MASTER THESIS
Reducing the capacity buffer
of a shared resource
A single case study at the physiotherapy department
Robin Kuiper – S2576651 MSc. Supply Chain Management
Abstract
Healthcare organizations are under increasing pressure to reduce costs, which forces them to become efficient without reducing the quality of care. Lean, a philosophy derived from the manufacturing sector, has been used to improve the performance. However, most of the results focus on direct waste on departmental level, while the essence of lean is minimising buffers for variability on a more interdepartmental level. This research seeks to find out how a shared resource, like a physiotherapy department, can cope with waste elimination to reduce the capacity buffer. A shared resource has been chosen, because the capacity is called upon by different patient flows and should coordinate their activity with those patient flows. Results show that it is hard to control variability, as the shared resource is not coordinating their activities with other disciplines. However, some variability and waste could be easily reduced by a better coordination of the shared resource and by making processes standardized. Furthermore, this study indicates that the shared resource can cope with the current demand by making use of flexible capacity, a quality buffer and a time buffer mechanism. This time buffer mechanism could be at the core of coping with variability, as it is not the patient who waits, but the capacity resource.
Content 1. Introduction ... 3 2. Theoretical background ... 5 3. Methodology ... 9 4. Results ... 13 5. Discussion ... 21 6. Conclusions ... 27 References ... 29
Appendix A. Interview protocol ... 32
1. Introduction
Society is demanding the highest quality and affordable care of a healthcare organization. Under this strong pressure the hospitals are challenging themselves to become more efficient and deliver higher quality (Poksinska, 2010). Lean provides a way to improve performance, being a continuous improvement approach. More specifically, lean has proven to have a positive effect on the performance of the healthcare organizations in terms of quality, access and efficiency (Mazzocato, Savage, Brommels, Aronsson, & Thor, 2010). However, lean has been developed in a manufacturing environment and is being implemented in a service environment (e.g. healthcare) for which it was not designed (Burgess & Radnor, 2013). In general, research has mainly focused on ‘quick wins’ to improve performance by applications with limited organization reach (Mazzocato et al., 2010). Healthcare organizations face a few challenges for the implementation of lean: a healthcare organization is a complex system with interdependent units and it is hard to make the value for the customer visible. It should be noted that healthcare units work in silos and are not always aware of the effects of their efforts outside their own departments (Poksinska, 2010). This research will contribute to a better understanding of how lean can cope with an interdepartmental approach.
The lean philosophy has been derived from the Toyota production system and focuses on continuous improvement (Taylor, Taylor, & McSweeney, 2013). According to Mazzocato et al. (2010), lean is more than a set of tools and project teams. Lean usually focuses on continuous improvement by reducing waste (Womack & Jones, 1996). Another specific focus of lean, adopted within this thesis, is on the more specific reduction of variability. It is suggested that reducing variability can reduce (capacity, time and inventory) buffers (Hopp & Spearman, 2004), which will result in better performance of the hospital. The overall conclusion of different studies is that lean in healthcare tends to focus on direct waste on departmental level (Brandao de Souza, 2009; Burgess & Radnor, 2013; Mazzocato et al., 2010). Further development of lean has to be made in the interaction between different departments where variability occurs as a result of the transmission of patients between departments (Brandao de Souza, 2009). Various departments are involved in the different steps of a care process. For this reason, it is a challenge for departments to determine how to cope with an interdepartmental approach to improve the flow of patients. Cross-departmental implications increase when different departments make use of a shared resource.
performance implications for a department which is dependent on and collaborates with other departments, like a shared resource (Drupsteen, van der Vaart, & van Donk, 2012). Therefore the aim of this research is to fill the gap between the desired performance and the realized performance of a shared resource. The next paragraph will further discuss the shared resource. According to van Donk & van der Vaart (2005), a shared resource is “a common-capacity source in two or more networks”. In a shared resource, control of variability is complicated because a lot of departments make use of it. Using a capacity buffer is a way to cope with variability. Capacity buffers could reduce the risk of patient flow delays in a shared resource. Though, this kind of buffer is not desirable, due to the high costs (Waring & Bishop, 2010). By reducing variability, through the implementation of lean, the buffers can be reduced. An example of a successful cross-departmental collaboration is provided by Drupsteen et al. (2012) and these authors concluded that cross-departmental planning reduces the variability and increases performance. Variability reduction is another motivation for this research as variability can be reduced when departments collaborate.
The preceding argumentation leads to the following research question: How can a capacity
buffer be reduced, with which a shared resource has to cope? Because of the explorative
nature of this question, a single case study will be conducted at one department which provides a shared resource, being physiotherapist capacity, for many patient flows in a regional hospital. This department has already concluded that it ‘spoiled’ high amounts of capacity due to unforeseen variability caused by other disciplines leading to for instance an inefficient planning. Therefore the physiotherapy department does not realize their desired performance.
The current focus of lean in healthcare is more on departmental level. This research contributes to theory by researching a shared resource about how variability, buffers and waste arise, which is a gap in literature. Practitioners will develop a better understanding about which factors cause variability and waste in a shared resource and how these factors can be taken into account to decrease capacity buffers as much as possible.
2. Theoretical background
The following section starts with a description of lean thinking. It continues with the characteristics of lean in healthcare. Followed up by an explanation of the role of waste, variability and buffers. This section ends with a research model.
2.1 Lean
Lean is a philosophy based on continuous improvement developed in a production system at Toyota. The core concept is the continuous improvement of processes by removing non-value adding steps (waste) and to focus on value-adding steps (Burgess & Radnor, 2013; Joosten, Bongers, & Janssen, 2009). Radnor (2011) stated that the general perception of lean is that it is concerned with waste reduction. However, this might be a limited understanding of lean according to Hines, Holweg, & Rich (2004). They found that there are two ways to increase value, (1) by reducing waste that leads to a reduction of cost of the product or service, or (2) by additional features or services without increasing cost for product or service. Non-value adding activities (Japanese: Muda) are causing delays in the process or require extra resources (Burgess & Radnor, 2013). Though, Muda is one of the three concepts of lean, the other two are Muri and Mura. Muri relates to ‘excessive strain’, and argues for good working conditions. Mura can be translated as ‘unevenness’ and relates to variability in a process (Burgess & Radnor, 2013; Joosten et al., 2009; Radnor, 2011). Variability is an important source of waste which is often neglected in lean (Hopp & Spearman, 2004). This research focuses on Muda and Mura. In addition, Shah & Ward (2007) state that reducing waste and variability should be done at the same time to reduce all waste. Additionally, this has been summarized in the continual improvement cycle which eventually reduces capacity buffers by reducing obvious waste sources and variability in a four-step-cycle (Hopp, 2011). Different studies have been done on the identification of different types of waste and variability in healthcare organizations (Fredendall, Craig, Fowler, & Damali, 2009; Litvak & Long, 2000), which will be discussed in the following section.
2.2 Lean in healthcare
However, the cases discussed in Mazzocato et al. (2010) are focussed on implementing simple tools and techniques through small projects and lack an interdepartmental view (Brandao de Souza, 2009; Radnor, 2011). For a better realization of the benefits of lean, healthcare organizations need to extend the organizational view to a holistic view, work across departments and create a continuous improvement cycle (Hopp, 2011; Joosten et al., 2009; Mazzocato et al., 2010; Radnor, 2011). As mentioned before, a more interdepartmental view is necessary to avoid a limited application of lean. Traditionally the focus of lean applications is on one department. By improving processes of a department that delivers a resource to multiple departments, like a shared resource, a more holistic view is intended to achieve. Because the multiple departments have to coordinate their activities with the shared resource (Curatolo, Lamouri, Huet, & Rieutord, 2014; Radnor & Osborne, 2013; Sewell, 1997).
2.3 Waste
Lean provides a way to improve processes by approaching waste inherent in all of the processes. Adapted from Toyota’s seven wastes, the following wastes could be relevant to healthcare: (1) staff walking to the other end of a ward to pick up notes (transport), (2) unneeded stocks and supplies (inventory), (3) unnecessary staff movement looking for paperwork (motion), (4) delays in diagnosis and treatment (waiting), (5) requesting unnecessary tests (over production), (6) overworked staff (over burden) and (7) medication errors (defects) (Fillingham, 2007; Radnor, 2011). The general perception of lean is that it is primarily concerned with waste reduction. Waste reduction is the first of four steps contributing to the continual improvement cycle (Hopp, 2011). This research takes it to the next steps of the continual improvement cycle by investigating both waste and variability of a shared resource.
2.4 Variability
Altogether natural variability is a result of differences in patients and medical practitioners (Joosten et al., 2009). According to Litvak & Long (2000) natural variability should be managed optimally by identifying the different sources and apply operations research methodologies like queuing theory. On the other hand, artificial variability is related to controllable and non-random factors and exists because of the way a system is designed (Litvak & Long, 2000). To illustrate, artificial variability can relate to a mismatch between necessary equipment and provided equipment, or more practically, a physician arriving too late for an appointment. In general, this kind of variability should finally be eliminated. However, the first step is the identification of the types of variability in the healthcare organizations (Andersen, Røvik, & Ingebrigtsen, 2014; Litvak & Long, 2000).
2.5 Buffers
A decrease in quality buffers could increase the capacity buffer. Referring to the research question, the capacity buffer should be minimized, without affecting the quality or time buffers.
2.6 Pull system
One way to expose buffers is by implementing a pull system, which is one of the principles of lean. A pull system can be defined as “one that explicitly limits the amount of work in process that can be in the system”. The opposite of a pull system is a push system, which is defined as “one that has no explicit limit on the amount of work in process that can be in the system” (Hopp & Spearman, 2004). A pull system is suggested because it is more efficient than a push system. An example of a push system is that patients arrive according to their scheduled appointment times, without having information about the status of their appointment (e.g. a physician is running late and the patient have to wait). In a pull system, the patient receives a call when the physician is behind schedule. This allows the patient to delay their arrival and hence reduce the time spend in the waiting room (Hopp, 2011).
2.7 Research focus
This study will focus on the identification of variability, buffers and waste to reduce the capacity buffer. It will answer the question how a capacity buffer can be reduced, with which a shared resource has to cope. The first step is to find out which activities encounter idle time in the process of a shared resource. The second step is to define these activities to a specific type of variability, buffer or waste. The last step is about knowing which activities put pressure on the capacity buffer, because they should be reduced.
Figure 2.1 Research model
Capacity buffer Lean Level of variability / level of waste
3. Methodology
In this section the methodology, applied in this research, will be described. This research is a single case study. A case study is a methodology for an objective, in-depth examination of a contemporary phenomenon where the investigator has little control over events (Yin, 2009). The aim of this research is to enhance knowledge on the use of shared resources and to explain how a shared resource can reduce waste, variability, capacity and time buffers by the use of lean. As this knowledge is currently underdeveloped, an extension of theory is required. According to Eisenhardt (1989), a case study is most suited for extending theory. Several purposes for a case study are the discovery and description, mapping and relationship building and theory validation and theory extension (Handfield & Melnyk, 1998). One of the purposes of this case study is to better understand and to describe a phenomenon. In this case, the relation of variability, buffers use of a shared resource in relation to variability and waste. Furthermore, while implementing lean in shared resources is an element in healthcare organizations, a lack in research can be found in the detailed case studies that describe the fine texture of how lean reduces variability in these settings (Poksinska, 2010). Therefore, a single case has been conducted, in order to gain in-depth understanding of the use of the shared resource and how this shared resource is influenced by applying lean. One of the limitations of a single case study is the generalizability of the conclusions, however this outweighs the advantage of an in-depth research.
3.1 Introducing the hospital
The hospital that was participating in this research is a university hospital with a total of 31 Result Responsible Units (RRUs) and a total amount of 643 beds. An RRU contains different related departments, organized around patient care and patient services. Across the RRU’s, several lean projects are applied, all supervised by black and green belts1. They provide data for the lean projects. On department level, orange belts are educated with the basic principles of lean. They take care of team driven issues.
3.2 Sample selection
The focus of this case study is on the physiotherapy department. This department can be classified as a shared resource for other departments as the physiotherapy capacity is called upon by different patient flows.
First, need to be understood which specific resources are shared. Physiotherapy has an outpatient clinic and serves several hospital wards by offering the physiotherapy capacity; examples are neurology, orthopaedics or internal medicine. A motivation for studying both an outpatient clinic and a ward is that these are organized differently, which could affect the outcomes. In the outpatient clinic, patients have an appointment, so the arrival rate is less variable compared to hospital wards. At the wards, the physiotherapists are dependent on multiple disciplines, which affects their activities.
Every morning, patients of thirteen different departments have to be visited by physiotherapists. The amount of patients differs per department, but in total there are around 150 patients a day. The procedures for patients are characterized per department. In this case, orthopaedics and general surgery is studied, because these departments require the highest capacity from physiotherapy. It is a small department, with 10-15 Full-Time Equivalent (FTE) in the morning and 5-10 FTE. in the afternoon. In the afternoon, the physiotherapist works in the outpatient clinic. The department is under high pressure to examine all clinical patients in the morning, due to rapid recovery targets set by the management, which make it complex to deliver high quality.
3.3 Data collection
The data is collected at the two selected wards. Multiple sources of data were used including semi-structured interviews, observations and desk research. Semi-structured interviews have been conducted in Dutch with unit managers2 and care coordinators3 from physiotherapy, general surgery and orthopaedics. The unit managers have a more organizational view, while the care coordinators have a more practical view on the activities of a physiotherapist. An overview of the interviews can be found in table 3.1. For the reliability and validity of the interviews, an interview protocol (Appendix A) is conducted (Karlsson, 2009). With permission of the interviewee, the interview was recorded on a voice recorder. The interviews lasted between thirty minutes and one hour and were recorded. The recordings are used to write the transcript of the interview and to ensure completeness. All interviewees got the opportunity to give final feedback on the total transcript of their interviews.
Interview Ward Interviewee function
1. General surgery Unit Manager
2. General surgery Care coordinator
3. Orthopaedics Unit manager
4. Orthopaedics Care coordinator
5. Physiotherapy Unit manager
Table 3.1 Overview interviewees
Beside interviews, data was collected through observations and by on-site visits. During the observations different physiotherapists were followed to get insights in their work routine, in regular processes and activities to identify underlying causes of waste, variability and buffers. As preparation for the observations a list of questions was conducted (Appendix B) based on the preliminary interviews with the physiotherapists and interview with the unit manager of the physiotherapy department. During the observations, field notes were taken and coded afterwards. Furthermore, the observations clarified the reasons for employees to make certain choices that can result in waste, variability and buffers. Thereby, regular on-site visits raised the involvement of employees, which have led to in-depth information from the employees. As described before, some employees are educated as orange belts. Collaboration with the orange belts was useful, because they were educated on departmental level and thus could help to get insights into the shared resource.
The final data source used in this research is archival data. Physiotherapy is implementing lean, therefore different studies have been conducted. These reports and presentations have been examined to gain more insight into the implications which are already identified.
Together, these different types of data collection methods facilitated in achieving construct validity which is essential for theory building, as theory validation increases when multiple sources and methods are used (Karlsson, 2009). These data collection methods were chosen to understand waste, variability, buffers and characteristics of lean in a shared resource and provided insights in the day-to-day problems that physiotherapists encounter.
3.4 Data analysis
All transcripts were analyzed on useful quotes, which count as interpretive codes (Huberman & Miles, 2002). After that, deductive reasoning has been applied to the quotes, so relevant data could be distinguished from irrelevant data.
All quotes were provided with comments on their influence on variability, waste and buffers and its influence on the physiotherapist’ activities. By recognizing patterns and their underlying argument results can be induced.
Variability Example
Arrival variability The arrival of patients is different. One week more patients are received compared to another week. This could be due to the planning of the OR, or seasonal accidents.
Clinical variability The degree of patient’s disease differs per patient, one patient needs more care compared another patient.
Professional variability
The physiotherapists vary in way of working, quality, speed and effectivity. One physiotherapist could do the activity quicker compared to another physiotherapist.
Artificial variability The organization of processes. An example is the agreement that all patients should be sitting in their chair before 09.00 a.m., however nurses cannot make this due to busy schedules.
Table 3.2 Overview variability
Buffers Example
Time buffer The patient has to wait before he receives his care.
Capacity buffer Extra capacity is needed to do the work, for instance an extra physiotherapist.
Quality buffer One consult differs in amount of activities, compared to the same consult at another patient.
Table 3.3 Overview buffer
Variables Explanation
Quality of care The quality of care, given by physiotherapist to a patient, could be influenced negatively or positively by the variability or buffer. Time spend on
patient
The time spend on a patient by a physiotherapist could be influenced negatively or positively by the variability or buffer.
Flow of patients The activities of a physiotherapist can influence the flow of patients negatively or positively due to occurred variability.
Predictability Interesting to know, according to the interviewees, if the occurred variability or buffer is predictable.
Way of handling Multiple disciplines have to collaborate for a smoothly patient flow. Therefore it is interesting to know how the other disciplines handle variability and buffers.
4. Results
In the following section an in-depth analysis will evaluate and discuss the activities of the physiotherapy department. The focus is on describing the idle time that occurs in the activities of a physiotherapist and to define it in a type of waste, variability or buffer. The first section will give a description of the case department. The second section will discuss activities on daily basis. The third section will discuss all activities that cannot be done within a day.
4.1 Case description physiotherapy department
In general, a shared resource can be for example a MRI scan, where different departments use the same resource. The explored shared resource is the physiotherapy department of the Martini hospital. Physiotherapists are divided over thirteen different wards at the Martini hospital and therefore a capacity resource will be shared. They are asking for extra capacity to keep up with the demand from the wards. The question is whether this is necessary?
To give an idea about a regular day of the physiotherapist, an impression will be given in table 4.1. A day from the physiotherapist starts at 08.00 a.m. and last until 05.00 p.m. During the observations it became apparent that the first capacity buffer could already be recognized, as some physiotherapists start earlier than the mandatory 08.00 a.m.
Time Activity
08.00 – 08.15 a.m. Preparing activities for wards 08.15 – 08.30 a.m. Moving to wards
08.30 – 10.30 a.m. Consults at wards 10.30 – 10.45 a.m. Coffee break 10.45 – 11.30 a.m. Consults at wards
11.30 – 11.45 a.m. Moving back to own department
11.45 – 12.45 a.m. Moving to outpatient clinic, based on planned consults 12.45 – 01.30 p.m. Break
01.30 – 05.00 p.m. Consults at wards or outpatient clinic
Table 4.1 Daily schedule physiotherapist
As mentioned before, this case focuses on the outpatient clinic and two wards; these are orthopaedics and internal surgery (further referred to as ward A and ward B). Ward A has a bed capacity of 28 beds, ward B respectively 32 beds.
Furthermore, four patient groups are defined as (1) rapid recoveries, (2) completion consults, (3) follow-up consults and (4) trauma consults. Table 4.2 provides an overview on the different patient roles, this distinction is important because physiotherapists have different roles in each patient group.
Patient group Activity Role physiotherapist
Rapid recovery consult
Clearly defined care pathway, patient should be mobilised 4 hours after surgery.
Makes decision to send the patient home or not. Completion
consult
Patient has been discharged by the physician and needs a referral letter from the physiotherapist.
Writes referral letter if needed. Follow-up
consults
Patients should be visited to make progress in their recovery.
Taking care of recovery. Trauma
consults
Physiotherapist visit trauma patient. Taking care of recovery.
Table 4.2 Patient group overview
Especially the rapid recovery patients put pressure on the capacity of the physiotherapy department, because the physiotherapist has a decision role in this care pathway. Another patient group, which puts pressure on the capacity, are the patients that stay around for 2/3 days, because they need a consult twice a day dependent on their care pathway.
4.2 Daily activities
4.2.1 Stage 1: Before the consult
From the moment that the physiotherapist arrives at ward A or B, the preparation of the consults starts. Physiotherapists bring their own laptop to the wards, where they have their own office. First, they receive a list of bed occupancy and an overview of patients per room. Both wards are defined as planned wards, because the patient arrival is known for at least 80%, which can be seen in the system one week earlier. The arrival variability is low, with a few peaks due to trauma patients, planning of the operating room and the amount of operations during a day.
The first step from a physiotherapist is to structure and prioritize the patients into the patient groups. During this process the structure slips away due to the attendance of multiple physiotherapists. No structure has been recognized in the co-operation between physiotherapists. Therefore, it could happen that physiotherapist 2 has picked up a patient, who was assigned to physiotherapist 1. Accidently two physiotherapists were preparing the same patient. Another point that was observed is the way of working in the administration system. Every physiotherapist works differently, due to the fact that the IT system has been implemented three months ago and the physiotherapist did not make rules or routines for this. Ambiguities concerning the workload allocation, unstructured preparing of patients, vagueness concerning co-operations between the physiotherapists and unstructured way of working in the administration system all relate to rules and routines, which could be considered as artificial variability. Furthermore, during the ‘administration hour’ at ward A, nurses arrive separately at the office for a quick update about the patients. This update contains information about the status of the patient. However, the same information is provided in the new-implemented system, which just has been read by the physiotherapist. According to an unit manager the quick update is out-dated: ‘You can read everything in the
system, take your benefit and adapt this in your activities’.
This has been recognized by all unit managers: ‘if we work with our regular physiotherapist
everything works smoothly, quick and efficient. We do not hear any peculiarities´. 4.2.2 Stage 2: During the consults
After preparation, it is time for the physiotherapist to start seeing patients. This is around 09.00 – 09.30 a.m., which depends on the amount of patients (arrival variability). A quick round on the ward resulted in a short list of available patients. There are several reasons for not being able to visit the patient, like the patient gets washed, the patient is tired from washing, the patient is not at the ward because by example he is at the x-ray. This is also noted during the observations and by all the interviewees. Different agreements have been made with the nurses to reduce this artificial variability. One of these agreements is that the patient should be washed and sit in his chair at 09.00 a.m., however, nurses simply cannot make it due to their strict schedule in the morning.
Every physiotherapist starts with his prepared patients. Arriving at the bed of the patient, the physiotherapist does a quick scan for the kind of materials the patient needs. At ward A, a search starts for the right materials somewhere on the ward. Some locations are fixed for materials, however it is not structured. Unnecessary moving of physiotherapists searching for the right materials could be considered as transport waste. Derived from the desk research, most of the materials were not present around 11:00 a.m., which sounds recognizable because materials are continuously being hauled back and forth to different places on the ward.
A big effect on the time of a consult is clinical variability. ´The intensity of care a patient
needs can vary” according to the care coordinators. Clinical variability can be recognized
during a recovery stage (first consult, right after surgery consult or recovering for a couple days), the degree of patient’s disease, age and way of living (one-floor or two-floor). As a result of the level of care that the patient needs, a consult could take a physiotherapist around 20 minutes, where most of the consults take about 10 minutes.
Around 10.30 a.m., during the coffee break, the physiotherapists check how many consults they have finished. Based on their experience, they decide whether they need extra capacity for the remaining consults or not. This results in a flexible capacity buffer, as physiotherapist, which were allocated to other wards, will receive a call to transfer to ward A or B. Interestingly, in this way, the physiotherapists have already developed a pull system in their way of working. However, noted from the observations and interviews, a disadvantage from this system is the fact that before the coffee break, physiotherapists do their job in a relaxed way. After the coffee break, time is limited and pressure increases to finish all consults, so every physiotherapist works a lot faster (example of a quality buffer). A second way to finish all consults is by taking no coffee/lunch breaks (extra capacity), which is undesired capacity.
4.2.3 Stage 3: After the consult
Finishing up the consult, the only thing that remains is the administration. Walking back and forth to their computers between consults can be seen as waste. One physiotherapist will do it after each consult, however some physiotherapists finish their administration after five consults. Working differently is a characteristic of professional variability. Another example of professional variability is to determine the next step in the consult, which influences the flow of patients. Some regulation has been provided to know when a patient is healthy enough. However, these regulations do not provide enough information for a physiotherapist. So they need to base the next step on their experience. Due to objective and subjective perception, a patient could be considered healthy, while not having the confidence he can do everything at home. One physiotherapist can decide to send the patient home in contrast to another physiotherapist who decides not to do so.
4.3 Weekly activities
Weekly based activities are activities, which cannot be assigned to a regular day. At a department where multiple disciplines have to cooperate for a patient, a kind of coordination meeting is expected. Once a week, all disciplines of a ward are gathering for a multidisciplinary meeting. These disciplines talk about the status of the patients. It could be argued that this meeting could be used for talking about coordination as well. However, it is not possible due to time limitations and strict rules for this meeting. The unit managers admitted that in the last couple of years no formal meeting was conducted about the coordination of the different activities. According to them, the physiotherapist stays up-to-date by informal talks on the ward. If the physiotherapist suggested an improvement for the whole department, the care coordinators placed it in the newsletter of that ward. However, this results in a short moment attention to the problem, but on the long term the department will forget the problem and falls back in old practices. Unstructured coordination is a form of artificial variability. All interviewees from the wards stated that, despite the fact that they do not have any formal meetings about the coordination of the shared resource on their wards, it does not influence the quality of care.
4.4 Overview
Table 4.3 provides an overview of the overall conclusion on how each variability influences the variable. Overall influence on the ward Influence on quality of care Influence on time spend on patient Influence on patient flow
Predictable How to handle
Arrival variability
About 80% planned patients on both wards.
The unit managers stated that this is a negative influence. The care
coordinators did not have an opinion about this point.
It only
influenced when too much people need a consult on a ward at the end of a shift.
The higher the demand of consults, the more it influences the patient flow negatively. Planning made it predictable for 80%. The other 20% is about trauma, which cannot be predicted. By making use of planning and adapt changes over the day in the daily planning. Clinical variability Due to higher complexity of care on ward B, the experienced clinical variability is much higher compared to ward A. Overall not an influence, the physiotherapists show a lot of flexibility in the need of care asked by the patient. The content of a consult could fluctuate by clinical variability, not the time spend. The variation of time spend is too small.
The higher the variability, the more it has a negative influence on the patient flow. From experience and/or the intake consultation it becomes predictable. Continuous coordination between nurse and physiotherapist. Professional variability Both wards agreed that the experienced professional variability is high. However, the unit manager from the
physiotherapist did not agreed to this. The experienced physiotherapist on a ward is efficient and quicker. A less experienced physiotherapist takes more time of the patient.
Does not influence the patient flow due to the fact that the consults will get done.
If someone else than the regular physiotherapist will do the consult at the ward, the professional variability becomes higher. So this is predictable. Planning of the physiotherapist. If it is a busy morning, do not use a less experienced physiotherapist on that ward. Artificial variability According to all interviewees, artificial variability has the highest impact on the performance of all disciplines. Good collaboration has a positive effect on the quality of care. Artificial variability is about the collaboration, so this influences the quality. Depending on the amount of variability that occurs, consults could be done quicker due to time pressure.
The higher the variability, the more complex it becomes to do the consults. So this has a negative effect on the patient flow. From different studies, all interviewees concluded different sources of artificial variability, which made it predictable. All disciplines should work together more for a more efficient patient flow.
Table 4.3 Overview results interviews
4.4 Capacity buffer
The aim of this research is to reduce the capacity buffer, an overview of the capacity buffer during a day is provided in figure 4.1.
07.00am 08.00am 09.00am 10.00am 11.00am 12.00pm 01.00pm 02.00pm 03.00pm 04.00pm 05.00pm 06.00pm (3) A
t co ffee
bre ak an u
pdate abo ut w ork load (1) S tart ing befo re w orktim e (4) P ressure on cap
acity, a s they sh ould m ove to outp atie nt-‐c linic (5) N o lu nch bre ak t o fin ish all cons ults (8) O verw ork (6) M idd ay, capa city is divid ed o ver outp atie nt clinic and the wards (7) C loser to end o f the d ay, pressu
re is get ting hig her to finis h all con sults (2) S pille d ca
pacity a s th ey can not do their con sults ye t
Figure 4.1 Overview capacity buffer
5. Discussion
This study could be considered as a single case study that examines the capacity buffer of a shared resource. The instruments of lean are used to describe the idle time that occurred in the activities of a physiotherapist. Although the studied wards and physiotherapist are different in their nature, interesting findings were derived from this study.
The conclusion of this research is that the focus of physiotherapists is on providing care to the patients. In the activities of the physiotherapist, variability and waste occur. However, the focus is on getting the job done and helping the patient to recover as soon as possible. Even if this reduces personal time of the physiotherapist. Another interesting point, like Poksinska, (2010) already stated, is the fact that the physiotherapy department works as a silo. This case study confirmed that a shared resource, despite the fact that they have to coordinate their activities on other wards, works as a silo as well. First, all involved departments did improvement projects on departmental level (Brandao de Souza, 2009; Burgess & Radnor, 2013; Mazzocato et al., 2010) without including other departments. Second, in the interaction between the different departments who collaborate, only quick updates could be discovered. The last official meeting about the collaboration between the shared resource and the disciplines acting on the ward was years ago, according to all interviewees.
5.1 Presence of variability
An interesting point regards the professional variability. According to the theoretical section, professional variability was defined as natural variability, because medical practitioners capabilities are not the same and therefore not in their ability to provide the same care equally. However, the physiotherapy department increases this form of variability by choices made by them. Physiotherapists are divided over three areas and have a regular ward they visited most of the time. As a result, they are efficient and quick at their own ward but inflexible at other wards. Therefore, a trade-off between flexibility and speed/efficiency occurs.
Artificial variability influences the activities of the physiotherapists the most. Literature concluded that this variability should be reduced (Litvak & Long, 2000). Unstructured ways of working in the system by the physiotherapists or how to co-operate when multiple physiotherapists are assigned to the same ward could be easily solved. However, some sources of artificial variability cannot be addressed in a shared resource. For example the regulations about patients should be sitting in their chair at 09.00 a.m. This is simply not possible due to the busy schedule of the nurses. Another example is that the patient is not available due to other disciplines who are treating the patient at that moment. Regulations and mutual agreements are seen as artificial. However every department works in a silo, and performs their own activities without aligning it with the physiotherapists. What can be learned from this case is that a shared resource has no control over some sources of artificial variability, because it is controlled by other disciplines.
It should be noted that literature argues that (Brandao de Souza, 2009; Burgess & Radnor, 2013; Litvak & Long, 2000) variability should be managed in order to reach an efficient department. This case study showed complications, due to the fact that most variability occurs by activities of other disciplines, which the shared resource cannot influence.
5.2 Presence of buffers
This way of buffering could be defined as a pull system (Hopp & Spearman, 2004), which is one of the principles of lean. However, a disadvantage is that a physiotherapist works relaxed in the morning, because they feel no pressure until the coffee break. One could argue, that if the arrival variability is stable, a calculation could be made to plan total consults for a fixed number of therapists. A trade-off occurs between fixed or flexible capacity.
Furthermore, physiotherapists use the mechanism of a time buffer on both wards. According to Hopp & Spearman (2004), a time buffer refers to the waiting time of a patient. Fact is, in contrast to literature, patients do not wait for their consult. It is the physiotherapist that waits when the patient is available. What also has been noticed is the fact that the physiotherapist makes use of the length of time that a patient is in the hospital. If the physiotherapist or the patient is not available for a consult, the physiotherapist decides to move the consult to another part of the day. This is possible because the patient will be there anyway. Clinical variability influences this mechanism as well. The higher clinical variability will be, the longer the patient will be in the hospital that result in more time for the therapist to do the consult. If the therapist has a strict schedule, their flexibility depends on the consult patient group. All patients will be prioritized in line with the patient groups, as a result they will erase the follow-up consults. Previously mentioned examples give a new insight of a time buffer and counter expectation from literature. As the case uses a time mechanism where patients do not have to wait.
Last type of buffer is the quality buffer. Suggested by Hopp et al. (2007), the quality of care given to a patient fluctuates as a result from variability. All interviewees concluded that the physiotherapist always gave the same treatment during a consult. However, the time spend on a patient could fluctuate, which by example could be done by giving less broad information or shorten the walking distance (obligatory during a consult). Therefore, this quality buffer could be renamed to ‘processing time buffer’ as it does not influence the quality of care but adjusts the time spend on a patient. So, when the pressure on physiotherapist increases, the processing time of a consult could fluctuate.
5.3 Presence of waste
First, the physiotherapist has to walk a lot to find the right materials for each consult, which is a type of transport waste. Second, the physiotherapist has to walk back and forth to a computer looking for paperwork or the administration, which is a form of motion. Lastly, the interviewees stated that it could happen that the patient needs to wait for a referral letter or a final check before they are discharged, because the physiotherapist is simply not working at that moment. Also stated by the interviewees is the fact that “a hospital become a 24 hour
company, however the physiotherapists are still working from 08.00 a.m. till 05.00 p.m.”. It
occurs that patients need treatment in the afternoon or in the weekends. From a patients’ view, delays in treatment could be seen as waiting waste.
5.4 Reducing the capacity buffer
The research question was aimed to find out how a capacity buffer of a shared resource could be reduced with the use of lean instruments. From the research can be concluded that the current capacity is enough to match supply with demand, which is confirmed by the unit manager ‘If everything runs smoothly, you cannot say that you need a lot more
physiotherapists, if you compare the amount of patients to the available time, it should be possible. However, this only happens if you exactly start at eight o'clock and finish at twelve o'clock’. However, due to disruptions in the activities of the physiotherapist, pressure on the
capacity buffer increases. Some clever ways to deal with the pressure is by using a time buffer mechanism, a pull system and flexibility of the patient groups. However, if demand increases or if the patient groups reduce to rapid recovery groups, the physiotherapist becomes important for all patient groups. In this way, the pressure increases on the capacity and it will become difficult to match supply and demand.
As stated in the introduction, departments are not naturally willing to coordinate their activities with other departments involved in the same care process (Drupsteen et al., 2012). The expectation was that a shared resource has to collaborate with the disciplines because they are dependent on each other. However, this case study confirmed that they are not coordinating their activities and that they work in a silo as well.
A way to coordinate the shared resource is by the theory of Thompson (1967) that consists of three approaches, namely standardization, plan and mutual adjustment. Different activities of the physiotherapy department expect different approaches for achieving coordination.
First, coordination by standardization involves the establishment of routines and rules. Every physiotherapist uses different ways to work with the IT system. By establishing a standardized way of working every physiotherapist works the same in the administration, which reduces the professional variability.
Second, coordination by plan involves the establishment of schedules. The results indicated that most of the pressure on the capacity buffer is in the morning. During the morning, most of the variability occurs in a natural way and cannot be reduced. Artificial variability occurs between 08.15-09.30 a.m., because other disciplines (e.g. nurses, physicians) are more important on that same moment. A possible solution is to change the schedule of the physiotherapist, so the physiotherapist arrives after the peak. In this way the physiotherapy department manages the variability optimally by reducing the chance of idle time, instead of reducing the variability.
Shah & Ward (2007) stated that the reduction of variability and the reduction of waste are intertwined. This study showed the opposite. Variability has been hard to control. However, waste could be reduced to decrease the chance of idle time. Therefore, waste has a direct influence on the performance of the capacity buffer. Using laptops/tablets and making someone responsible for the materials on the wards, could reduce motion and transport waste. After each consult, the physiotherapist should bring the materials back to the inventory. Another way could be that someone collect all materials after each shift and store them. Combining all results and comparing it with the initial research model (figure 5.1), it becomes clear that a shared resource rarely has influence on variability. Although, parts of variability occurred by choices the physiotherapy department made by themselves and have a direct influence on the capacity buffer. The results showed that variability has to be buffered, as also stated by Hopp & Spearman (2004). This results in trade-offs between the capacity buffer and the time - or quality buffer. Furthermore, waste has a direct influence on the presence of idle time. An increase in idle time results in a higher capacity buffer. By reducing waste, the capacity buffer could be reduced as well.
Time buffer / Quality buffer Capacity buffer Level of waste (Muda) Lean Level of variability (Mura) -‐ + -‐ + + +
6. Conclusions
The aim of this research was to determine how lean instruments could reduce the capacity buffer of a shared resource. According to the literature, a more holistic view was necessary to benefit from lean in healthcare organizations. Therefore, the idea was to identify variability on an interdepartmental approach, as this should be improving performance and decrease the capacity buffer. A shared resource is used as the capacity of that resource is called upon by different patient flows. At the case department used in this research, many sources of variability were found. However, in this case, reducing all variability could only be done in collaboration with other departments. An example is the patient flow of rapid recovery patients, were all procedures are harmonized.
6.1 Theoretical implications
The results of this case study provide some useful thoughts for theory. According to literature a more interdepartmental approach was needed to reduce variability. The expectation was that a shared resource coordinates their activities more with other disciplines and so reaches an interdepartmental approach. However, the shared resource works as a silo as well and is not willing to coordinate their activities with other disciplines. This research showed that it is hard to control variability on departmental level, as variability of a shared resource involves multiple disciplines. Within this research, the most frequent buffers to cope with the variability are time, capacity and quality buffer. Trade-offs occur, as reducing the time or quality buffer will increase the capacity buffer. Counter to expectations on literature, time buffers are not about the patient, but about the capacity resource. Moreover, the quality buffer, which is suggested by Hopp et al. (2007), has been identified as well. However, ‘processing time buffer’ would be more suitable, where the processing time of a treatment fluctuates depending on the pressure of the moment.
6.2 Managerial implications
Another interesting trade-off with managerial relevance is the fact that a pull system works in a shared resource. Flexibility is an advantage of pooling, however a disadvantage could be that employees work more relaxed because they do not feel pressure on the capacity during a day.
6.3 Limitations and future research
Despite the fact that it was tried to take a more holistic view, as advised by literature, this case study does not provide a holistic view. This is due to the fact that the shared resource is acting like a silo on the different wards, without aligning their activities with other disciplines. Thereby, no formal meetings are conducted between the departments, which should collaborate. In this way, departments are not aware of each other’s problems and issues. The second limitation lies in the methodology. As this is an explorative study, a single case study is suitable. For generalizing the outcomes, more cases should have been used. However, due to time limitations it was not be possible to extend the number of cases.
A third limitation, which was recognized in this case study, is the fact that the studied department has been working with lean for over a year. The interviewees understand the basic principles, but some had problems identifying waste and variability in the patient flows. A consequence of this lack in lean knowledge was that the researcher provided information and examples about lean and this way the interviewee was not objective in their answers given. Although this research provided some interesting insights, further research is needed to get a more holistic view. It is suggested to research patient flows where different departments are forced to collaborate, for instance a rapid recovery flow. Further research could test whether the delivered trade-offs and conclusions of this research hold in other shared resources and hospitals. Another interesting point, which could be an implication for further research, is to increase the interaction between the disciplines by setting up multidisciplinary teams on the wards. Part of this team could be the physiotherapy capacity, which is no longer part of the shared resource. In this way, the disciplines are triggered to communicate and coordinate their activities, which could result in reducing variability to reduce the capacity buffer.
References
Andersen, H., Røvik, K. A. & Ingebrigtsen, T. (2014). Lean thinking in hospitals: is there a cure for the absence of evidence? A systematic review of reviews. BMJ Open, 4, 1–8. Bendell, T. (2006). A review and comparison of six sigma and the lean organisations. The
TQM Magazine, 18(3), 255–262.
Brandao de Souza, L. (2009). Trends and approaches in lean healthcare. Leadership in Health
Services, 22(2), 121–139.
Burgess, N. & Radnor, Z. (2013). Evaluating Lean in healthcare. International Journal of
Health Care Quality Assurance, 26(3), 220–35.
Curatolo, N., Lamouri, S., Huet, J. & Rieutord, A. (2014). A critical analysis of lean approach structuring in hospitals. Business Process Management Journal, 23(3), 433–454.
Drupsteen, J., van der Vaart, T., & van Donk, D. P. (2012). Integrative practices in hospitals and their impact on patient flow. International Journal of Operations & Production
Management, 33(7), 919–933.
Eisenhardt, K. M. (1989). Building Theories from Case Study Research. Academy of
Management Review, 14(4), 542-550.
Fillingham, D. (2007). Can lean save lives? Leadership in Health Services (Bradford,
England), 20(4), 231–241.
Fredendall, L. D., Craig, J. B., Fowler, P. J., & Damali, U. (2009). Barriers to swift, even flow in the internal supply chain of perioperative surgical services department: a case study.
Decision Sciences, 40(2), 327–349.
Handfield, R. & Melnyk, S. (1998). The scientific theory-building process: a primer using the case of TQM. Journal of Operations Management, 16, 321–339.
Hines, P., Holweg, M. & Rich, N. (2004). Learning to evolve: A review of contemporary lean thinking. International Journal of Operations & Production Management, 24(10), 994– 1011.
Hopp, W. J. (2011). Supply Chain Science. Long Grove: Waveland Press Inc. Hopp, J. & Spearman, L. (2004). To Pull or Not to Pull: What Is the Question?
Manufacturing & Service Operations Management, 6(2), 133–148.
Hopp, W. J., Iravani, S. M. R., & Yuen, G. Y. (2007). Operations Systems with Discretionary Task Completion. Management Science, 53(1), 61-77.
Hopp, W. J. & Spearman, M. L. (2008). Factory Physics: foundation of manufacturing
Huberman, M., & Miles, M. (2002). The qualitative researcher’s companion. Sage Publications.
Joosten, T., Bongers, I., & Janssen, R. (2009). Application of lean thinking to health care.
International Journal for Quality in Health Care, 21(5), 341–347.
Karlsson, C. (2009). Researching operations management. New York: Routledge.
Litvak, E., & Long, M. C. (2000). Cost and Quality Under Managed Care :, 3(6), 305–312. Mazzocato, P., Savage, C., Brommels, M., Aronsson, H. & Thor, J. (2010). Lean thinking in
healthcare: a realist review of the literature. Quality & Safety in Health Care, 19(5), 376– 82.
Poksinska, B. (2010). The Current State of Lean Implementation in, 19(4), 319–329.
Radnor, Z. (2011). Implementing Lean in Health Care : Making the link between the approach , readiness and sustainability, 2(1), 1–12.
Radnor, Z. J., Holweg, M., & Waring, J. (2012). Lean in healthcare: The unfilled promise?
Social Science and Medicine, 74, 364–371.
Radnor, Z., & Osborne, S. P. (2013). Lean: A failed theory for public services? Public
Management Review, 15, 265–287.
Sewell, N. (1997). Continuous quality improvement in acute health care: creating a holistic and integrated approach. International Journal of Health Care Quality Assurance
Incorporating Leadership in Health Services, 10, 20–26.
Shah, R., & Ward, P. T. (2007). Defining and developing measures of lean production.
Journal of Operations Management, 25, 785–805.
Taylor, A., Taylor, M., & McSweeney, A. (2013). Towards greater understanding of success and survival of lean systems. International Journal of Production Research, 51(March 2015), 37–41.
Thompson, J. D. (1967). Organizations in action: Social science bases of administration. New
York et al.
Van Donk, D. P., & van der Vaart, T. (2005). A case of shared resources, uncertainty and supply chain integration in the process industry. International Journal of Production
Economics, 96, 97–108.
Waring, J. J., & Bishop, S. (2010). Lean healthcare: rhetoric, ritual and resistance. Social
Science & Medicine (1982), 71(7), 1332–40.
Yin, R. K. (2009). Case Study Research: Design and Methods. Essential guide to qualitative
Appendix A. Interview protocol
The following questions have been asked to each interviewee. 1. General questions
• What is your function within the hospital?
• Can you describe the work activities performed on your department? 2. Variability (arrival, clinical, professional and artificial)
• In general, how does this variability influence your daily activities?
• To what extent does this variability influence the quality of care given by a physiotherapist?
• To what extent does this variability influence the time spend on a patient given by a physiotherapist?
• To what extent does this variability influence the flow of patients? • To what extent is this variability predictable for a physiotherapist? • How do you cope with this variability for a physiotherapist?
• Is there something left that you experienced which we have not discussed yet about this variability?
3. Time Buffer
• In general, how do you experience the time buffer on your department? • Which sources are the basis for the need of a time buffer?
• To what extent does a patient have to wait?
• To what extent does the time buffer influence the quality of care given by a physiotherapist?
• To what extent does the time buffer influence the flow of patients? • How can a time buffer be reduced for the physiotherapists?
4. Capacity buffer
• In general, how do you experience the capacity buffer for the shared resource?
• Which sources is the basis for the need of a capacity buffer?
• To what extent does the capacity buffer influence the quality of care given by physiotherapist?
• To what extent does the capacity buffer influence the flow of patients? • How can a capacity buffer be reduced for the physiotherapists?
5. Concluding
• Are there sources of variability that you experienced which we have not discussed yet?
Appendix B. Observation protocol
For the case study several observations have been done to collect data. The following questions were used to structure the observations. These questions were at least asked or observed during the observations.
A. Prepare a consult
• How does a physiotherapist prepare for the daily activities? • How does a physiotherapist plan patients?
• How does a physiotherapist prioritize patients?
• How does a physiotherapist detects s/he knows that he needs help on the ward?
• How does a physiotherapist react when it is very busy on the ward? • How does a physiotherapist react when it is less busy on the ward? • Which materials does a physiotherapist prepare before the consult? • What are the regulations on a ward according to the physiotherapist? B. During the consult
• Is the patient available?
• What happens when the patient is not available?
• Does a physiotherapist have to search for information, materials or people during the consult?
• Does a physiotherapist do the same steps during a treatment? C. After the consult
• How does a physiotherapist work with the administration? • How long does the administration take?
• When does a physiotherapist do the administration? D. Between consults
• What does a physiotherapist do between consults?
• Does a physiotherapist have the most efficient walking route to patients? E. Other questions
• What does a physiotherapist do when s/he has been asked for a trauma patient?
• What does a physiotherapist do when the amount of consults decreases? • How many times does a physiotherapist checks the system on new
information?
