Assessing overall resource effectiveness of nursing wards
Master Thesis
MSc Technology & Operations Management
University of Groningen, Faculty of Economics and Business
July 3, 2017
Tristan Reinsma - S2054175 t.reinsma.1@student.rug.nl
Supervisor University dr. ir. D.J. van der Zee
Co-assessor University dr. ir. W.H.M. Alsem
Supervisor field of study W. van Dort
L. Lolkema
University Medical Centre Groningen
Abstract
Due to decreasing resources and increasing patient admissions, waiting times at hospitals are growing longer. Their nursing wards are under mounting pressure to increase effectiveness in order to cope with these changes in the health care sector. To do so, measurement and root cause analysis of the current situation is required. The clear reality in the industry is that the great majority of health care providers fail to assess patients’ flow with respect to their length of stay and the use of beds. Moreover, they lack the tools to do so. This research will develop and test a framework for measuring and assessing overall resource effectiveness of nursing wards.
To address this problem, a design science method is used to develop the overall resource effectiveness (ORE) framework. The nursing ward of the surgeon department at the University Medical Centre Groningen (UMCG) servers as a case example.
The ORE framework builds on a literature review on existing frameworks from the manufacturing and logistics industry, related literature in the health domain and observations in practice on nursing ward set-up and operations. Where current frameworks focus on the effectiveness of equipment in manufacturing or trucks in transportation, this framework will focus on the effective use of beds at nursing wards.
Existing frameworks exist out of the three categories availability, productivity and quality.
Each category has its respective loss factors, which identify root causes for losses in effectiveness. The three categories and their loss factors are redefined and linked to the domain specific characteristics of nursing wards to make it applicable in the healthcare context and measure the effective use of beds.
The ORE framework is designed to systematically measure and assess resource effectiveness of nursing wards and provides starting points for root cause analysis. Implementing the framework allows to use it as a tool for continuous monitoring and control of the most important factors influencing current performance.
Keywords: nursing wards, ORE, performance measurement, resource management, root cause analysis
Preface
This master thesis project is the final step towards the completion of the MSc Technology and Operations Management at the University of Groningen. I would like to express my gratitude towards all who have aided me and made this research to a success.
First, I would like to thank W. van Dort and L. Lolkema from the UMCG for their invested time. Their meetings provided me with valuable insights and all the needed information regarding the workings of their nursing ward. Their advice and involvement proved to be of great help to keep the research going in the right direction. Furthermore, I would like to thank T.J.J. Hoogstins for providing me with the needed historical patient data. Also I would like to thank all of the other personnel that invested their time to help me.
I would also like to thank my supervisor D.J. van der Zee for providing me with feedback and sharing his experience on the process of writing a structured report. His positive approach always motivated me to start writing again and outperform myself. The same goes for my co- assessor W.H.M. Alsem, who I would like to thank for his constructive feedback on earlier drafts. Last, I would like to thank my fellow students of the healthcare thesis theme group for the feedback given during our meetings.
Groningen, June 2017 Tristan Reinsma
Contents
1. Introduction...6
2. Problem statement and research design...8
2.1 Problem background...8
2.2 Research objective...9
2.3 Conceptual model...9
2.4 Research Design...10
2.5 Data Sources...11
3. Theoretical background...12
3.1 Nursing ward terminology...12
3.2 Tools for assessing effectiveness...13
3.3 Overall equipment effectiveness frameworks...13
3.3.1 Overall equipment effectiveness in manufacturing...14
3.3.2 Overall transport effectiveness in logistics...15
3.4 Overall resource effectiveness in healthcare...15
3.5 Summary...16
4. System Description...17
4.1 Overview nursing ward...17
4.2 Process nursing ward...18
4.3 Timeline process...20
4.4 Key resources...21
4.5 Operational offline planning...21
4.5.1 OR planning...21
4.5.2 Ward planning...22
4.6 Operational online planning...23
5. Framework Design...24
5.2 Overall resource effectiveness framework...25
5.3 System elements...25
5.4 Loss factors...26
5.4.1 Availability...26
5.4.2 Productivity...28
5.4.3 Quality...28
6. Case Study Validation...30
6.1 Framework Results...30
6.1.1 Overview overall resource effectiveness...30
6.1.2 Availability...31
6.1.3 Productivity...33
6.1.4 Quality...34
6.2 Recommendations...35
6.2.1 Main recommendations...35
6.2.2 Availability...36
6.2.3 Productivity...36
6.2.4 Quality...37
7. Discussion...38
7.1 Main Findings...38
7.2 Limitations...38
8. Conclusion...40
8.1 Main Conclusion...40
8.2 Future Research...40
References...41
Appendix I: Loss factors overall equipment effectiveness manufacturing...43
Appendix II: Loss factors perspectives overall transportation effectiveness logistics...44
Appendix III: Variation in admissions at C4 ward UMCG...45
1. Introduction
Over the last decade hospitals have been increasingly busy with improving the effectiveness of their operations (Vanberkel et al., 2010). This attention is due to developments in volume and care needs in the health care sector such as increasing expenditures, an aging population, government cuts and the increased patient demand (Vanberkel et al., 2010; Okma et al., 2013). Forced by these developments, hospitals recognize the importance of using their resources more effectively (Lega and DePietro, 2005) and improve their operational performance. However, to improve processes in any field, measurement and evaluation of performance is required. The clear reality in the healthcare industry is that hospitals lack the tools to do so (Porter and Lee, 2013).
The nursing ward of the surgery department at the University Medical Centre Groningen (UMCG) seeks to improve their effectiveness as well. With a growing list of patients who have to undergo surgery, the nursing ward has difficulty to cope with higher inflow of patients. Furthermore, each patients’ care needs tend to differ significantly, causing high variety in healthcare services and the duration of their stay. As a result of this, beds are occupied more often and longer, which causes waiting lists to grow larger. Various causes may explain the current performance of the ward. However, hospitals managers have few tools to systematically analyse the wards effectiveness.
Literature acknowledges these findings at the UMCG. Health care facilities are under mounting pressure to become more effective in providing care (Vanberkel et al., 2010). As a result of longer waiting times and increasing patient admissions, hospital wards are becoming more overcrowded (Bachouch et al., 2012). Although both patients and care providers seem to have accepted waiting is an inevitable part of healthcare provision (Vissers and Beech, 2005), effective use of nursing wards has to be assured (Litvak, 2008) in order to handle rising patient demand (Pencheon, 1998).
Current academic research in the field of healthcare operations provide few tools for measuring and assessing effective use of hospital operations (Shah et al., 2008; Vanberkel et al., 2010; Porter and Lee, 2013). Where tools such as Lean Six Sigma and Total Quality Management focus on reducing variability and improving performance, overall equipment effectiveness (OEE) frameworks are used to measure and assess current performance and provide a root cause analysis. They categorize major loss factors into the three categories
performance, provide the basis for setting improvement priorities and beginning of root cause analysis by providing insights on effectiveness in each category (Pintelon and Muchiri, 2008).
As wards currently have multiple causes for current performance, OEE can benefit wards by systematically measuring and assessing domain specific system characteristics and address them to such loss factors regarding effective use of resources. When loss factors are identified, hospitals can engage in operational improvement activities directed at minimizing those losses. To apply OEE at nursing wards, a redesign of the loss factors of the framework is needed by relating them to nursing ward characteristics.
This research will develop a framework to measure and assess overall resource effectiveness (ORE), of nursing wards. Where current frameworks focus on the effectiveness of manufacturing equipment or transportation trucks, this framework will focus on effective use of ward beds. As staff is planned accordingly, beds are considered to represent associated resources, such as staff and medical equipment, as well. In doing so, it will provide insights in how hospitals can improve effective use of their nursing wards. Therefore, the research objective is:
Develop and test a framework for measuring and assessing overall resource effectiveness of nursing wards.
To design such framework, a design science approach will be adopted as mentioned by Holmström et al. (2009). It consists out of four phases, from which the first three will be implemented: (1) Solution incubation, (2) Solution refinement and (3) Explanation I – Substantive theory. When designed, the framework will be tested by using historical patient data of the year 2016 to assess resource effectiveness at the nursing ward of the UMCG.
This research will contribute academic research in the field of health care by developing a framework on how to measure and assess overall resource effectiveness at nursing wards.
Furthermore, the theory behind the framework is generalizable and therefore can be applied at other nursing wards in the healthcare industry.
This thesis continuous with chapter 2: Problem statement and research design. In chapter 3 the theoretical background of the research will be discussed. Chapter 4 describes the system and its characteristics. In chapter 5 the design of the framework will be presented. Testing and evaluation of the framework will be discussed in chapter 6. Chapter 7 discusses the results and limitations of the thesis. Finally, the conclusion and suggestions for future research can be found in chapter 8.
2. Problem statement and research design
This chapter provides an overview of the content, scope and methodology of this research.
Section 2.1 provides the background of the problem. Next section 2.2 gives an overview of the research objective. Section 2.3 provides the conceptual model to give an overview of which elements are incorporated, where section 2.4 discusses the research design. Finally, section 2.5 gives an overview of used data sources.
2.1 Problem background
Hospital wards are becoming more overcrowded due to waiting times and increasing patient admissions (Bachouch et al., 2012). As a result of this, health care facilities are under mounting pressure to become more effective in providing care (Vanberkel et al., 2010;
Bachouch et al., 2012). Porter and Lee (2013) acknowledge these problems and state that resources at nursing wards have to be used more effectively in order to handle rising demands. However, the multiple sources of variability, such as patient characteristics and treatment uncertainties, harden hospital organisation improvements (Litvak and Long, 2000).
The question is then; how to organize the use of beds at nursing wards in an effective manner, to meet the demands of patient to be served from the same resources (Aronsson et al., 2011).
To improve ward effectiveness, measurements and root cause analysis of the current situation is required (Nachiappan and Anantharam, 2006). The clear reality in the industry is that the great majority of health care providers fail to assess outcomes for patients regarding their length of stay (LOS) and the use of beds (Shah et al., 2008; Porter and Lee, 2013) and currently lack the tools to do so.
At the nursing ward of the surgeon department at the UMCG, a shortage of beds is experienced due to decreasing resources and increasing patient admissions. To cope with this, an assessment of the current situation is needed in order to give an indication of which areas to target for improvement enhancing practices. Currently hospitals have little means for a tool which can be used for a systematic analysis of the wards effectiveness. Overall Equipment Effectiveness (OEE) and Overall Transport Effectiveness (OTE) are tools used in the manufacturing and logistics industry to assess effectiveness of equipment or trucks. By measuring current performance and identifying operation losses, the tools provide starting points for root cause analysis. In order to apply such tools to a healthcare setting, the loss factors of current framework have to be redefined and linked to domain specific system
2.2 Research objective
This research will provide a framework to measure and assess ORE of nursing wards. The research is motivated by an effectiveness problem at a nursing ward at the UMCG, which will serve as a research vehicle. Where current frameworks focus on the effectiveness of equipment or transportation, this framework will focus on effective use of beds, as the capacity of a ward is measured in terms of operational beds. Staff is planned accordingly to handle the patients occupying the number of beds. Therefore, beds are considered to represent associated resources such as staff and medical equipment as well. By applying the framework, insights are obtained on how hospitals can improve effective use of beds at their nursing wards.
Therefore, the research objective is:
Develop and test a framework for measuring and assessing overall resource effectiveness of nursing wards
Essentially, the framework design boils down to a modification and extension of existing frameworks available for manufacturing and logistics systems. When designed, the framework will be tested and evaluated at the nursing ward of the UMCG. As a result of this, hospitals know on what loss factors they should focus in order to improve the effectiveness of their nursing wards.
This research will contribute academic research in the field of health care by developing a framework on how to measure and assess overall resource effectiveness at nursing wards.
Furthermore, it will be shown that the theory behind framework is generalizable and therefore can be applied at other nursing wards in the healthcare industry.
2.3 Conceptual model
Figure 1 provides an overview of the conceptual model. Loss factors regarding availability, productivity and quality effectiveness from current OEE frameworks have to be linked to characteristics of nursing wards beds. By addressing such loss factors to beds, current effectiveness of this resource can be assessed. This will result in an overall resource effectiveness (ORE) framework which can be used to systematically measure and assess the effective use of beds at nursing wards.
Figure 1: Conceptual model
2.4 Research Design
The design science method, as mentioned by Holmström et al. (2009), is used as a guideline.
Other methodologies, such as a case study or survey are not found optimal since those will not develop an artefact which can be used to solve the practical problem. The design science method consists out of four phases from which the first three will be used: (1) Solution incubation, (2) Solution refinement and (3) Explanation I – Substantive. In table 1, all steps are addressed and linked to the corresponding chapter in the thesis. In the first phase, the framework will be designed by assessing existing literature. In phase 2, the framework will be tested and evaluated at the nursing ward of the UMCG, which serves as a research vehicle.
The theory behind the framework will be generalized in phase 3. Following these steps will ensure that the framework will be applicable at the nursing ward at the UMCG and, by generalizing the findings, at other nursing wards as well.
PHASE CHAPTER
1. SOLUTION INCUBATION 5. Framework Design 2. SOLUTION REFINEMENT 5. Framework Design
6. Case Study Validation 3. EXPLANATION I –
SUBSTANTIVE THEORY
7. Discussion 8. Conclusion
Table 1: Phases and corresponding chapters
Phase 1: Solution Incubation
In this phase, an initial framework to measure overall resource effectiveness at nursing wards will be designed. A literature review on the subject has identified the elements of existing frameworks. The loss factors of current frameworks will be applied to nursing ward resources.
The corresponding nursing ward performance indicators which have to be included in the
framework will be determined by doing a literature review on nursing ward metrics and by conducting non-structured interviews with the UMCG staff.
Phase 2: Solution Refinement
This phase exists out of testing and evaluating the framework. It will be an iterative process which will refine the initial solution with the help of experts at the UMCG by doing non- structured interviews. In doings so, the feedback will validate the design. By doing quantitative data analysis, the framework and its outcomes will be tested and evaluated.
Alterations will be made until no significant improvements can be done anymore. The eventual framework will be designed and visualized in Excel. This will ensure easy use and data capture as it provides a highly adaptable and self-explanatory user interface.
Phase 3: Explanation I – Substantive theory
In phase 3 the theoretical relevance of the solution design will be established. It will generalize the findings of the first two phases and demonstrate a theoretical contribution. This involves an examination and evaluation of the designed framework from a theoretical point of view. As a result, the framework can be used at other nursing wards at other hospitals to measure and assess resource effectiveness.
2.5 Data Sources
Data used for the system description will be derived from non-structured interviews with the ward-matron, planning manager and nurses of the surgeon department of the UMCG.
Furthermore, qualitative and quantitative historical patients information of the year 2016 is used regarding patients admitted to the nursing ward and their length of stay. This data is also used to evaluate the ORE model at the UMCG.
3. Theoretical background
This chapter will provide a theoretical background, starting with section 3.1, which gives an description of nursing ward terminology. Section 3.2 gives an overview of available tools for assessing effectiveness and explains the choice for OEE. Section 3.3 characterizes existing OEE frameworks available for various industries by highlighting their structure and elements.
Section 3.4 is meant to identify domain specific aspects of the frameworks and discuss if they are usable in a healthcare context. Section 3.5 will provide a summary of the literature findings.
2.6 Nursing ward terminology
In figure 2 a simplified flow of patients through nursing wards is shown. The flow parameters are described below.
Figure 2: Simplified flow of patients through wards
Admissions
A patients gets scheduled for an operation and gets admitted at the ward beforehand. After registrations the patients gets its own bed in which he or she is prepared for surgery. Patients are registered by the administrative staff of the ward and prepared for surgery by the nursing staff. The total amount of admissions is equal to the sum of a number of production parameters such as admissions, day treatments, and transfers (patients coming from another medical discipline) (Eldenburg et al., 2009).
Beds
Each ward has a fixed number of operational beds which determine the capacity. Staff is planned accordingly and enough medical equipment is kept on stock to handle the patients occupying the number of beds. Therefore, beds are considered to represent associated resources such as staff and medical equipment as well. On a day to day basis, the actual number of open or staffed beds slightly fluctuates (due to illness, holiday and patient demand) (Green et al., 2007).
Discharge
The process is finished when patients have recovered from surgery and are discharged from the hospital information system. They can either go home or to an aftercare institution if further care is necessary. It occurs that patients whom have been discharged returns to the hospital due to complications and need to get admitted again.
Length of stay
Time spent at the ward can be defined as length of stay (LOS). LOS is the term used to measure the duration of a single episode of hospitalization (Preedy and Watson, 2010).
2.7 Tools for assessing effectiveness
There is a broad scale available when looking at tools regarding process improvement such as Lean, Total Quality Management (TQM) and Lean Six Sigma (LSS). Lean focuses on reducing waste and to be efficient in terms of ‘just in time’, where TQM and LSS focus on reducing variation and making incremental steps towards the perfect process. Before one can engage in such performance improving activities, the areas to target, or the root cause of a performance problem, has to be known. OEE is a dedicated resource focused tool which can target domain specific system characteristics. It can systematically measure and assess current effectiveness at wards by providing starting points for root cause analysis. It is an advantage that OEE allows continuous monitoring of the most important factors influencing equipment performance, and it clearly identifies root causes for losses in manufacturing effectiveness (Dalmolen et al., 2013). The tool is proven to be a viable route and the basis for performance improving activities in both the manufacturing and logistic industry.
2.8 Overall equipment effectiveness frameworks
OEE is a quantitative metric tool which is used in the manufacturing industry to measure effectiveness in the form of production losses (Aminuddin et al., 2016) and is used to track and trace improvements over a period of time (Huang et al., 2003). The tool is used to identify the related resource losses for the purpose of improving total performance (Dal et al., 2000) and beginning of root cause analysis (Pintelon and Muchiri, 2008). The OEE is calculated in percentages and it uses time as the central metric unit (Dalmolen et al., 2013). Variations of the framework such as OTE are applied in the logistics industry to identify effectiveness losses from a truck perspective (Dalmolen et al., 2013). OEE was originally designed to monitor and control performance, but it has also been used to identify process improvement opportunities and as an approach to measure and achieve them. (Garza-Reyes et al., 2010).
2.8.1 Overall equipment effectiveness in manufacturing
OEE is a tool much used in the manufacturing literature to assess equipment effectiveness. It is a function of the mutually exclusive domains availability (A), performance (P) and quality (Q) and is essentially the result achieved by multiplying these three categories together as shown by the following equation:
OEE = A x P x Q
where
A = Operating time / Loading time * 100%
P = Theoretical cycle time x Actual output / Operating time * 100%
Q = Total production– Defect amount / Total production * 100%
The bottom-up approach strives to achieve overall equipment effectiveness by eliminating six big losses (Nakajima, 1988). The framework is shown in figure 3, which links the six losses to the three categories. The loss factors for the availability component of the framework are (1) poor productivity and lost yield due to poor quality, (2) setup and adjustment for product mix change, and for productivity they are (3) production losses when temporary malfunctions occur, (4) differences in equipment design speed and actual operating speed. The quality effectiveness exists out of the loss factors (5) defects caused by malfunctioning equipment, and (6) start up and yield losses at the early stage of production. The six big losses are given in figure 3 below which links them to the three categories A, P and Q. In appendix I the loss factors are further defined.
Figure 3: Overall equipment effectiveness framework
2.8.2 Overall transport effectiveness in logistics
In the logistics industry, the OTE framework is derived from OEE frameworks developed for manufacturing systems. OTE focuses on effective resource use from a truck perspective. It provides a hierarchy of metrics to show how effectively transportation is executed (Dalmolen et al., 2013). This measure converted the OEE losses from manufacturing to transport operations (Villareal et al., 2012). OTE is achieved by multiplying three categories together as shown by the following equation:
OTE = A x P x Q
where
A = Real running time / Used time * 100%
P = Real operation time / Real running time * 100%
Q = Effective time / Real operation time * 100%
In appendix II the loss factors of the OTE framework are defined. When comparing the transportation framework to its manufacturing equivalent, the availability and productivity component address similar loss factors in both frameworks. The quality component, however, addresses different loss factors and is redesigned for the logistics industry. Where the focus in the manufacturing framework is on defects caused by malfunctioning equipment and start up and yield losses, for the transportation framework the losses are the orders when a truck arrives outside its timeslot, speed losses caused by traffic jams and kilometres driven empty.
2.9 Overall resource effectiveness in healthcare
To design an ORE framework, the factors regarding the categories availability, productivity and quality have to be linked to the nursing wards its beds. To get an understanding of what can and cannot be used from current OEE and OTE frameworks, the similarities and differences for each category and their loss factors are discussed.
The availability component reflects the time a machine or truck is available by subtracting loss factors as equipment failure (manufacturing) and loading time (transportation). Although the loss factors address slightly different aspects, the goal is to get an understanding of the real operation time for both frameworks by subtracting the time a machine or truck is not available. It is expected that this can be done for the nursing ward as well.
The productivity category looks at loss factors as speed losses due to idling and technical failures in both manufacturing and transportation frameworks. They give an overview of net operating time by subtracting time used for non-value adding activities. This is done in both frameworks and thus it is expected that this will not be different when redesigning the framework for a nursing ward.
Last, the quality component, which shows differences among the OEE and OTE framework.
As mentioned before, the manufacturing framework loss factor is based on malfunctioning equipment and start up and yield losses, where the transportation framework losses are the orders when a truck arrives outside its timeslot, speed losses caused by traffic jams and kilometres driven empty. The loss factors clearly addresses different factors among the industries, which suggest that the quality effectiveness of nursing wards needs a redesign of the loss factor as well to be applicable in a healthcare context.
2.10 Summary
The terminology of nursing wards has been identified to get a clear understanding of the process at wards. Root cause analysis of the ward is needed before performance enhancing activities can be started. OEE is a dedicated tool which can systematically assess current effectiveness at wards by facilitating a complete root cause analysis The key elements of current OEE and OTE frameworks have been identified in order to define differences and similarities among them. The loss factors regarding the categories availability, productivity and quality have to be redesigned in such a way that they give a complete and systematically measurement of the effective use of beds at nursing wards. Where the first two categories address the similar loss factors among industries, the quality effectiveness addresses different loss factors.
3. System Description
In this chapter, the primary process of the nursing ward of the surgeon department at the UMCG will be described. Section 4.1 gives an overview of the nursing ward. Furthermore, it will discuss the scope of the research. In section 4.2, a description of the primary process of the nursing ward of the surgery department is given. Section 4.3 will link the primary process to a timeline and the variation that may occur. Section 4.4 describes the key resources of nursing wards, section 4.5 focuses on the operational offline planning of the ward, where section 4.6 elaborates on the online planning.
3.1 Overview nursing ward
The surgery department of the UMCG is one of its largest medical departments. The department handles the admission of patients whom have to undergo surgery. It consists out of the wards C4, A3, K4 and J2, which each has its own speciality. Table 2 gives an overview of the four nursing wards with their specialties, number of beds, admissions and average LOS.
In 2016 a total of 4478 patients were admitted at the nursing wards and 1515 takeovers from other wards took place. The four wards have a total availability of 103 standard and eight intensive care beds, for patients whom are in need of extra, divided over the specialisms. The latter has special medical equipment attached to the bed and is used for intensive care needs only, therefore not counting towards total number of available beds.
The focus of the research will be on the C4 ward, as it handles the most patients on a yearly basis; 1273 admissions and 416 takeovers. The ward is specialized in vascular surgery, hepatobiliary surgery and liver transplantation and has a total of 30 available beds; 15 for vascular surgery and 15 for hepatobiliary surgery and liver transplantation. The average LOS of patients at the C4 ward is 4.7 days, where 8.0 days is the average LOS of all wards.
WARD SPECIALISM BEDS AVAILABLE
ADMISSIONS TAKE- OVERS
AVERAGE LOS (DAYS)
A3 Abdominal surgery 20 588 248 11.8
A3 CIVZ 8 16 476 11.6
C4 Vascular surgery Hepatobiliary surgery &
liver transplantation
15 15
1273 416 4.7
J2 Trauma surgery 30 1184 198 6.8
K4 Oncology 23 1417 177 5.2
TOTAL 103 + 8 4478 1515 8.0
Table 2: Characteristics wards
3.2 Process nursing ward
Figure 4 provides a description of the primary process of the nursing ward. The process steps of the ward are explained below, where the resources and planning are discussed in the next sections.
Figure 4: Patient flow nursing ward
Admission
When patients get called in for a surgery by the operation room (OR) planning, they are admitted at the ward by the administrative staff. This is usually done between 10:00-14:00.
Admission variability per month, week and day is graphically shown in appendix III. When patients are dismissed from the hospital, the bed has to be cleaned before it can be used again.
This takes up to ten minutes and is usually done within the hour. During the cleaning, the next patient is admitted right away. If the registration of patients is finished before the bed is cleaned, they can wait at designated waiting areas before entering the bed. Elective patients get called in by the OR planning, who schedule patients if there is enough availability at the OR or by the radiology department, which schedule radiologic interventions for patients to run diagnostics for research. Patients get admitted at the ward as soon as they are called in by the OR planning. Emergency patients can arrive 24/7 from the trauma or intensive care department or are returning patients with complications.
Preparation
Before surgery patients get prepared by the medical staff of the hospital. The doctor gives further explanation about the surgery and what to expect. Preparation for the surgery is done by the nursing staff. This includes activities such as disinfecting and a visit to the anaesthetist if needed. The needed medical equipment e.g. drips or heart monitors are placed alongside the bed.
Surgery
Surgeries are performed in one of the ORs. The surgery ward has a different number of ORs available each day. As a result of this, the number of surgeries performed a day may differ significantly. Patients wait at the ward until their surgery is scheduled. When ready, patients and their beds are being moved to one of the ORs, in which they undergo surgery. The duration of the surgery can vary from half an hour to approximately eight hours, depending on the type of procedure and condition of the patient. When complications arise during surgery, it might happen that the surgical intervention takes longer than usual. As a result of this, surgeries scheduled afterwards might be postponed till later that day or even the next day.
When patients leave the ward to undergo surgery, there is no dismissal.
Recovery
After their surgery, patients return to the ward to recover. The nursing staff ensures that the patients get proper care. Their condition gets checked multiple times a day. When all goes well, they may leave the ward.
Dismissal
At the output of the process, patients whom had surgery and are recovered are dismissed from the hospital and discharged from the information system. When patients are transferred to another department, such as the intensive care, there is no discharge and the ward keeps a bed available for when they return. When further care is needed aftercare is arranged for the patient. If complications arise when patients have left the ward, they may get re-admitted.
3.3 Timeline process
Figure 5: Timeline activities
Figure 5 shows the timeline of the total LOS with its process steps preparation, surgery and recovery. Beforehand, patients get admitted and afterwards, they are dismissed. For each step, variation may occur due to the earlier or later start of previous steps. As a result of this, the actual LOS may differ significantly from the scheduled LOS. This is further elaborated on in section 4.5.2. The process steps and causes for variation are discussed below.
Preparation
Patient may arrive earlier or later than scheduled. When patients arrive to early, they have to wait until a bed is available. When arriving to late, the bed made available is not occupied, or idle, until they arrive. A resource is idle when it is not being used to produce value that it might otherwise produce (Buchanan, 2001). If a patient is admitted later, the preparation gets done faster in order to get the patient ready for the surgery.
Surgery
It may occur that a surgery starts later because the previous surgery took longer than expected. This has effect on the upcoming surgery as well, as it has to start later. As a result of this, the OR planning may disrupt. This is further discussed in section 4.5.1. When a surgery is finished early, the next surgery does not start earlier. The patients just return to the ward earlier and their recovery can start.
Recovery
It might also occur that patients get dismissed earlier or later, depending on the patients’
condition. Their condition gets checked multiple times a day in order to decide if further treatment is needed. When no further treatment is needed and patients have recovered sufficiently, they are dismissed. It might happen that complications arise, which causes patients to stay longer, affecting their LOS.
3.4 Key resources Staff
Each ward has its own fixed team of staff. The surgeon department itself ensures that there is always enough nursing staff scheduled to treat all patients. During holidays, when less personnel is available, the hospital decreases the number of available beds at the wards accordingly. As a result of this, the maximum number of admitted patients at the ward is limited by the number of beds.
Beds
Capacity of the ward is determined by the number of operational beds. There are 30 beds available at the C4 nursing ward. When a bed cannot be used again after patients are dismissed, a bed can be borrowed from another ward as it is a shared resource in the hospital.
Each bed is a standardized version and does not have any medical equipment attached. When patients are dismissed from the hospital and leaves the bed, it has to be prepared for the next patient. Cleaning is done during the whole day and beds are cleaned within the hour after the dismissal. If patients arrive at the bed when it is still being cleaned, they can wait at the designated waiting area before occupying the bed. When, for unforeseen reasons, a bed cannot be made available again within the hour, the ward can borrow a bed from another department to ensure the next admission can be handled and the patient has a bed to lie in.
Medical equipment
Medical equipment, e.g. drips, heart monitors and blood pressure cuffs, are kept on stock by every ward itself. When medical equipment is needed for a patient, the nursing staff arranges that it is moved towards the patients’ room and attaches it to their bed. If, for unforeseen reasons, a piece of equipment is not available, it can be borrowed from other wards as they also are a shared resource in the hospital.
3.5 Operational offline planning 3.5.1 OR planning
Offline planning involves the detailed planning of operations done in advance. The OR planning maintains a schedule of two weeks and gives an indication of how many patients there will be arriving in the following weeks, although the actual number of admissions can
differ significantly due to emergency patients. Patients also get scheduled by the radiology department, which schedule radiologic interventions to run diagnostics for research.
The number of surgeries performed a day can differ significantly, as the time to perform a surgery can differ from e.g. half an hour to eight hours and the number of available ORs can differ from one to four. As a result of this, there is a significant variety in the arrival of patients, which the ward has to cope with. The variety in patient arrival on the C4 ward per month, week and day is graphically shown in appendix III. Furthermore, there is little to no slack time scheduled in-between surgeries. The slack time is the difference between the scheduled and actual expected surgery times (Bukkapatnam et al., 2006). When a surgery takes longer than expected, the schedule gets disrupted. As a result of this, following surgeries are postponed till a later that day, or even the next day.
3.5.2 Ward planning
Each ward has its own planners and is responsible to match their own schedule with ORs and plan their resources accordingly, therefore making it a decentralized operational planning. The ward itself maintains a schedule of one week, which gets prepared the week before. The ward schedules the total LOS of a patient, which is determined by the surgeon, and ensures needed resources are available. When patients do not stay at the ward as long or longer then scheduled, there is no check afterwards.
When the OR planning creates the schedule, there is no link with the ward to check the availability of beds in advance. As a result of this, it may happen that patients are scheduled to get admitted and the ward has no available beds. This might also happen due to patients whom need more time to recover then scheduled. When this occurs a bed can be borrowed from another ward. As soon as a bed is available at the own ward, it is being switched.
Staff illness
When there is staff illness, the number of available beds does not decrease. The ward tries to find substitutions by letting the remaining staff fill in or lending staff from other departments.
It has not occurred that beds are made unavailable due to illness.
Holidays
During holidays, the total capacity of the hospital is decreased to enable staff members to leave for vacation. For the wards, the number of available beds is decreased accordingly. The reduction per ward in available beds is shown in table 3. This happens for a total of ten weeks a year, so the C4 ward has a decreased availability from 30 to 24 beds for a total of 10 weeks a year and has 30 beds available for the remaining 42 weeks.
WARD NORMAL AVAILABILITY DURING HOLIDAY
C4 30 24
J2 30 24
A3 20 16
K4 23 20
TOTAL 103 84
Table 3: Bed availability during holidays
3.6 Operational online planning
Online planning involves planning on a daily basis. Every morning the planners and the medical staff kick-off the day with a meeting regarding planned admissions and current occupation of the ward. Throughout the day the planners are responsible for controlling the flow of patients and ensuring that every arriving patient has a bed to lie in.
The main concern for the wards planners and their schedule are emergency patients. As the demand is unpredictable, it may disrupt the offline planning significantly. The ward does not keep a few beds available for such emergency patients. As a result of this, it might happen that the ward is fully occupied when an emergency patient arrives. The medical staff tries to ensure that the arriving patient can be admitted at the ward by checking the condition of patients throughout the day to determine if they can be discharged or need further care. If there is no bed available and cannot be made available, it may occur that patients are being moved to a bed at another ward until there is a bed available again.
Differences in scheduled and actual discharge dates are not reflected on. Patients may lay shorter, or longer, in the bed than anticipated, which may influence the availability of the ward.
4. Framework Design
This chapter will discuss the proposed overall resource effectiveness (ORE) framework.
Section 5.1 will explain the approach for creating the ORE framework. Section 5.2 gives an overview of the structure of the framework and explains the set-up. Section 5.3 gives definitions of used elements, where section 5.4 will address and elaborate on the loss factors of the framework.
4.1 Approach
The ORE framework builds on the literature review on existing frameworks from the manufacturing and logistics industry, related literature in the health domain and observations in practice on nursing ward set-up and operations. The OEE framework, as designed by Nakajima (1988), discussed in section 3.3.1, is used as a basis. The framework is well accepted in the manufacturing industry and has a good fit with healthcare operations. Where current frameworks focus on the effectiveness of equipment in manufacturing or trucks in transportation, the ORE framework will focus on the effective use of beds at nursing wards.
As section 2.3 illustrates, loss factors for the categories availability, productivity and quality are linked the domain specific characteristics of nursing wards. The loss factors in each category identify starting points for root causes of losses in effectiveness. A literature review on existing frameworks, as mentioned in section 3.4, showed that the first two categories address similar loss factors among industries.
Quality is defined differently among industries and addresses different loss factors. A fundamental issue in service quality research, is the conceptualization and measurement of service quality (Basu and Biswas, 2012). Even though many researchers have agreed that service quality is a multi-dimensional construct, there is no consensus on how many and what the dimensions are (Clemes et al., 2008). The Healthcare service quality model, as defined by Sumaedi et al. (2015), is used to identify quality dimensions and measures in healthcare. To make it applicable in the framework, the loss factor is redesigned and validated in association with experts at UMCG.
The framework gives insights on a strategic, tactical, and operational level and visualizes current performance. ORE is a hierarchy of metrics which show how effectively beds are used at nursing wards. The framework systematically measures performance and provides starting
4.2 Overall resource effectiveness framework
Figure 6: Overall resource effectiveness framework
The ORE framework is proposed, as shown in figure 6, and exists out of three categories:
availability (A), productivity (P) and quality (Q). ORE is calculated by multiplying the three categories with one another, as shown below:
ORE = A x P x Q
Each category has its loss factors (Losses), which provide insights on effectiveness in each category. By subtracting loss factors from the total available time, the ward gets less effective.
The availability category gives insights in the time a bed is available and the time a bed is actually being occupied. The productivity category gives insights in speed losses and delays, where the quality category defines quality losses and their respective time loss. The calculation behind each category is shown in the section 5.3.
4.3 System elements
ORE is calculated by multiplying the three categories with one another:
ORE = A x P x Q
Where
A = (Used time / Total time) * 100%
P = (Actual running time / Used time) * 100%
Q = Actual running time (Patients without complications) / (Actual running time (Patients without complications) + Actual running time (Patients with complications)) * 100%
The elements used in the framework are defined in table 4.
ELEMENT DEFINITION
Total time This contains the selected time period e.g. one year ((365 days * 24 hours) * number of beds)
Used time Total LOS (preparation + surgery + recovery time)
Down time Time lost due to early/late admissions, holidays, breakdown, cleaning, shortages and when a bed is not occupied
Actual running time Surgery + recovery time
Production losses Time lost due to speed losses and delay in schedules Valuable time Actual running time for patients without complications
Readmission losses Time lost for a re-entry because a patient returns with complications
Table 4: Definition elements
4.4 Loss factors
Below the loss factors for the three categories availability, productivity and quality are discussed. A literature review on current frameworks provided similarities and differences between OEE and OTE frameworks and gave insights on which areas their loss factors are focused. As mentioned in section 3.4, the categories availability and productivity address similar loss factors, where the quality category focuses on different aspects among industries.
In association with experts of the UMCG and observations in practice, the loss factors at nursing wards have been identified and linked to domain specific characteristics.
4.4.1 Availability Not available (breakdown)
Similar to machines, beds can breakdown and need to be maintained in order to admit patients. This influences the availability at wards, as beds may not be directly available at other wards as well and it takes time to find one. This loss factor creates an overview of the time beds are not available from the moment they break down till the moment they are fixed and/or replaced.
Not available (cleaning)
Just as in the manufacturing industry, where a machine needs to be set-up in order to produce, a bed needs to be prepared in order to use it. When a patient is dismissed from the hospital, cleaning activities have to be done. These activities might be delayed due to e.g. no available personnel or equipment. During this time a new patient cannot occupy the bed, as they have to wait until cleaning activities are finished.
Not available (holiday)
During holidays, as mentioned in section 4.5.2, the number of available beds is decreased to enable staff members to leave for a vacation. By incorporating this loss factor, the wards gets insights in how much available time is lost due to the decrease of the number of available beds during holidays.
Not available (staff shortages)
When a patient lies in a bed, it needs staff to be treated. It might occur that there is a shortage of staff, due to e.g. illness or planning mistakes. This loss factor gives an indication of time lost due to staff shortages.
Not available (equipment shortages)
When a patient lies in a bed, the needed medical equipment gets attached to a bed by the nursing staff. These types of medical equipment might not always be available. This loss factor gives an indication of time lost due to equipment shortages.
Not occupied
Just as a machine, a bed can be in an idle state, as it is not always occupied by a patient. This loss factor gives an indication of the utilization of beds by looking at the time a bed is available and when it is actually occupied by a patient. As the number of surgeries, scheduled by the OR planning, can differ significantly, the ward may be under- or over-utilized at certain times.
Early/ late arrival of patients
When patients get scheduled for their surgery, the LOS is determined by the surgeon based on the patients’ health issues. As mentioned in section 4.3, patients might arrive earlier or later than scheduled. If patients get in earlier, the bed might not be ready for them and they have to wait. When patients arrive later, the bed is already made available as scheduled. However, as there is no patient yet, it is not occupied and in an idle state. As a result of this, patients’
actual LOS might differ significantly from the duration proposed by the surgeon. By incorporating this loss factor, the actual preparation time can be compared to the proposed preparation time. This way the norms set by surgeons on preparation times can be reflected on and adjustments can be made if necessary.
Early/ late dismissal of patients
Just as patients may arrive early or later, they can get dismissed early. Depending on the condition of the patient, the staff decides if further treatment is needed, or if the patients may go home early. This loss factor gives an indication in the difference between the scheduled recovery times and actual recovery times. This way the norms set by surgeons on recovery times can be reflected on and adjustments can be made if necessary.
4.4.2 Productivity Speed loss (surgery delayed)
This factor will be incorporated by looking at the planned and actual schedule of the surgery.
As mentioned in section 4.3, the starting time of the surgery may differ from the schedule.
When a patient is scheduled for a surgery, it happens that it is delayed when, for example, the previous surgery takes longer than expected. The patient is already brought to the OR but has to wait before his operation can start. During this period, little to no value adding activities are being undertaken up to the moment the surgery starts. When a surgery finishes earlier, the next surgery is not done earlier, but at the scheduled time. By incorporating this loss factor, the ward can get an understanding of losses regarding delayed surgeries.
Waiting time (preparation)
When patients get admitted, they get prepared for their surgery. A significant amount of this time, no value adding activities are done. This is incorporated in the framework by looking at the total LOS of the patient divided by the time actual value adding activities are done;
operating time and recovery time. This loss factor gives insight in how much time of the LOS of the patient is used for the admission and preparation of the patient and how much time actual value adding activities are done for surgery and recovery. This enables the ward to assess if the preparation time is representative for the needed activities.
4.4.3 Quality
Patients with complications
The quality of the ward in terms of medical considerations will be defined by patients whom leave the ward with or without complications. Patients who return to the ward with
the availability and productivity of the ward, as the LOS of the readmitted patient increases significantly. Furthermore, the bed is not available for a new elective admission and available time is lost. This loss factor creates an overview of the amount of patients with complications that return to the hospital, and its effect on the available time.
5. Case Study Validation
In this chapter the framework is validated by applying it at the UMCG. All results are checked and evaluated with UMCG staff to ensure correct design and use of the framework. The framework functions are illustrated by addressing all indicated loss factors. In section 6.1 the results are presented where section 6.2 provides the contributions made by the framework in terms of recommendations linked to said loss factors for the UMCG.
5.1 Framework Results
The framework, as discussed in chapter 5, is tested by using UMCGs historical patient data from the C4 ward in 2016. The results of the framework are presented first in section 6.1.1.
The categories and their corresponding loss factors are further explained afterwards in section 6.1.2.
5.1.1 Overview overall resource effectiveness
For the ORE component the following equation is used: ORE = A x P x Q
Using the outcomes per category, the ORE of the C4 ward is: 74% x 75% x 85% = 47%
Below the results per category are shown. The figures are defined afterwards.
Figure 7: Availability category C4 Ward Figure 8: Productivity category C4 Ward
5.1.2 Availability
For the availability component the following equation is used:
A = (Used time / Total time) * 100%
The Used time is the sum of the LOS of all patients who have been on the C4 ward in 2016.
The summation of the LOS of all patients at the C4 wards is 192.768 hours.
For the Total time, the number of available bed is multiplied by the number of weeks and the hours per week. This comes to a total of 30 x 52 x 168 = 262.080 hours.
Therefore, the availability effectiveness of the C4 wards is:
(192.768 / 262.080) * 100% = 74%
There is a loss of 26% of the available time. In table 5 the percentages per loss factor are shown, which are explained afterwards.
LOSS FACTOR PERCENTAGE
NOT AVAILABLE (BREAKDOWN) -
NOT AVAILABLE (CLEANING) -
NOT AVAILABLE (STAFF SHORTAGES) -
NOT AVAILABLE (EQUIPMENT SHORTAGES) -
NOT AVAILABLE (HOLIDAY) 4%
NOT OCCUPIED* 22%*
EARLY/ LATE ARRIVAL OF PATIENTS -
EARLY/ LATE DISMISSAL OF PATIENTS -
* This loss factors also includes the first four and the last two loss factors, as they are currently not being measured.
Table 5: Loss factors availability effectiveness
Not available (breakdown)
This factor is currently not being measured and therefore cannot be reflected on. When breakdown occurs the ward usually can lend a bed from a different ward, but the time it takes to do so is not registered.
Not available (cleaning)
This factor is currently not being measured and therefore cannot be reflected on. When patients leave their bed, the bed has to be cleaned, but the time it actually takes is not registered.
