from observations and both formal and informal conversations with employees. Second, the results of the 13 semi-structured interviews were analyzed. Transcripts were screened and read thoroughly. This process was done several times to discover underlying and deeper meanings. Coding was used to organize and interpret the data from the transcripts. For coding and analyzing all this textual data ATLAS.ti was used and in total 35 codes were created. Strauss and Corbin (1990) define coding as conceptualizing, elaborating, reducing and relating obtained data. The results were visualized in a codebook, a formal system to organize the data obtained (Bradley, Curry & Devers, 2007). The codebook in this study is separated into deductive (top-down) and inductive (bottom-up) coding approaches. The deductive codes were based on factors regarding organizational and individual change readiness literature. Figure 5 presents the data structure used to assess the change readiness. Deductive codes were applied to the transcripts. Afterwards, an inductive coding process allowed new concepts to emerge from the data besides the concepts considered important according to the literature. After the analysis of the interview codes, findings were compared to the literature to analyze the level of change readiness on the unit.

(23)

23 3.4.2 Analysis: Designing interventions

The diagnosis of the readiness on the pulmonary unit, based on the data structure, was analyzed and compared to readiness literature. Based on this comparison, a forces figure was made which

provided an overview of the driving and restraining forces of PROM integration. The arrow sizes representing the strength of the force at work were based on the relative salience of that force in the interview data. The Next, based on this diagnosis and with the help of organization change literature the researcher developed an intervention plan which fits the pulmonary unit and the organizational context to enhance organizational readiness.

Chapter 4 Results diagnosis change readiness (sub-question 1)

In this chapter the results of the read documents, observations and interviews held within the pulmonary unit are summarized and described. Using these results a diagnosis can be made of the change readiness of employees of the pulmonary unit. The change readiness issues are categorized in the four factors change content, change process, change context and individual (Holt et al., 2007).

First, a short summary of the interpretations of staff members’ views on quality of life and patient value will be given. Next the four factors with aligned items are described one by one. The items include: appropriateness, management support, participation, communication climate, perceived management ability, organizational restructuring, self-efficacy, commitment, personal valence and general attitude toward change. Due to the exploratory nature of this research the answers of interviewees are described instead of presented in a table. This would force the researcher to generalize the answers and, hence, lose valuable information.

4.1 Change Content: use of PROMs in palliative care

This section focuses on experiences and thoughts that staff members of the pulmonary unit had regarding the change initiative and its characteristics. For the change initiative regarding integration of PROM indicators the experimental project PROTECT was used as an example to gain information from interviewees. In the interviews staff members shared thoughts about using outcomes of PROM indicators, how the process should be designed (i.e. measuring and analyzing). In general,

interviewees were optimistic regarding the integration of PROMs, but acknowledged some difficulties.

(24)

24 Quality of life. At the start, all interviewees were asked about their vision on quality of life for non- operable lung cancer patients. All responded that that is up to patients to decide. The quotation of RN2 covers opinions perfectly: “If you are dealing with quality of life it is in my opinion very important to see what the patients want. What do they find important themselves? That is quality and that is not for someone else to determine.” To the question if everybody on the unit deals with quality of life in a sufficient way, VCLO3 answered: “Oh, I don’t know actually. No, I work from my own vision.”

More interviewees answered this question with “I try to give it enough attention, though I don’t know about my colleagues.” All interviewees found quality of life an important topic, but most did not talk about it with their colleagues.

Need for change. The perceived need for this specific change is quite low, as there is no real urgency noticeable among the interviewees. However, Manager 2 and PM4 argued that lung cancer is becoming more like a chronic disease as treatments keep improving. Therefore, people live longer, but the question whether these patients also perceive their quality as relatively good remains. The longer timespan also means that (pulmonary) oncological nurse consultants follow these patients for a longer period of time, which in turn creates opportunities to help patients by using PROMs.

Furthermore, some interviewees showed dissatisfaction with some current situations within the organization. PM1: “If you would ask patients how they experienced it here? I think you’ll be shocked.

That they didn’t like it so much here. Well, the nature of being here is not too nice in the first place.

But if certain things within the hospital are not organized that well..”

Appropriateness. All 13 interviewees clearly expressed that they believe integrating PROM indicators is an appropriate and beneficial change in the long-term. Many interviewees mention that measuring quality of life should play a bigger role in research. PM4: “Most of the time research focuses at primary outcomes, life extension, but very little at quality of life.” SEC: “I believe that this is a very important topic. I think that it creates a clear overview of how people experience treatments. That has never been done here really. And it is a very meaningful study.” Staff members expressed that this change is particularly appropriate, because of the nature of the unit: because of the limited time that these patients have left, quality of life becomes extra important. Despite the positive remarks about this change, efforts which have invested to integrate PROMs have not yet succeeded. Manager 1 said: “I think we still live in the Stone Age regarding outcome measurements. The past couple of years we have been undertaking lots of efforts.”

Usability of the outcomes PROM questionnaires. This section describes how interviewees view the use of the outcomes of PROM questionnaires. Their arguments on this topic varied. Table 5 provides an overview of their opinions. Next to their individual ideas, Manager 1 addressed his concerns