Master thesis, MSC Supply Chain Management
University of Groningen, Faculty of Economics and Business
May 8, 2017
Hendrik Dekker – 2011611
h.w.dekker.1@student.rug.nl
Supervisor: Prof. Dr. J. Wijngaard
Co-assessor: Dr. S.A. de Blok
Supervisors Medical Centre Leeuwarden:
J. Marra
B. Jeeninga
The impact of shared resources on the
integration of patient planning in
1
ABSTRACT
Hospital professionals find it difficult to integrate planning activities between departments. The results are that waiting times for patients reach undesirable levels and that throughput targets of hospital departments are not met. Literature on this matter has been focussed on addressing these issues in a low complexity care process (i.e. the orthopaedics department) identifying operational antecedents and their effect on integrating planning activities within hospitals. Therefore, the aim of this paper is to expand on existing knowledge by addressing these findings in a complex care process (i.e. the oncology department). The study was conducted by means of a single case study at a hospital in the Netherlands in which the findings stem from 12 conducted interviews with hospital professionals active in the mamma care pathway. The research contributes to theory by the identification of newly discovered operational antecedents and knowledge on how integration can be achieved in a more complex setting. These insights can be valuable to hospital professionals who operate within complex care processes and explain the potential impact these antecedents can have when wanting to integrate planning activities between functional departments.
2
Content
Chapter 1. Introduction ... 3
Chapter 2. Theoretical background ... 5
2.1 Integration of patient planning in hospitals ... 5
2.2 Shared resources in healthcare ... 7
2.3 Shared resources in operations management ... 9
2.4 Operational antecedents of integration ... 10
2.5 Conceptual model ... 13
Chapter 3. Methodology ... 14
3.1 Research method and case selection ... 14
3.2 Data collection/Data sources ... 15
3.3 Data coding and analysis ... 16
Chapter 4. Results ... 17
4.1 Integrative practices: within case analysis ... 17
4.2 Antecedents of integration: within case analysis ... 0
Chapter 5. Discussion ... 7
5.1 The impact of operational antecedents ... 8
3
Chapter 1. Introduction
Treating patients requires hospitals to coordinate process steps within care pathways to limit the waiting times for patients. This is especially important for patients were waiting time is detrimental to the health outcomes of patients. Integrative practices, for instance combining appointments, are therefore valuable in tackling waiting times by means of integrating process steps (Aronsson et al., 2011). However, due to conflicting preferences within hospitals such as the allocation of limited resource capacity, processes have to unite many stakeholders (such as surgeons, nurses, patients and hospital management), which could make coordinating practices difficult to manage (Cardoen et al., 2010).
Research has been conducted regarding the integration of patient planning in hospitals were the integral planning of activities leads to an increase in hospital performance (Drupsteen, Van der Vaart & Van Donk, 2013). However, health care systems are slow in embracing these integrative planning practices (Sampson et al., 2015; Cardoen et al., 2010). An exploratory research has been conducted by Drupsteen et al. (2016) which uncovered several reasons why hospitals find it difficult to integrate process steps and what factors enhance the establishment of integration. These were labelled as the operational antecedents of integrative planning: process visibility, performance management, information technology, shared resources and uncertainty/variability. These antecedents were categorized in different roles: initiating, inhibiting or facility, based on the effect they had on the integration of patient planning.
However, in the study of Drupsteen et al. (2016) the internal supply chain within the orthopaedics departments was investigated, which is characterized by low complexity care (e.g. low variety and uncertainty making the care processes highly similar) and high volumes of patients. These characteristics make it possible to create critical pathways that consist of clear procedures on the sequence and timing of actions to be taken in the process in order to achieve optimal efficiency (Pearson, Goulart-Fisher & Lee., 1995). Hospitals are able to do so because these patient groups are characterized by high-volume and patient routing is highly predictable from one process step to the next. This simplifies the process of allocating the limited resources present at hospitals (Drupsteen et al., 2016).
Therefore, the adequate allocation of limited resources (e.g. operating rooms, diagnostic equipment, beds, and staff) has a positive effect on reducing waiting times for patients (Hyer et al,. 2009). These shared resources have been researched widely in a manufacturing context where the presence of shared resources are characterized as an important barrier to integration of patient planning due to the difficulty in allocation of the limited capacity Drupsteen et al., 2016). Despite these insights it is unclear whether the findings of Drupsteen et al. (2016) on the role of shared resources as an operational antecedent as well as the role of other antecedents also hold within the context of a complex care pathway (e.g. multi-morbidity, higher risk on complications, higher urgency). Therefore, in this research the oncology supply chain is investigated and more specifically the mamma care pathway.
4 impact they have on integrating planning activities. Findings could contribute to the knowledge on the influence of complexity of care on the reasons why hospitals find it difficult to integrate patient planning. Furthermore, the research could provide more knowledge on the use of integrative practices for hospital professionals when managing and coordination processes within complex care pathways.
The research question is therefore as follows:
“How do shared resources as an operational antecedent influence the integration of patient planning in a complex care process?”
5
Chapter 2. Theoretical background
2.1 Integration of patient planning in hospitals
The focal subject of the investigation is the integrative patient planning in hospitals. Therefore it is important to make a brief distinction of what is meant with integration of patient planning in this research setting. Integration in a hospital setting is mostly defined as bringing together different activities and organizational units for the benefits of patients (Axelsson, Axelsson, Gustafsson & Seemann., 2014). Moreover it is important to state that integration of patient planning can be horizontal as well as vertical. Horizontal integration is achieved when patient planning activities in a hospital are linked on the same organizational level (Gröne & Garcia-Barbero, 2001) i.e. sequencing of appointments between two functional departments in a hospital, i.e. the radiology and oncology department, in order to create an efficient flow for the patient and sequence resources efficiently. On the other hand there is vertical integration which brings together patient planning on different levels in a hospital (Gröne & Garcia-Barbero, 2001; Hulshof, Boucherie, Hans & Hurink 2013 ). In contrast to horizontal integration, which is relatively straightforward, vertical patient planning levels can be subdivided in the hierarchical levels of strategic, tactical, and operational patient planning activities (Hans et al., 2012).
Vertical integration
The strategic level defines the direction in which hospital’s want to move when it comes to the future state of their process of health care delivery (Hulshof, Kortbeek, Boucherie, Hans & Bakker, 2012). Therefore the planning horizon is long and examples could be the location determination of a certain facility, the acquisition of new equipment and decisions on the patient service and case mix (Hulshof et al., 2012). These decisions are translated into the tactical planning of activities in the health care delivery process in which resources are allocated to different specialties and patient groups (Vissers, Bertrand & De Vries, 2001). In addition the use of temporary capacity by means of hiring extra staff or working overtime are also classified as tactical decisions. Demand for the allocation of resource capacity is forecasted based on seasonal demand, waiting list information and demand of patients already under treatment (Hulshof et al., 2012). On the operational level, elective demand and therefore the resource needs at the individual patient level is known and only the emergency demand needs to be forecasted. At this level the flexibility in adjusting capacity of resources is low since the operational and tactical level have imposed constraints on the scope for decision making on the operational level (Hulshof et al., 2012). Decisions at this level can be divided into ‘offline’ and ‘online’ planning activities in which the offline planning comprises of the in advance planning of operations and the offline planning reflects the decisions of controlling and monitoring unplanned events, such as the rescheduling of elective patients for the benefit of an emergency patient who requires immediate attention (Hulshof et al., 2012). Similar to the tactical planning, operational planning can also be subdivided into both ‘offline’ ‘and ‘online’ planning activities and represent the short-term decision making in the operational process of health care delivery (Vissers et al., 2001)
6 that the resources between departments and specialties are shared and should make the allocation of patients possible, because proper integrated patient planning ties these activities together (Rhyne & Jupp, 1988). However, the allocation of capacity to certain patient groups in a hospital is extremely difficult due to the functional silos that are present. Supplying departments have to make decisions on which capacity is given when and to which demanding department. The allocation of Magnetic Resonance Imaging (MRI) capacity for example, is decided upon at the tactical level. If at a certain moment the number of patients needing an MRI exceeds the expected number on which the allocation is based, problems concerning throughput and waiting times occur. At that moment it is all hands on deck for the practitioners at the operational level who have to deal with these effects originating from inflexible capacity allocation.
Benefits of integration
Coordinating and combining functions between departments, require integrative activities between departments in a hospital, which could be caused by their functional separation. This field has been studied and adopted in supply chain management industries with sound evidence of the benefits of achieving more integration between departments. Integration is described by Barki & Pinsonneault (2005) as the extent to which the functionally separated departments constitute a unified whole. Consequently, research conducted by Narasimhan & Das (2001) and Droge, Jayaram & Vickery (2004) show that in the manufacturing industry more integration between departments leads to a higher performance of the organization overall. Health care systems however are slow in embracing these practices (Sampson, Schmidd, Gardner & Van Orden, 2015; Cardoen et al., 2010). Furthermore, it is still uncertain whether these findings also apply in specific healthcare settings and what the precise effects of integrative practices are (Drupsteen et al., 2015; Thrasher, Craighead & Byrd, 2010). Insights on how integration takes place in a manufacturing context can therefore be useful in describing integration in more detail for a healthcare setting. The next paragraph is therefore dedicated to showing how the disintegration of integration in manufacturing is used in order to describe it in a health care context
stages of integration
Explorative research has been conducted on the content of integration by Van Donk & Van der Vaart (2004) who discovered three distinct stages of integration. These findings are in line with the argument of Narasimhan and Kim (2001), who argue that integrating planning and control activities takes place in different stages, from independently operating departments towards external integration. And the assumption by Pagell (2004) that organizations move a long a distinct path from no integration to a fully integrated supply chain. Van Donk & Van Der Vaart (2004) three stages of integration in which they uncovered four integrative mechanisms which are necessary in achieving integrated patient planning between departments. These were identified as: sharing waiting list information, sharing relevant planning information, cross-departmental planning, and creating combined appointments.
7
2. The commitment and coordination stage: Departments not only share all relevant information, but
also agree upon certain reciprocal responsibilities (e.g. resource capacity reservation). This relates to the resources planning and control stage where decisions are made on a tactical level regarding the allocation of leading shared resources.
3. The integrative planning stage: Planning of certain departments is relatively centralized. In this stage one of the major barriers is the existence of shared resources. Furthermore, at this stage decisions are made on a tactical level in order to centralize planning activities between departments. For example combining appointments of patients who have to go through multiple steps between different departments in the care process. These decisions can usually not be made on an ad-hoc basis unless they are planned in advance on a tactical level.
These stages describe the integrative practices needed to achieve a certain stage of integration. However, due to the fact that decision making is performed on three hierarchical levels it is difficult for functional departments to achieve a ‘higher’ level of integrative planning. The main reason for this, is the existence of hierarchical decision making explained earlier, where the flexibility in making adjustments, particularly on resource capacity, imposes constraints at lower levels. Therefore, shifting capacity problems eventually onto the operational level, where practitioners have to solve these problems. Therefore it is valuable for hospital professionals to know what operational antecedents help or hinder integration. Since a better understanding of these concepts help achieving higher levels of integration and in turn prove to be beneficial for a hospital’s performance (Drupsteen et al., 2016).
2.2 Shared resources in healthcare
It became clear in section 2.1. that higher hierarchical level impose boundaries on the amounts, availability and timing of resources at lower hierarchical levels (Vissers et al., 2001). This often causes healthcare managers to be confronted with operational problems regarding shared resources where a quick response is necessary in solving imminent health care needs. In these situations it is often claimed by managers that “more capacity” is the answer to solve these problems (Hans et al., 2012). This is referred to as the ‘real-time hype’ of managers. The result is that managers take drastic measures at the strategic level, where in most cases it would be far more effective to tactically allocate and organize the available resources (Hans et al., 2012). Therefore it seems that imminent operational problems can result in radical decision making at higher levels, when it is not necessarily effective.
8 decided that there is a capacity problem for the radiography modalities (CT, US, etc.), which causes radical investment decisions and are not always the most effective solution (Hans et al., 2012). Therefore it could be valuable to make a distinction between different types of shared resources to shed light on how and why health care professionals make certain planning decisions at the described hierarchical levels (operational, tactical and strategic).
Types of shared resources present within hospitals
The ‘problem’ described above on radical decision making would be described by Vissers et al. (2001) as a problem regarding the time-shared resources within a hospital. They have distinguished four types of shared resources that appear in a hospital: Time shared resources, dedicated and shared patient group resources, specialist-time as shared resource and other shared resources. These types are useful in characterizing the shared resource present in hospitals because they shed light on the multiplicity of shared resources and characterize the subjects of decision making at the hierarchical levels.
Time-shared resources
These are resources which are allocated to a certain specialist or department for a specified period of time. Departments or patient groups can indicate their preferred periods for a time-shared resource in which the allocated time depends on the total allocation requests of all specialties. Examples of these resources are operating theatres and outpatient facilities (Vissers et al., 2001). Decisions made on the allocation of time-shared resources can be viewed as tactical decisions since it does not involve a long time frame. The main resource in the allocation of time-shared resources is specialist capacity which is allocated first to the different specialities and are then allocated further to different patient groups or used as shared resource. (Vissers et al., 2005).
Dedicated and shared patient group resources
For some patient groups a focused factory could be created, where resources are dedicated to certain patient groups. Therefore it is necessary to have a resource structure that is clearly defined where part of the resources are dedicated and are used by a specific patient group only. Other resources will be easily accessible and other resources are contracted from departments managed at hospital level (Vissers et al., 2001). Decision making regarding allocating resources only to a certain patient group is considered as part of strategic policy making by the hospital and the reasons for dedicating resources could be quality, costs, and control of resources. These resources do not play an active role in the issues of integration because they have already been dedicated to a certain patient group. However, it is valuable to understand the existence of these resources in the care process.
Specialist-time as shared resource
9
Other shared resources
The last type of resources are resources which are generally available for all departments and do not have special commitments to certain specialties. Examples would be diagnostic imaging departments such as the X-ray department and the intensive care unit.
Using these types of shared resources could provide new insights into what the role is of shared resources, within the process of integrating planning activities between departments. Furthermore, it could give healthcare professionals a better understanding of the role of shared resources in comparison to the ‘shared resources’ as described and used by Drupsteen et al. (2016).
2.3 Shared resources in operations management
Shared resources have been explored in operations management to much greater extent than in health care settings. Therefore, applicable insights regarding the use of shared resources in an operations management context are discussed in this paragraph, in order to discuss shared resources in more depth and see how they could relate to the healthcare environment.
Fawcet, Magnan & McCarter (2008) investigated causes why supply chain partners within a supply chain were unable to successfully share resources and effectively use these resources. Two categories of main causes were identified: inter-firm rivalry and managerial complexity. Inter-firm rivalry is concerned with the misalignment of motives and behaviour between departments within the supply chain (Park & Ungson, 2001) where departments compete with each other rather than cooperate (Fawcet et al., 2008). These characteristics might as well occur within hospitals where shared resource capacity and therefore planning of these resources has to be divided between departments. Functional dispersed departments within a hospital setting are in an sense competing for shared resources. Furthermore, the lack of cooperation between departments also uncovered that there was a lack of partner trust which was an underlying explanation of the inter-firm rivalry (Fawcet et al., 2008). Managerial complexity was regarded as the misalignment of an organizations processes and structures, which were caused by incompatible information systems, conflicting organizational structure and culture (Fawcet et al., 2008). They also stated that departments only felt comfortable using their own information systems for their own tasks, which resulted in inconsistent information and lack of sharing information between departments (Fawcet et al., 2008). These could be valid reasons why hospital departments find it difficult to effectively share resources between each other.
10 the health care delivery process (Lluch, 2011). Furthermore, the extent of uncertainty within a firm determines the appropriate strategy to use to integrate planning and control activities (Van Donk & Van der Vaart, 2005).
Van Donk & Van der Vaart (2005) stated that the level of integration mostly depends on the degree of uncertainty within the supply chain. The framework that they constructed shows how an organization should organize its shared resources at different levels of uncertainty, varying from low volume, low mix/specifications towards high volume, high mix/specifications. If we would translate these back to a healthcare context, one could argue that the orthopaedics department investigated by Drupsteen et al. (2016), characterized by a high volume and low mix/specifications care process would require a different strategy towards shared resources than a hospital department characterized by a low volume and high mix/specifications care process such as oncology department. According to Van Donk & Van der Vaart (2005) an environment characterized by high volume, low mix/specifications would have difficulty in the planning of capacity due to the high volume levels and would therefore necessitate the use of integrating shared resources in order to improve the physical flow to cope with these levels. In contrast an environment characterized by low volume, high mix/specifications requires a higher level of integration in order to react to the high mix and low volume characteristics. In this case resource capacity needs to be reserved to enable coping with the broad scope of inflow. The latter situation would roughly translate to the complex care process getting every attention and therefore the needed allocation of capacity to match the patients demands with the required resources. Therefore the use of shared resources to integrate practices in order to improve patient flow at different levels of uncertainty varies. Consequently research in a more complex environment could provide new insights in literature as to how different types of shared resources should be managed.
2.4 Operational antecedents of integration
11 Within the orthopaedics department a three-way split was found in which five antecedents where characterized as having an inhibiting, facilitating or initiating role on the stages of integration. These are described and evaluated below.
Process visibility
The visibility of the process was regarded as an initiating factor for integration. Health care professionals need to have some insights in what the steps are within the process to be able to integrate and have staff being aware of relevant planning information for other departments and steps in the process (Drupsteen et al., 2016). Furthermore, process visibility as described by Buchanan (1998) states that hospital professionals have difficulty in communicating relevant planning information to other professionals in steps preceding or subsequent to the process step of patients they are involved in. Process mapping would be a way of counteracting this problem according to Barrat & Oke (2007) to increase the visibility in the process. Giving hospital professionals insights into the planning functions and processes in the adjacent process steps would lead to communication and sharing of information between departments to improve the flow in the care process (Barrat & Oke, 2007).
Performance management
To make integration feasible between departments, communication on performance goals is necessary to fuse requirements together (Stank et al., 2001). However, within healthcare settings departments function as silos, focussing on their own performance and in most cases optimal capacity utilization is the key objective (Vissers & Beech, 2005). Therefore integral performance management focussing on the sharing of performance goals between departments is seen as an initiator of integration as opposed to local performance management focussed on the performance of individual functional departments (Drupsteen et al., 2016).
12
Information technology
Departments depend on information technology to better coordinate patient flow by the effective exchange of relevant planning information (Devaraj, Ow & Kohli, 2013). Furthermore, the use of information technology helps in facilitating and executing the integrative practices in hospitals (Pagell, 2004). These characteristics of information technology in hospitals were discovered to be acting as a facilitator to the use of integrative practices in all three stages of integration (Drupsteen et al, 2016)
Uncertainty/variability
Within the research performed by Drupsteen et al. (2016) these operational antecedents were found and are further distinguished in routing uncertainty and demand variability. Routing uncertainty can be described as the degree to which a patient process from one step to the next is subject to uncertainty. Demand variability can be explained by means of the degree to which demand patterns for a certain pathway or process is known in advance. Therefore these antecedents are sometimes used interchangeably since variability can lead to uncertainty but not the other way round. Both uncertainty and variability form barriers in the final two stages of integration (Drupsteen et al., 2016).
Shared resources
13
2.5 Conceptual model
Guided by the findings of earlier research and the combination of operations management literature translated to a healthcare context has lead towards the visualization of this research into a conceptual model (figure 2.5.1). The model shows the stages of integration as described earlier together with the operational antecedents and the three-way-split based on the impact that was found by Drupsteen et al. (2016). However, the findings that form the basis of this model have been investigated in a non-complex healthcare environment which immediate raises questions as to the generalizability of the role that the operational antecedents have on the stages of integration. Furthermore, research regarding shared resources show different characteristics they can pertain and how they should be managed by hospital professionals. Therefore, it could be valuable to see what impact these individual types of shared resource have on the integration of planning activities within a more complex healthcare setting (I.e. the oncology department). The classification of antecedents in the model is given by their initiating, facilitating and inhibiting role on the stages of integration as found in the orthopaedics supply chain by Drupsteen et al. (2016). An addition to their model has been made regarding the shared resources. These are viewed as having an inhibiting role in achieving the last two stages of integration (i.e. the commitment and coordination stage and the integrative planning stage). However, by disintegrating shared resource into the four types as described earlier we hope to discover new theory on the impact these types can have and their role towards the integrative stages.
14
Chapter 3. Methodology
3.1 Research method and case selection
The aim of this research is providing insights on the role of operational antecedents in a complex care process, with a focus on shared resources and how healthcare professionals on different hierarchical levels behave and cope regarding these antecedents. These will be investigated in a mamma care pathway which is part of the oncology department. By choosing this particular department it provides the ability to make a comparison with the orthopaedics department which was investigated by Drupsteen et al. (2016). Since these cases represent extremes (i.e. high volume/low complexity versus low volume/high complexity a comparison can be made with regard to the results in the terms of Miles and Huberman (1994) to be able to ensure a theoretical replication as mentioned by Yin (2009). Furthermore, due to the complex characteristics of the mamma care pathway (e.g. multi-morbidity, higher risk on complications, higher urgency), the findings could provide a better understanding of the role of these operational antecedents in achieving integrative planning between hospital departments and the hierarchical levels in which hospital professionals operate within these departments.
A single case study will be conducted due to the explorative nature of the research. It is the preferred method since not all characteristics are understood (Eisenhardt. 1989; Meredith, 1998). The aim is to expose the characteristics of the operational antecedents present in the pathway and their role on the integrative stages. An answer to the research question will be found by means of interviews to investigate the complex social interactions by collecting qualitative data. Since the research will be done as a single case study it is important to compensate for the lack of external validity. This can be done by analysing the results of the findings of research done by Drupsteen et al. (2016) and comparing both results. The research of Drupsteen et al. (2016) has specified a set of clear propositions and circumstances on which theory has been built. However, the theory can be extended by the use of an extreme single case to determine whether these propositions also hold in a case characterized by different conditions, but still reflect the characteristics that are identified in the conceptual model (Yin 2009). Furthermore, the mamma care pathway could provide some alternative explanations that might be relevant for expanding theory on the subject.
15 Drupsteen et al. (2016) visualized these steps which is included in appendix II. Comparing this to the care process of breast cancer patients included in appendix III, and described in appendix IV, shows the difference in complexity of routings.
The mamma care supply chain consists of four main departments who execute planning activities in the patient care process. These are the oncology, radiology, operation room and integral planning department. The care pathway that is selected for the single case study is part of the oncology department at one of the largest general hospitals in the Netherlands. A number of characteristics of the hospital are listed in table 3.1.1.
Type Number of beds FTE employees Patients treated per year
Regional 671 3046 288.455
Table 3.1.1: hospital characteristics *approximate
3.2 Data collection/Data sources
Data is collected by conducting semi-structured interviews with subjects from each department in the mamma care supply chain. A standardized interview protocol (Appendix I) is used for all subjects to increase the validity of collected data (Voss et al. 2002). The protocol is based on the interview protocol used in the research of Drupsteen et al. (2016). The protocol is divided into two parts where the first part will give insights in the used integrative practices. The second part of the protocol focusses on the operational antecedents that help or hinder integration as visualized in the conceptual model. The protocol by Drupsteen et al. (2016) was adapted by adding extra questions focussing on shared resources in order to gather more in depth data regarding the four types of shared resources. The participants that are interviewed were selected by several key informants at the departments of oncology, radiology, and the key informant at the hospital control centre. They were selected based on the knowledge each informant has on his or her involvement in the planning and execution of the mamma care pathway. To be confident of a multifaceted view of the pathway, interviewees have been chosen based on their distinctive role within the pathway of mamma care patients to create a good representation of actors within the chain, these are listed in the table 3.2.1.
interviewee department hierarchical level
manager oncology strategic
team manager oncology tactical
staff member oncology operational/tactical
Mamma care secretary oncology operational
nurse consultant oncology operational
surgeon oncologist oncology operational
manager radiology strategic
team manager radiology radiology operational/tactical
radiographer radiology operational
manager operating room strategic
planner operating room/integral planning operational
16 By interviewing multiple interviewees within each of the four departments and on the three distinctive hierarchical levels, it is possible to increase the internal validity because insights on whether and how the constructs are related can be combined from each distinctive role in the chain. Transcripts of the 12 interviews were sent to each informant to verify the transcribed text. This resulted in minor changes within the acquired data.
3.3 Data coding and analysis
17
Chapter 4. Results
In this section both the degree of integration as well as the underlying operational antecedents are established. This is done by means of a within case analysis to determine the integrative practices present within the breast cancer pathway as well as the factors that influence achieving a certain stage of integration.
4.1 Integrative practices: within case analysis
In the breast cancer supply chain, six integrative practices were identified: Three concerning the transparency stage; two practices of cross-departmental scheduling and one form of integrative planning. These are presented in figure 4.1.
Sharing information
Within the breast cancer supply chain the oncology department is responsible for referring and scheduling patients for the diagnostic phase at the beginning of the process. Therefore the
responsibility of scheduling the patients lies with the mamma care secretaries. They keep track of the maximum date of surgery which is corresponded with the integral planning department who are responsible for the planning of surgeries. A necessity for the correct referrals is information on the patients and other valuable information regarding the process. These data are accessible for everyone in the information system and are not ‘actively’ shared between departments. However, within the planning process certain information is actively shared by means of a weekly meeting regarding patient planning. In these meetings the mamma care secretaries as well as someone from the integral planning department discuss the current waiting times and throughput of patients to see if adjustments or immediate action has to be taken on particular cases. An example would be trying to expedite an ultrasound to ensure that a patient is ready before the target date of surgery.
Cross-departmental planning
In the planning process there are two forms of cross-departmental planning: The ability of the
oncology department to schedule radiology appointments themselves and slot reservation of the OR. The outpatient clinic is able to schedule patients within a template carried out by the radiology department. However, most of the appointments are still discussed with the radiology department whether it fits. A reason for this is the complexity surrounding mamma care patients where, due to their diverse medical conditions the same procedure or scan can take 20 minutes for one patient and 40 minutes for another. Furthermore, the mamma care secretaries are not able to schedule the placement of iodine (radioactive) seeds since the legal planning procedure of this procedure is so complex that it is best carried out by the radiology department.
The other form of cross-departmental planning stems from the reservation of slots for combined surgery. In these cases the surgery is performed with both a surgeon as well as a plastic surgeon present. Each month two slots are reserved for these patients. All other surgeries are planned into the OR templates without reciprocal agreements between the oncology department and the OR.
Combined appointments
There is one form of combined appointments present in the process which concerns the first process steps of new patients. New patients arrive at the hospital with an intake at the outpatient clinic and after that are able to directly be seen by the radiology department for a mammography, an
Integrative practice type stage informant's quote Informant department Waiting list times of
radiology is shared with other departments
Sharing
information Transparancy
"Departments are able to open my balanced scorecard and see what the waiting times are, but we are not used to sharing this kind of information yet"
Manager (radiology)
Maximum date of surgery is corresponded with the planning department
Sharing
information Transparancy
"The oncology department provides us with information on the
maximum date for surgery to schedule patients for OR" Planner (integral planning)
A meeting regarding patient planning is scheduled each week
Sharing
information Transparancy
"Every Wednesday morning we have a patient planning meeting scheduled concerning the waiting time and throughput targets." "A short while ago we decided to have someone from the integral planning department attend these meetings"
Team manager (oncology)
Oncology schedules radiology appointments Cross-departmental scheduling Commitment and Coordination
"The oncology department is able to schedule certain
appointments themselves" Radiologist (radiology)
Reserved OR slots for surgery Cross-departmental scheduling Commitment and Coordination
"There are two slots reserved each month for this type of surgery
at the OR with the surgeon and plastic surgeon" Planner (integral planning)
Appointments are combined between OCL and radiology – 3 spots per morning
Combined appointments
Integrative planning
"Every morning three patients are examined by a doctor, after which they directly have an appointment with the radiology department for a mammography, ultrasound and if needed an ultrasound guided biopsy"
Mamma care secretary (oncology)
4.2 Antecedents of integration: within case analysis
After analysing and coding the data, seven operational antecedents were discovered that have an impact on the integration of patient planning: performance management, process visibility, information technology, uncertainty/variability, patient autonomy and shared resources. Each antecedent is justified by means of multiple subject quotes who span multiple departments within the unit of analysis. The antecedents are discussed in turn in which the latter, shared resources, showed a noticeable 4-way split and is therefore dissected into four parts.
Performance management
Within the investigated pathway it became apparent that departments move in their own way and on their own terms. This is reflected by subjects saying that coordination of patient planning is difficult due to the silo-mentality of departments. Several reasons were mentioned such as the lack of responsibility for the entire chain that causes a barrier for cooperation which is caused by departments focussing on their own production figures and that every department has its own interests:
Manager OR: Departments themselves are moving and the other departments have to keep up. Those are the bugs in the system. / At the moment that financial subjects are on the agenda you see that interests are separated.
Manager OCL: Personal interests and interests of departments hinder cooperation in which too less attention is paid to the interests of patients and subsequently acted upon.
Team manager OCL: It is important that everyone feels the responsibility for moving patients through the entire chain and we notice that it forms a barrier if someone does not feel that responsibility.
Nurse consultant: That is the reason why the OR planner is now always present at the weekly planning overview meeting so that everyone knows the current situation of patients and silo mentality is breached.
Another observation that surfaced from the data showed that cooperation between the departments is focussed mainly on reactive problem solving instead of preventative steps to improve patient planning. This is illustrated in the monitoring of patients within the pathway which takes place every week between the mamma care clinic and the operating room planners. This shows that only those people who have the responsibility of the actual patient planning (e.g. operational practitioners) initiate steps to improve patient flow and that all involved departments in the chain do not share goals on higher levels concerning the lead time of patients:
Manager oncology: Issues regarding patient planning are put on the agenda when lead times and the norms that have been set are not being met.
1
Mamma care secretary: We steer upon the goals within the pathway since we have a clear overview of those particular goals. Therefore the weekly monitoring meeting is useful to check and act upon acute situations in which a patient risk of not being ready for surgery within 5 weeks.
Despite the silo-mentality between departments, subjects mention that a more centralized approach would be beneficial for patient planning and that departmental cooperation should be guided more towards continuously improving on commitments between departments to improve patient planning and therefore lead times:
Team manager radiology: You should revise the current commitments between departments more frequently to check whether these commitments still suffice and try to prevent these issues in the future.
Manager oncology: More actively coordinate with each other, inspire each other and exchange expertise and knowledge.
Manager OR: We do not share performance goals well enough with everyone. We do not share production figures but we should.
Centralizing the complete mamma care was also mentioned by every department which is viewed as an enabler of major improvements for patient planning. Factors such as patient volume uncertainty, capacity allocation and coordination between departments were mentioned as areas in which those improvements would take place:
Nurse consultant: If you have all referrals from the screening program, inflow of patients would become more even.
Team manager radiology: A one-stop shop would be great. We all know the phenomena and in part we implement it. But for these type of patients it would be a wonderful trajectory.
Surgeon oncologist: We want to centralize that particular pathway very badly to prevent planning problems.
Team manager oncology: It would be great if we would have the sole responsibility over the entire chain. To have dedicated diagnostics and have everything under one roof. Therefore the mamma centre would be an improvement for patient planning in the sense that things don’t have to go through as many players as they do now.
Process visibility
Due to the fact that departments mostly function as compartmentalized units process visibility on the exact working methods is low and therefore non transparent. Decisions are made without knowing what the effect will be on other departments and how it will affect the entire chain. Knowing where the patient is at what particular moment is therefore seen as a benefit for the exposure of bottlenecks within the chain to improve patient flow by means of having more visibility in the process and acting upon it.
2
Manager radiology: I think that one of the biggest factors is not being able to oversee what the effect of your decisions is for the other department
Planner OR: I am not aware how patient planning is done at the mama care clinic.
Process visibility is not only limited by the fact that process information is not shared, but can also be caused by departments not knowing what information on the process is relevant for the other department. Furthermore, it is not always clear what particular information on the process is actually available and were it can be found:
Manager radiology: My balanced scorecard is visible for everyone in which waiting times are shown. However, we are still unaware of sharing that kind of information.
Nurse consultant: Not having visibility in the process of other departments limit cooperation. From the data is was clear that limited process visibility negatively influences patient flow due to slow responsiveness of departments caused by other departments not sharing significantly important information which affect patient planning:
Manager radiology: Sometimes the surgeons get extra time, but we do not get notified. / It is useful if we know in advance that the outpatient clinic is going to process more patients due to an increase in referrals from the screening program or the GP’s. However, you notice that this is often not corresponded.
Radiographer: If I get the request from the outpatient clinic to schedule an extra patient and I have a free spot somewhere to fill, but I don’t know that the patient is using an anticoagulant, then it interferes with the patient schedule.
Team manager radiology: I notice that relevant information such as maintenance on equipment or not having a radiologist is shared quicker than before, meaning that I can steer the process more adequately.
These issues show that hospital professionals sometimes find it difficult to communicate relevant planning information to other professionals in steps preceding or subsequent to the process step they are involved in. Therefore, it forms a major barrier in achieving the transparency stage and subsequently achieving internal integration. Process visibility is therefore seen by subjects as a prerequisite for improving patient flow and being able to more adequately ‘steer’ the process.
Information technology
3
Radiographer: Using the new technology takes a lot of time. Everything is locked with an enormous amount of questions which you all have to go through to before you can do something.
Team manager radiology: They are working on making it easier for us to use the program.
Manager OR: At the moment we are in have special circumstances in which the new information system feels as a vacuum, but that’s part of the phase that we are in and I expect that we can get a lot out of it.
Team manager OCL: We now have to go to the optimization phase and I think that the system can be an important facilitator, therefore I am keen on the new system.
The intention of the new system is to facilitate integration between departments and data shows that the step by step improvement in the use of the system enable better coordination between departments on all fronts and that departments already see the potential in comparison to the earlier system. The new system should eventually be able to monitor the entire pathway of individual patients from registration to the last hospital visit. This will facilitate a better provision of information for the hospital itself as well as for external parties. Ultimately the new system should also make it easier to discover bottlenecks within the process and where they occur, to be able to control and react upon the process more adequately.
Uncertainty/variability
Uncertainty plays an important role in the mamma care pathway. This is reflected by the vast complexity of the variation in routings that patients follow. There are clear guidelines as to what procedures need to be undertaken from one process step to the other, but almost the entire pathway can be characterized as a decoupled process in which you only know what to do next if the preceding step has been completed and the course of action is discussed in interdisciplinary meetings:
Nurse consultant: The entire routing is not always known. It depends on the diagnosis and what is being discussed in the interdisciplinary round. Only on rare occasions is it clear what will be done.
Mamma care secretary: The patient visits the hospital and comes back three days later for the results. These can be good or bad and if its shows malignity then you know what to do and only then can you start planning.
Radiographer: You don’t know it in advance. You get 3, sometimes 4 new patients and it all depends on what you see. They have been referred with a suspicion, but you never know. Sometimes there is one abnormality in the breast, but sometimes you have 2 or three abnormalities which can all be different. So you never know.
4
Volume variability
Variability was also seen as a barrier to both the commitment and coordination stage as well as the integrative planning stage, where a two-way split was found in variability (i.e. volume variability and artificial variability):
Mamma care secretary: Scheduling two multidisciplinary rounds per week is not feasible at the moment because of the low volume (15 patients per week) and because it is too difficult to coordinate it in such a way that all specialists are there.
Nurse consultant: At the moment it is not feasible to organize two multidisciplinary rounds a week, since we are not that busy. However, we had moments when we had 30 patients on the list where it would be better to divide those patients over two rounds.
Volume variability was therefore seen as a barrier to the commitment and coordination stage because it interfered with the use of interdisciplinary rounds in the process. Committing to more rounds within one week accelerates the decision making for the proceeding steps for patients, but not all actors within the chain are able to comply due to volume variability of patients and committing to certain agreements. The other type of variability that was found was artificial variability, stemming from flaws within distinct processes in the chain:
Artificial variability
Planner OR: If the patient had an intake with the surgeon and decides that she wants an operation with direct reconstruction and we get notified, then we will reserve a spot for combined surgery (surgical oncologist and plastic surgeon). However, the patient had an intake a week later with the plastic surgeon and wants a different type of reconstruction, resulting the slot being free again.
Head of logistics and registration: It would be more ideal if the patient would already have all information regarding the possible treatments and decide in a couple of days what she wants. But in the case of having a gap of two weeks and after the first intake you assume she will have a combined surgery but it becomes a different reconstruction, you keep lagging behind the patient in a reactive manner.
Radiographer: Some GP’s do not follow the right procedure when it comes to the referral of a new patient. This happens quite often and they are referred to us instead of the mamma clinic. In those cases you have to send them back to the GP who has to refer them to the mamma clinic for additional diagnostics, already losing two weeks.
5 Shared resources
Several bottlenecks for integration regarding shared resources were distinguished by the informants. These bottlenecks showed clear patterns between the characteristics of the departments in which they occurred. The radiology department and OR department were both viewed as the bottleneck departments since they have limited capacity both in terms of equipment and personnel. These departments have to deal with the difficult task of coordinating all incoming requests of the hospital and allocating capacity accordingly. These were considered to be the time shared resources which form a barrier reaching the commitment and coordination stage, as well as the integrative planning stage:
Staff member quality and safety: For example we only have one mammograph and that is a limitation which is not solved in the blink of an eye, but it is important to take into account.
Manager radiology: In relation with the total amount of work we have too little radiologists.
Surgical oncologist: However, a prerequisite is of course that an operating room is available, because most of the time we are not the problem.
The above data shows the impact time shared resources have on the use of integrative practices and gives rise to the question whether resources should be dedicated. Informants gave the preference to resources being dedicated in order to improve patient planning. However, due to uncertainty in patient routing and complexity, supplying departments (radiology and OR) are reluctant in making more commitments:
Radiographer: If you get a patient who requires all types of diagnostics and only 20 minutes are scheduled, you are not going to make it. But if you have patients who only need a follow check-up, you can do five or six. Same goes for an ultrasound, then you can do a lot more. You never know what you get.
Team manager radiology: The problems that occur are often a matter of diagnostics taking longer than you might have expected or that other departments refer to many patients.
The importance of making more commitments between departments is apparent when the circumstances demand it, but when dedicated slots would take up more time than scheduled, waiting times will increase for other patient groups. Caused by the uncertainty revolving around mamma care patients. Furthermore, data showed that specialist-time shared resources can form an enabler to patient planning in the commitment and coordination stage:
Surgical oncologist: One of the patients who was planned for surgery wanted to be operated solely by me, but I was scheduled for the outpatient clinic. Therefore my colleague and I switched our schedules for that particular day so that he would perform the intakes at the mamma clinic and I would perform surgery all day.
Patient autonomy
6
Mamma care secretary: A patient could prefer a certain surgeon for surgery and if that surgeon is not available than the patient could be operated after the set goal of 5 weeks until surgery. This will be visible in the performance score we get from pink ribbon.
Manager OCL: Sometimes patients want to go on a holiday first and then go for surgery. These patients are overdue for the 5 weeks until surgery and this could be the difference between acquiring the performance goals or not.
7
Chapter 5. Discussion
The findings of the research are discussed in this section, where the impact of the operational antecedents on the integrative stages is explained in connection to the literature. The data shows similar outcomes to research of Drupsteen et al. (2016).The same three way split of the operational antecedents was discovered in which antecedents can be described as having an initiating, facilitating or inhibiting role on the stages of integration. However, the operational antecedents that were discovered differ from what was known from theory.
Shared resources have been split up according to the different characteristics they can have, by using theory of Vissers et al. (2011) and showed an important difference from the literature in which shared resources as a whole do not always act as an inhibiting antecedent of integration, but can also, in some cases, be regarded as an initiator as the data showed in the case of specialist time shared resources. The other types of resources however (e.g. dedicated and other shared resources), do not have any direct influence on the integrative practices of patient planning.
Furthermore, a new inhibiting antecedent emerged within the mamma care pathway which is described as patient autonomy. Giving patients the possibility to influence certain process steps leads to uncertainty on the routing of the patient and therefore make departments reluctant of indulging in more commitments and integrative planning practices with the outpatient clinic. The revised model is shown below (figure 5.1). Although uncertainty and variability were bundled in the initial model due to their interchangeable characteristics, they are split up in the revised model. The reason for this is that it clarifies the distinction between routing uncertainty and demand variability and makes further discussion of these antecedents easier in relation to literature.
8
5.1 The impact of operational antecedents
As visualised in the revised model all operational antecedents that were found are part of a particular role on the degree of integration: Facilitating, initiating and inhibiting antecedents. These roles and their underlying antecedents are not different from what was known from theory. However, this research adds to literature on the fact that these also hold in a pathway characterized by complexity.
Facilitating antecedents
Literature suggested that information technology would have a facilitating role in each of the three stages of integration, this corresponded with the outcomes of the study. The information technology for the hospital changed over the course of the previous year and therefore provided insights in the pre phase of the new information technology system as well as the post phase of the new information technology. The system is viewed as troublesome for most people within the chain and form barriers to the daily operational practices, but the technology in itself is static and can be continuously improved in such a way that the wishes of all actors in the chain are reflected in the system. For that reason the information technology is still regarded as a facilitating antecedent.
Initiating antecedents
Process visibility is closely linked to the uncertainty and unawareness that is facilitated by the functional silos found in hospitals. The mamma care pathway showed a clear resemblance to the findings in literature but is characterized by closer cooperation between departments in terms of the medical process and routing variability. Therefore the dependencies between department are valued as extremely important, but do not have the effect of enabling more process visibility between functional silos.
The second initiating antecedent that occurred was performance management which as Drupsteen et al. (2016) suggested would enable the use of integrative practices in the second and third stages of integration. However, it must be mentioned that the connotation of performance management as an initiator lies in the integral values it upholds, where the sharing of performance goals between departments is key. This in contrast to local performance management as described by Vissers & Beech (2005) where the focus of departments lies in the optimization of capacity utilization. Therefore the transfer of departmental performance management towards integral performance management creates an incentive for integration. These results correspond with the findings of Drupsteen et al. (2016) and Elg, Palmberg & Kollberg (2013) who concluded that integral performance management by means of sharing of goals, initiate collaboration between professional groups.
9 viable to have more commitment between departments and be able to improve patient flow. The flexibility that time-shared resources in this particular case have changed the pre-set surgical time in the OR schedule. This is in line with the argument of Van Donk & Van der Vaart (2004) where they state that flexibility of leading shared resources, in this case the specialist time, is necessary to reach the commitment and coordination stage, where the initial optimal utilization of the resource, the OR schedule, does not satisfy the need of the buyer. The latter being the patient in the breast cancer chain who requested a particular surgeon to perform the operation.
Inhibiting antecedents
Within the mamma care chain almost all shared resources can be described as time shared resources and they prove to form a barrier to integration which is consistent with the findings of Drupsteen et al. (2016) for shared resources as a whole. However, it is important to denote that specialists can have a dual role in the distinction made between shared resources. Surgical oncologists for instance can shift tasks in different phases of the process and are therefore distinguished as time-shared resources. In contrast, all other specialists; radiologists, radiographers, plastic surgeons, etc. cannot shift tasks in different phases since they only perform a certain task in the process. Demanding departments have to compete with other departments for the time shared resources which are available at the supplying departments. This causes inter-firm rivalry between the functional silos as suggested by Fawcet et al. (2008) and in this setting showed that silos mostly compete with each other rather than cooperate, especially when financial matters are at stake.
In addition, other shared and dedicated resources as described in theory did not seem to play a role in connection with the three stages of integration. However, dedicated resources themselves did not form a barrier to integrative practices, the reluctance of making use of dedicated resources such as dedicated slots did. Functional silos are mainly reluctant to create dedicated slots due to the complexity of mamma care patients in terms of routing uncertainty. Showing valuable information that the complexity of a care pathway might be a barrier to the usage of integrative practices.
A new inhibiting antecedent was also discovered, patient autonomy. Due to the large impact treatment has on patients, they have the ability to decide the precise routing for surgery and the following therapies. However, the autonomy interferes with integrative practices in the way that it increases the uncertainty as to what the proceeding step will be in the process. These findings shed new light on the inhibiting antecedents when patient autonomy is present within a pathway, and adds new insights to literature.
11
Chapter 6. Conclusion
Implications for theory and practice
Integrating planning activities within hospitals is known to improve patient flow. Despite this knowledge, hospitals have difficulties in making use of integrative practices in order to achieve a consecutive stage of integration. This research tried to extend knowledge on the antecedents that influence the use of integrative practices by investigating the following research question: How do shared resources as an operational antecedent influence the integration of patient planning in a complex care process? Theory provided the starting point for this research and after analysing the data a split was seen between two different types of shared resources (i.e. time shared resources and specialist-time shared resources). Time shared resources (e.g. OR time, time of radiologists, etc.) proved to be an inhibitor in the last two stages of integration. This is mostly caused by the limited capacity of these resources, making it difficult for departments to commit capacity to a particular pathway. However, data also revealed that the higher urgency surrounding breast cancer patients initiate the use of integrative practices. The flexibility of specialist-time shared resources is able to improve patient flow on an operational level and stimulates the use of integrative practices between the departments of oncology and OR. This result can provide insight for practitioners in the way that flexibility can act as a stimulant for the use of integrative practices.
12
Limitations and future research
13
Chapter 7. References
Aronsson, H., Abrahamsson, M., & Spens, K. (2011). Developing lean and agile health care supply chains. Supply Chain Management: An International Journal, 16(3), 176-183.
Axelsson, R., Axelsson, S., Gustafsson, J., & Seemann, J. (2014). Organizing integrated care in a university hospital: application of a conceptual framework. International journal of
integrated care, 14(2).
Barki, H., & Pinsonneault, A. (2005). A model of organizational integration, implementation effort, and performance. Organization science, 16(2), 165-179.
Bradley, E. H., Curry, L. A., & Devers, K. J. (2007). Qualitative data analysis for health services research: developing taxonomy, themes, and theory. Health services research, 42(4), 1758-1772.
Cardoen, B., Demeulemeester, E., & Beliën, J. (2010). Operating room planning and scheduling: A literature review. European Journal of Operational Research, 201(3), 921-932.
Devaraj, S., Ow, T. T., & Kohli, R. (2013). Examining the impact of information technology and patient flow on healthcare performance: A Theory of Swift and Even Flow (TSEF) perspective. Journal
of Operations Management, 31(4), 181-192.
Drupsteen, J. (2013). Treating planning flaws in patient flows (Doctoral dissertation, University of Groningen).
Drupsteen, J., van der Vaart, T., & Pieter van Donk, D. (2013). Integrative practices in hospitals and their impact on patient flow. International Journal of Operations & Production
Management, 33(7), 912-933.
Drupsteen, J., van der Vaart, T., & Van Donk, D. P. (2016). Operational antecedents of integrated patient planning in hospitals. International Journal of Operations & Production
Management, 36(8), 879-900.
Eisenhardt, K. M. (1989). Building theories from case study research. Academy of management
review, 14(4), 532-550.
Eisenhardt, K. M., & Graebner, M. E. (2007). Theory building from cases: Opportunities and challenges. Academy of management journal, 50(1), 25-32.
Elg, M., Palmberg Broryd, K., & Kollberg, B. (2013). Performance measurement to drive improvements in healthcare practice. International Journal of Operations & Production
Management, 33(11/12), 1623-1651.
