Lean in Healthcare: Applying Value Stream Mapping and Lean in the Accident & Emergency Unit at G.F. Jooste Hospital A Research Report Presented to The Graduate School of Business University of Cape Town In Partial Fulfilment of the Requirements for the Masters of Business Administration Degree By Antonio S. Booysen December 2006 Supervisor: Prof. Norman Faull
Note of Acknowledgement This report is the copyright of the University of Cape Town and is intended for use by the Graduate School of Business. Firstly, thanks must be given to Norman Faull. Without his inspiration, knowledge on TPS/Lean and guidance this research report would not have been undertaken. Thanks must also be given to Numaan Mohamood whose paper titled A Strategy Model for G.F. Jooste Emergency Unit provided additional inspiration in doing this project. The greatest vote of thanks goes to the staff at G.F. Jooste Hospital whose involvement and support facilitated the undertaking of this research report. Specifically, thanks is given to Dr. Neshaad Schrueder and members of the A&E Unit, Dr. Gio Perez and members of the hospital administration team, Mr. Chris Reyneveld and members of the Medical Records department, and the Triage team consisting of Dr. Moosa Parak, Nurse Frances Goliath and Nurse Andea van Schalkwyk. Gratitude also goes to all other staff members not mentioned by name who were involved within this research study. Lastly, acknowledgement must be given to Stephen Brooks, whose research report, titled A lean services environment experiment: Introducing TPS and Lean principles to a pathology laboratory, aided the author in drafting this report. Brooks paper was issued as part of the Research Methodology lecture series on the MBA Full-Time course of 2006. Where applicable, the author has attempted to acknowledge Brooks. Any instance(s) where recognition is not afforded Brooks is/are unintended, as the author understands that plagiarism is wrong. I certify that, except as noted above, the report is my own work, and all references used are accurately recorded in the Bibliography section of this document. Signed: Antonio S. Booysen (BYSANT001) Date: 08 December 2006
Abstract The Accident & Emergency Unit at G.F. Jooste Hospital (GFJ) has been experiencing significant operational problems in terms of resources, infrastructure, low morale amongst staff members, and over-capacity in terms of patient numbers and patient flow and tracking. The aim of this research study was to address the above areas of concern through the use of Value Stream Mapping, and the implementation of Toyota Production System/Lean principles and philosophies, as the hypotheses stated in this report seek to determine whether Toyota Production System/Lean principles and philosophies are transferable in a service delivery environment. To facilitate this process, an Action Research methodology was employed. This methodology was believed to be the most feasible means of determining the validity of this report s hypotheses. The research study found that Value Stream Mapping could be used to describe patient flow accurately in the Accident & Emergency Unit, and that Value Stream Mapping could be understood by the primary role players and used as a tool for continuous improvement initiatives. Furthermore, this research study also found that Toyota Production System/Lean principles and philosophies could be implemented within the Accident & Emergency Unit, and that these principles and philosophies could form the basis for change within the environment of the Accident & Emergency Unit in an effort to address the challenges mentioned above. 3
Table of Contents 1. Introduction... 7 2. Environment of Research Study... 7 2.1. Background... 7 2.2. Problem Definition... 8 2.3. Research Objectives... 9 2.4. Research Hypotheses... 9 2.5 Assumptions & Limitations... 10 3. Literature Review... 10 3.1. Value Stream Mapping... 11 3.2. Lean thinking... 12 3.3. Toyota Production System (TPS)... 15 4. Research Methodology... 16 4.1. What is Action Research?... 16 4.2. Implementation of Action Research... 17 4.3. Applicability of Action Research in the environment under study... 18 4.4. What was the role of the researcher?... 19 5. Hypothesis Testing... 20 5.1. Hypothesis 1... 21 5.1.1. Value Stream Mapping... 21 5.1.2. Conclusion... 27 5.2. Hypothesis 2... 30 5.2.1. Testing of Hypothesis... 30 5.2.2. Conclusion... 31 5.3. Hypothesis 3... 32 5.3.1. Testing of Hypothesis... 32 5.3.2. Conclusion... 52 5.4. Hypothesis 4... 54 5.4.1. Testing of Hypothesis... 54 5.4.2. Conclusion... 55 6. Conclusion... 55 7. Bibliography... 56 Appendix... 59 Exhibit 1: Head of Department Report... 59 Exhibit 2: Meeting Transcripts... 61 4
Glossary of Terms and Abbreviations used in text Term Definition A&E Unit Accident & Emergency Unit: The environment under study in this research study GFJ G.F. Jooste Hospital: The location of the study environment (A&E), situated within the Cape Town metropolitan area Secondary level referral hospital A level of hospital between primary health, as offered by municipal clinics, and tertiary hospitals, as offered by academic hospitals Lean (Thinking) A cyclical process of seeking perfection by eliminating waste and thereby enriching TPS value from a customer perspective (Brooks, 2005) Toyota Production System: The method of automobile production developed by Toyota after World War II with the aim of developing better cars for more people AR Action Research: The selected method of research for this study VSM Value Stream Mapping: A map of all the essential actions/steps (both value added and non-value added) required to bring a product or service to a customer. It shows material and information flows as a product makes its way through the value stream (Brooks, 2005) JIT Just-In-Time: TPS production method that seeks to eliminate waste 5
List of Figures Figure 1: The 5S Process... 14 Figure 2: The Action Research Cycle... 17 Figure 3: The Action Research Spiral... 18 Figure 4: Layout of the A&E Unit... 28 Figure 5: Value Stream Map as at 20/21 September 2006... 29 Figure 6: Coding of patients by severity of presenting condition... 38 Figure 7: Key changes undertaken in Experiment 2 with respect to layout... 46 Figure 8: Comparison of VSM for Triage on 20/21 September and 16 October... 47 Figure 9: Comparison of VSM for ER on 20/21 September and 15 November... 53 List of Tables Table 1: Testing of Hypothesis 1... 21 Table 2: Key Findings in Triage during Value Stream Mapping on 20/21 September... 23 Table 3: Key Findings in ER during Value Stream Mapping on 20/21 September... 25 Table 4: Testing of Hypothesis 2... 30 Table 5: Testing of Hypothesis 3... 32 Table 6: Outline of Guidelines for Experiment 1... 33 Table 7: Findings during Experiment 1... 34 Table 8: Comparison of patient throughput in Triage after Experiment 1... 35 Table 9: Breakdown of patients entering A&E Unit by category... 37 Table 10: Outline of Guidelines for Experiment 2... 39 Table 11: Results obtained during Data Gathering in Experiment 2... 43 Table 12: Problems encountered during Experiment 2... 44 Table 13: Outline of Guidelines for Experiment 3... 48 Table 14: Results obtained during Data Gathering in Experiment 3... 51 Table 15: Testing of Hypothesis 4... 54 6
1. Introduction The Accident & Emergency (A&E) Unit at G.F. Jooste Hospital (GFJ) has been experiencing significant operational problems in terms of resources, infrastructure, low staff morale, over-capacity in terms of patient numbers and patient flow and tracking (Exhibit 1, Appendix). The principles of Lean thinking are based on the Toyota Production System (TPS). Through the application of Action Research (AR) principles, the intention of this research study was to test whether Lean principles are relevant in the A&E Unit environment, and whether application of these principles could enhance operational performance within the A&E Unit. 2. Environment of Research Study 2.1. Background G.F. Jooste (GFJ) is a public, secondary level referral hospital situated in Heideveld on the Cape Flats. The hospital serves the low-income surrounding suburb of Manenberg and the townships of Gugulethu, Nyanga and Phillipi on a 24-hour basis, 7 days a week, 365 days per year, and receives patient referrals from surrounding Clinics, Day Hospitals and General Practitioners (Mohamood, 2006). The hospital forms part of the University of Cape Town s training rotation scheme for medical students, with Groote Schuur Hospital as the supporting tertiary level hospital 1. The A&E Unit serves, on average, between 150 and 200 patients per day. The area of focus in this study was the A&E Unit at GFJ. The A&E Unit offers a wide range of emergency medical and auxiliary medical services. 1 An academic hospital offering the full spectrum of medical services 7
This research study focused on patient flow within the A&E Unit. There are various functional areas within the A&E Unit that form part of the patient flow process. These functions include: Administration (primarily clerking of patients) Triage (sorting of patients by category) Porter services (porters are responsible for intra- and inter-departmental conveyance of patients) Emergency Room (where patients are assessed, diagnosed and decisions on further treatment are taken) Auxiliary Medical Services: o Radio-diagnostic services (primarily X-Ray and CT Scanning) o Pathology (an in-hospital laboratory service) o Pharmacy 2.2. Problem Definition In the words of the Head of Department for the A&E Unit at GFJ, the problem under investigation in this study is described as follows: The Emergency Unit of GF Jooste Hospital is a high stress, high intensity unit that is constantly overwhelmed by numbers and severity of illness of patients. The turnover and pressure to perform is unique. The hospital is grossly under resourced for the patient numbers and the capacity is stretched to the limit on a regular basis. The patient flow through the unit is always a major issue with delays in admission times to the wards a major factor in clogging the casualty and occupying staff with an extra patient care load (Exhibit 1, Appendix). It is in the context of the above problem description that the problem under investigation in this study is defined as follows: the intermittent progress of patients through the A&E Unit places significant pressure on capacity and impacts negatively on the level of patient service. 8
2.3. Research Objectives This research study sought to answer the following questions: Can patient flow within the A&E Unit at G.F. Jooste be mapped via Value Stream Mapping (VSM), and can the map be used to clarify that flow to the significant role players, namely doctors, nurses and support staff? Can VSM be used as a tool to facilitate improvement initiatives in improved patient flow and improved patient service level? Can TPS/Lean principles, philosophies and tools be applied in the A&E Unit to improve patient flow and patient service level, as measured by reduced time in the system? Can the introduction of TPS/Lean thinking in the A&E environment result in a positive attitude amongst the significant role players doctors, nurses and support staff which could drive sustained performance improvement initiative within the A&E Unit? 2.4. Research Hypotheses The primary hypotheses of this research study are as follows: Value Stream Mapping (VSM) can be used to describe patient flow within the A&E Unit, and can be used to clarify that flow to the significant role players, namely doctors, nurses and support staff VSM can be used as a tool to assist improvement initiatives in enhanced patient flow and improved patient service level Application of the principles, philosophies and tools of TPS/Lean thinking can contribute to improved patient flow and patient service level, as measured by reduced time in the system 9
Introduction of TPS/Lean thinking in the A&E Unit can result in a positive attitude amongst the significant role players doctors, nurses and support staff which could drive sustained performance improvement initiative in the A&E Unit 2.5 Assumptions & Limitations This research study within the A&E Unit at GFJ was subject to the following assumptions and limitations: The principles and philosophies of TPS/Lean, if proven of benefit within this research study, can be applied to environments similar to the A&E Unit at GFJ, and therefore outcomes of the case study will be applicable to these environments Intermittent progress of patients affects the quality of service offered to patients in the A&E Unit Applications of TPS/Lean principles, philosophies and tools were focused on patient flow in the A&E Unit at G.F. Jooste 3. Literature Review The focus of this literature review was guided by the questions and hypotheses that were under investigation in this study. This literature review therefore focuses on Value Stream Mapping (VSM), the Toyota Production System (TPS) and Lean thinking as the principles and philosophies of TPS/Lean thinking form the basis of this study, with VSM applied as a tool within these approaches. A literature review was also undertaken on the research methodology employed within this research study. The relevant review of literature pertaining to the research methodology is discussed elsewhere in this document (Section 4). 10
3.1. Value Stream Mapping A value stream is all the actions (both value added 2 and non-value added) currently required to bring a product through the main flows essential to every product: (1) the production flow from raw material into the arms of the customer, and (2) the design flow from concept to launch (Rother & Shook, 2003: 3) Value Stream Mapping (VSM) is a visual aid that describes the various stages in the consumption process. Womack & Jones (2005: 42) propose that a complete value stream consists of the consumption stream plus the provision stream, where the consumption stream visualises consumption as experienced by the customer and the provision stream visualises the activities involved in delivering a product or service by the provider. Liker & Meier (2005: 34) suggest that seeing the VSM in reverse, from the customer s perspective, can form the basis for the future desired state VSM by creating pull 3 within the process flow 4. Liker & Meier (2005: 37) suggest the following tips, which aided the researcher in this study, with regard the use of VSM: The current state VSM can be used as the foundation for the desired future state The future state map represents what the organization is attempting to achieve Future state map facilitation should be performed by a lean expert The purpose of the VSM is a basis for action The VSM should be developed timeously A senior person should lead improvement initiatives PDCA cycles 5 should occur on a continuous basis in order to facilitate continuous improvement 2 Those work elements that actually transform the product in a way that the customer is willing to pay for (Rother & Shook, 2003: 21) 3 A form of flow in which a supplier process sends a product to its customer process only when signalled to do so (Mishina & Takeda, 1995: 2, Liker & Meier, 2006: 100) 4 cutting back to zero the amount of time that any work project is sitting idle, waiting for someone to work on it (Liker & Meier, 2006: 9) 5 An acronym for Plan-Do-Check-Act: prior to undertaking any improvement initiatives it is advisable to define the problem under study through a thorough root cause analysis (Liker & Meier, 2006: 364) 11
Liker & Meier (2006: 75) contend that the 80/20 rule is useful when considering divisions in products that will isolate variation. To reduce variability in processing time we consolidate similar products based on required processing time. The 80/20 rule 6 provided a useful guideline in selecting the patient categories to focus on during VSM and in undertaking improvement initiatives within this research study. VSM proved a powerful tool in aiding this research study by defining and clarifying the patient flow process in the A&E Unit. It served as a basis for action towards continuously improving patient flow and service delivery under the guidance of the researcher. PDCA cycles, performed on an iterative basis, underlie the drive for continuous improvement 7. 3.2. Lean thinking By understanding processes and identifying non-value adding waste through root cause identification, the Lean organisation seeks to eliminate muda 8 and create continuous, pull flow within the processes of the value stream by involving key partners in this process that seeks to eliminate non-value adding activities 9 (Liker & Meier, 2006). Within the A&E Unit, one of the intended outcomes of this study was to reduce the time patients spend in the system through elimination of waste and non-value adding activities. A literature review, to justify this viewpoint, follows in the ensuing paragraphs of this sub-section. 6 Also known as the Pareto Principle (Wikipedia, 2006): the tricky thing about variation is that 20 percent (the minority) of the product often provides 80 percent of the total variation (Liker & Meier, 2006: 75) 7 Also known as kaizen in Japanese: a key to learning in Toyota is reflection. It is the driver of kaizen (Liker & Meier, 2006: 8) 8 Japanese term meaning waste (Liker & Meier, 2006: 345) 9 Activities regarded as waste (Liker & Meier, 2006: 35) 12
The term Lean thinking was first defined in The Machine that Changed the World: The Story of Lean Production (Womack et al, 1991). In Lean Thinking (Womack & Jones, 1996) five core concepts are identified which embody Lean thinking: Specify value in the eyes of the customer Identify the value stream and eliminate waste Make value flow at the pull of the customer Involve and empower employees Continuously improve in the pursuit of perfection Liker & Meier (2005: 14) propose that a Lean learning organization seeks to achieve its objectives with minimum waste by continually getting better. Harrison and van Hoek (2005: 171) also add that Lean thinking is a cyclical route to seeking perfection by eliminating waste and thereby enriching value from the customer perspective. Liker & Meier (2005: 6) assert that the principles of Lean thinking are embodied in Toyota s 4 P Model, namely: Philosophy the foundation of Lean thinking is to first create value for the customer, society and associates of the organisation Process to create pull-flow within the value stream by identifying sources of value to the customer and also to identify and eliminate waste People/Partners to grow leaders, to develop exceptional people and to respect the extended network in order to help partners improve Problems by adopting a go and see 10 approach in order to understand problems, and identifying the root cause of problems and instituting countermeasures which are implemented through Plan-Do-Check-Act cycles organizational learning can be achieved through continuous reflection and improvement 10 Referred to as genchi genbutsu in Japanese, the principle is to go and see the actual place and understand the real situation through direct observation (Liker & Meier, 2006: 8) 13
Lean thinking identifies eight non-value adding activities (Liker & Meier, 2005: 35): Overproduction Waiting Transportation and conveyance Over-processing Excess Inventory Unnecessary movement Defects Under-used employee creativity Underlying the efforts to eliminate waste is the 5S process 11, the primary purpose of which is likened to clearing the clouds 12 (Liker & Meier, 2006). The sub-components of the 5S process are illustrated graphically in Figure 1: Sort Clear out rarely used items by red tagging Sustain Use regular audits to stay disciplined Straighten Organise and label a place for everything Eliminate Waste Standardize Create procedures to maintain the first 3 S s Shine Clean it Figure 1: The 5S Process (Source: Liker & Meier, 2006) 11 A process that seeks to remove clutter in a work area (Liker & Meier, 2006: 64) 12 A metaphor comparing processes to a photograph: often cloudy conditions obscure the underlying view of a photograph. By removing the clouds the underlying view is enhanced (Liker & Meier, 2006: 58) 14
3.3. Toyota Production System (TPS) We get brilliant results from average people managing brilliant processes. We find that our competitors get average results (or worse) from brilliant people managing broken processes. A Toyota View (Source: The Economist, 10 September 2005) TPS is based on two guiding principles (Mishina & Takeda, 1995: 2), namely: Just-In-Time (JIT) produce only what is needed, only how much is needed, and only when it is needed. Any deviation from true production needs is condemned as waste. Jidoka make any production problems instantly self-evident and stop producing whenever problems are detected. It insists on building in quality in the production process and condemns any deviation from value-addition as waste. In TPS needs and value are defined from the viewpoint of the next station down the line, that is, the immediate customer (Mishina & Takeda, 1995: 2). TPS is based upon the scientific method which is described by Toyota s Four Rules (Spear & Bowen, 1999): Rule 1: All work shall be highly specified as to content, sequence, timing and outcome Rule 2: Every customer-supplier connection must be direct, and there must be an unambiguous yes-or-no way to send requests and receive responses Rule 3: The pathway for every product and service must be simple and direct Rule 4: Any improvement must be made in accordance to the scientific method, under the guidance of a teacher, at the lowest possible level of the organization By applying these Toyota s Four Rules the organisation aims for output that is defectfree, that can be delivered one request at a time and that is produced without waste (Brooks, 2005). 15
4. Research Methodology The chosen methodology for this research study was Action Research (AR), which is defined as an approach to research that aims both at taking action and creating knowledge or theory about that action (Coughlan & Coghlan, 2002: 220). AR was chosen above alternative methodologies for this research study as this methodology relates to describing an unfolding series of actions over time in a given group, community or organisation; understanding as a member of a group how and why their action can change and improve the working of some aspects of a system; and understanding the process of change or improvement in order to learn from it (Coughlan & Coghlan, 2002: 227). These principles are aligned with the objectives and hypotheses of this research study (Section 2.3 and Section 2.4). 4.1. What is Action Research? The term Action Research was first used by Lewin in 1946 (Saunders et al, 2003: 93). Within the literature, there are three common themes that describe AR (Saunders et al, 2003: 94). These focus on: The purpose of the research: the management of a change (Cunningham, 1995). Involvement of the practitioners in the research, and involvement with members of an organisation over a matter which is of genuine concern to them (Eden & Huxham, 1996: 75). The research should have implications beyond the immediate project (Saunders et al, 2003: 94) with regard the development of theory (Eden & Huxham, 1996) that could be applied in other contexts. The purpose of (action) research and discourse is not just to describe, understand and explain the world, but also to change it (Coghlan & Brannick, 2001). Critical to the success of AR is the importance of employee buy-in through active involvement in the process (Schein, 1995) as they are the agents most likely to implement the change they 16
have helped create. Schein (1999) also describes AR consultant activities as process consultation having two distinct foci (Saunders et al, 2003: 94): Fulfilling the agenda of those undertaking the research rather than that of the sponsor The needs of the sponsor and involves those undertaking the research in the sponsor s issues, rather than the sponsor in their issues 4.2. Implementation of Action Research Action research (AR) employs a spiral of sequential steps (Thornhill et al, 2000). These steps are as follows (Saunders et al, 2003): The research process commences with an initial idea and criteria for change intervention, generally expressed as an objective Reconnaissance about the intervention is undertaken in order to generate an overall plan and a decision about the first steps to be taken Implementation of planned action steps Monitoring of implementation and the effects thereof Evaluation of implemented action steps Context & Purpose Data Gathering Evaluation Data Feedback Monitoring Implementation Data Analysis Action Planning Figure 2: The Action Research Cycle (Adapted from: Coughlan & Coghlan, 2002) 17
Subsequent AR cycles involve revising the change intervention to ensure it meets the needs of the organisation using information gathered through the Monitoring and Evaluation process. Planned action steps are amended and Implemented to take account of unforeseen changes, their effects monitored and evaluated and further amendments made. Saunders et al (2003: 95) suggest that this stepwise approach requires the completion of at least two cycles within the AR methodology (Figure 3). Figure 3: The Action Research Spiral (Source: Thornhill et al (2000) as cited in Saunders et al 2003: 95) 4.3. Applicability of Action Research in the environment under study Action Research (AR) is applicable when the research question relates to describing an unfolding series of actions over time in a given group, community organisation; understanding as a member of a group how and why their action can change and improve the working of some aspects of a system; and understanding the process of change or improvement in order to learn from it (Coughlan & Coghlan, 2002: 227). 18
As outlined in Section 2.2, the A&E Unit at G.F. Jooste Hospital is presently experiencing issues of concern around over-capacity in terms of patient volumes with consequent sub-optimal service delivery to patients. Within the context of the above literature review, the researcher proposes that the A&E Unit, through the application of AR methodology, provided an ideal environment for the implementation of the principles, philosophies and tools of TPS/Lean thinking. Through this process it is proposed that changes introduced can generate theory that will be applicable to other environments of a similar nature. 4.4. What was the role of the researcher? Within AR the researcher s role is that of an outside agent who acts as a facilitator of action and reflection within an organisation (Coughlan & Coghlan, 2002). Schein (1999) describes two main models of helping: The expert model which is akin to the doctor-patient model where the patient approaches the doctor for expert diagnosis and prescriptive direction The process consultation model where helpers work as facilitators of client inquiry into their own issues and create and implement solutions It is this latter approach that is to be adopted by action researchers. Additionally, the action researcher has the following responsibilities (Burnes, 2004): Development of those involved in the research Creation of a learning environment that allows participants to gain insights into themselves and their environment In support of the methodology chosen for this research study, Nutt (2002) highlights two approaches to arriving at solutions during the decision-making process: Idea imposition, in which an orderly path is seldom followed because decision makers jump to conclusions and then try to implement the solution they stumbled upon (Nutt, 2002: 49). This process lacks proper reflection and thereby offers 19
sub-optimal solutions. Nutt found that solutions arrived at through idea imposition were neither cheaper nor faster Discovery, where decision makers work their way through a process that stresses claim validation, implementation, and direction setting. A premium is placed on learning through the discovery of decision topics, barriers to taking action, and desired results (Nutt, 2002: 46). Discovery thus encourages questioning and seeking the best solution, thereby offering superior results The researcher, within the permitted time frame, endeavoured to achieve superior outcomes by undertaking the Discovery process, concurrently seeking to create a learning environment in which participants themselves could develop further knowledge and insight on their working environment. 5. Hypothesis Testing Research findings and analysis of these findings was undertaken within the context of testing of the various hypotheses suggested within this research paper. Hypothesis testing was guided by the AR methodology steps illustrated in Figure 2. A table following a statement of each of the hypotheses under investigation within this research study intends to illustrate the following information: The stage(s) of the AR cycle during which the hypothesis in question was tested The method(s) applied in testing the hypothesis in question Source material which intends to provide evidence in support of hypothesis testing 20
5.1. Hypothesis 1 H 0 : Value Stream Mapping (VSM) can be used to describe patient flow within the A&E Unit, and can be used to clarify that flow to the significant role players, namely doctors, nurses and support staff Stage(s) Applied in Action Research Cycle Data Gathering Data Feedback Table 1: Testing of Hypothesis 1 Method of Hypothesis Testing Qualitative Feedback on VSM was obtained from role players following Data Gathering during Data Feedback stage of AR cycle Source Material Exhibit 2, Appendix 5.1.1. Value Stream Mapping Value Stream Mapping (VSM) was conducted within the Context of the A&E Unit, with the Purpose of Data Gathering on patient lead time 13 in the A&E Unit (Coughlan & Coghlan, 2002), and was conducted over a two-day period on 20/21 September 2006. Prior to constructing the VSM, however, it was necessary to construct a map of the layout of the A&E Unit to obtain an understanding of the environment in which this research study was conducted (Figure 4). The VSM as at 20/21 September 2006 is graphically illustrated in Figure 5. Key findings from initial VSM are expanded upon below: The Triage 14 section of the A&E Unit consists of the following processes with key findings within these processes elaborated upon in Table 2: o Assessment by a Triage Nurse during which a patient s vital signs 15 are measured, and additional investigations finger prick testing of blood 13 The time it takes one piece to move all the way through a process or a value stream, from start to finish (Rother & Shook, 2003: 21) 14 The section of the A&E Unit comprising the processes of Triage Nurse, Medical Records and Triage Doctor as noted on VSM as at 20/21 September 2006 (Figure 5) 15 Vital signs are comprised of Blood Pressure, Heart Rate, Respiratory (Breathing) Rate & Temperature 21
glucose 16 and Haemoglobin 17 levels, urinalysis 18 for screening of kidney function and pregnancy testing, and ECG 19 are undertaken as appropriate o Processing of a folder with Medical Records for new patients this entails creation of a new folder, and for past patients of the hospital this entails retrieving the patient s existing folder from the archive section at the rear of the Medical Records department o Medical screening 20 by a Triage Doctor who decides on subsequent patient management. Management options include: Discharge of patients from the A&E Unit with or without medication Discharge of patients from the A&E Unit with referral for subsequent management at a Day Hospital Discharge of patients from the A&E Unit with referral to the quent patient appropriate Out-Patients department where subse management options are considered and undertaken Referral of patients to the Emergency Room 21 (ER) section of the A&E Unit for assessment, conduction of special investigations primarily this includes X-Ray investigations and laboratory investigations of blood samples and decision-making on subsequent patient management 16 The elementary form of sugar found in the human body. The level of glucose in the human blood stream acts as an indicator of the level of function of the pancreas, an organ found within the human body 17 The molecule attached to Red Blood Cells within the human body, and which is responsible for transporting Oxygen through the human bloodstream 18 Chemical analysis of urine 19 Also known as Electrocardiography, which tests the electrical activity within the human heart 20 Involves identifying the appropriate subsequent treatment of patients 21 The section of the A&E Unit in which Medical Assessment, Review and Treatment of patients occurs 22
Triage Process Key Findings during Value Stream Mapping on 20/21 September Triage Nurse Pre-Assessment: o 10-20 patients waiting in queue o 10-45 minute waiting period o Patients wait in Triage Waiting Area (Figure 4) Process: o Triage Nurse measures patient s vital signs, and does additional tests/observations as appropriate o Stable Cycle Time 22 6-10 minutes (variability in Cycle Time due to number of additional tests/observations required per patient, as appropriate) Medical Records Pre-Assessment: o 10-20 patients waiting in queue o 25-120 minute waiting period o Patients wait in Triage Waiting Area (Figure 4) Process: o Medical Records Clerk issues patients with a folder in which a record of all activities is kept o Stable Cycle Time 5-10 minutes (variability in Cycle Time due to need to retrieve old folder from archives for patients previously treated at G.F. Jooste) Triage Doctor Pre-Assessment: o 5-10 patients waiting in queue o 30-60 minute waiting period o Patients wait in Triage Waiting Area (Figure 4) Process: o Triage Doctor does medical screening (not treatment) of patients, and decides on further treatment as appropriate o Stable Cycle Time of approximately 10 minutes (variability in Cycle Time due to non-adherence to Rule 1 of TPS: generally due to Triage Doctor treating patients additionally to screening) Table 2: Key Findings in Triage during Value Stream Mapping on 20/21 September The ER section of the A&E Unit consists of a team of doctors (between four and seven doctors) and nurses (between six and seven nurses). The processes within this section are discussed below, with key findings elaborated upon in Table 3: 22 The time, from initiation, it takes to complete a process (Liker & Meier, 2006: 97) 23
o Medical Assessment by a Medical Doctor, assisted by a Nurse, during which a detailed assessment of a patient s presenting complaint(s) is undertaken, their past medical history is evaluated, and a clinical examination is conducted as guided by the patient s presenting complaint(s) and past medical history o The patient is then referred for special investigations, namely X-Rays and/or Pathology Laboratory testing of blood samples. The steps of these parallel processes are: X-Ray: Patients are initially registered within this department with the attendant Radiology Clerk. Thereafter, the patient undergoes the relevant X-Ray investigation, as requested by the ER Doctor, with one of the attendant Radiographers Pathology Laboratory: All blood samples that arrive at the laboratory are first registered onto the department s electronic system by the attendant Laboratory Clerk. Thereafter, each sample is batched (as appropriate) by the nature of testing requested. Most samples undergo centrifuging for approximately 10 minutes. Thereafter, samples are loaded into the appropriate machine for testing by one of the attendant Laboratory Assistants. Testing was measured between 5 and 45 minutes, depending on sample type. Following testing, blood tests need to undergo verification by one of the attendant Laboratory Technicians. Once verified, a test result is loaded onto the hospital s intranet portal from where it can be accessed by the medical staff for review o Medical Review is undertaken by the attendant Medical Doctor once the results of X-Rays and Pathology Laboratory tests are available. A decision is then taken on the further management of each patient following review o For those patients who are to be discharged from the hospital with medication, these medicines are ordered from the Pharmacy. Ordering occurs per individual patient, or in batches of up to five patients medication at a time. Orders are taken by the attendant Pharmacy 24
Assistant and processed by one of the attendant Pharmacists. Hereafter medicines are dispensed to a member of the nursing staff who generally waits for the medication in the case of single patient orders. In the case of batch orders a member of the nursing staff would generally tend to return after a period of between 30 to 60 minutes to collect medication orders o Patients are then either Discharged home, to an alternative department for admission, or to an alternative facility for treatment options that are not offered at G. F. Jooste Hospital Emergency Room Process Medical Assessment Key Findings during Value Stream Mapping on 20/21 September Pre-Assessment: o 10-15 patients waiting in queue o 5-60 minute waiting period o Patients wait in ER in non-designated area (Figure 4) Process: o Medical Doctor assesses patient s presenting complaint, past medical history and conducts physical examination of the patient as guided by presenting complaint and past medical history. Additional procedures are conducted with the assistance of a member of the nursing staff as appropriate o Cycle Time 20-45 minutes (variability in Cycle Time due to complexity of patient, and number of additional procedures conducted on each patient) X-Ray Pre-Assessment: o 0-15 patients waiting in queue o 5-15 minute waiting period Pathology Laboratory o Patients wait in X-Ray Waiting Area (X-Ray department does not form part of the A&E Unit) Process: o Radiology Clerk registers patients for X-Ray o Radiographer performers X-Ray as requested o Stable Cycle Time 5-15 minutes (variability in Cycle Time due to nature and number of X-Ray investigation(s) requested) Pre-Assessment: o 3-5 samples waiting in queue o 15-30 minute waiting period o Samples wait in designated container within ER until collected by a porter for conveyance to laboratory 25
requests) Medical Review Pre-Assessment: Process: o Laboratory Clerk registers sample details onto departmental computer system o Laboratory Assistant(s) centrifuges blood and then loads sample into appropriate machine for testing o Laboratory Technician verifies test result and loads result(s) to hospital intranet portal from where results are accessible to medical staff o Cycle Time 30-120 minutes (variability in Cycle Time due to nature of test required and volume of preceding o 5-10 patients waiting in queue o 30-90 minute waiting period o Patients wait in ER in non-designated area Process: o ER Doctor reviews results of X-Ray and/or Pathology Laboratory investigations o Cycle Time 5-15 minutes (variability in Cycle Time due to nature and number of i requested) Pharmacy Pre-Assessment: o 0-5 patients waiting in queue o 5-20 minute waiting period (includes conveyance) o Patient folders either conveyed immediately to Pharmacy or kept in designated slot in ER before conveyance to Pharmacy by a member of nursing staff Process: o Pharmacy Assistant accepts order(s) for medication from member of nursing staff. o Pharmacist prepares medication order(s) nvestigation(s) o Cycle Time 5-20 minutes (var due to number of medication Discharge Pre-Discharge: o 0-5 patients waiting in queue o 0-60 minute waiting period iability in Cycle Time orders requested) o Patients await discharge by ER Doctor either in allocated bed or designated seat (if available) in ER Process: o ER Doctor discharges patient home, to alternative department for admission, or to alternative facility that offers treatment options that are not offered at GFJ o Cycle Time 1-5 minutes (variability in Cycle Time due to nature of discharge for patients going home this involves instruction on use of medication; for patients 26
referred to alternative department/facility this involves conducting a discussion on the reasons for referral ) Table 3: Key Findings in ER during Value Stream Mapping on 20/21 September A workshop was held on 20 September 2006 (Exhibit 2, Appendix) to introduce key role players to the concept of VSM and the potential benefits of VSM as a tool that could assist initiatives in enhanced service delivery as measured by patient lead time in the A&E Unit. All attendees indicated that they understood the content of a sample VSM as illustrated in Rother & Shook (2003: rear cover) which was shown during the workshop. Furthermore, attendees acknowledged the value of the VSM as a tool that highlighted a broader awareness of all the processes of a particular value stream, and that the VSM could be a tool that aided improvement initiatives. 5.1.2. Conclusion within the A&E Unit, and can be used to clarify that flow to the significant role players, Cannot Reject H 0 : Value Stream Mapping (VSM) can be used to describe patient flow namely doctors, nurses and support staff 27
Men s Toil et Ladies Toil et? Sluice Room Medical Records Records Admin POP Room To X-Ray & Pharmac y Surgical Holding Area Holding Area 35 Seats Resuscitati on Area ECG Room Walk-In Access Window Security Gate Nursing Stati on Medical Holding Area Security Desk Ambulance/Wheelchair Access i?? Ladies Men s Toil et Toil et Booking Unit Clerks Manager Bed Medical Assessment Area Asthma Treatment Area 4 Seats Bed Relatives Waiting Room To Pathol ogy Laboratory Relatives Comfort Room Tri age Waiting Area 37 Seats Porters Lodge Bed Bed Bed Work Stati on Bed Bed Bed Tri age Assessment Area Bed Bed Bed Bed Work Stati on Bed Figure 4: Layout of the A&E Unit Note: Map Not Drawn To Scale
Supplier TAKT = Availabl e Ti me = 1440 min = 7.2 minutes No. of Patients 200 pts Customer Intravenous Infusions X-RAY Medical Complaint I 0-15p 0-15m 2 Rooms 5-6 R 1 C C/T: 5-15m S/U: 0m U/T: 95% I 10-20p 10-45m Tri age Nurs e 1 I 10-20p 25-120m Medical Records 1-2 I 5-10p 30-60m Tri age Doc tor 1-2 I 10-15p 5-60m Assess 6-7 N 4-7 D I 3-5sp 15-30m 1440m Path Lab 08-24h I 2 T 2 A 1 C 5-10p 30-90m Medical Revi ew Ditto I 0-5p 5-20m Pharmac y 7.30-16h 3-5 P 4 A I 0-5p 0-60m Discharge Ditto C/T: 6-10m C/T: 5-10m C/T: 10-30 m C/T: 20-45 m C/T: 30-120 m C/T: 5-15m C/T: 5-30m C/T: 1-5m S/U: 0m S/U: 0m S/U: 0m S/U: 0m S/U: 60m S/U: 0m S/U: 0m S/U: 0m U/T: 100% U/T: 100% U/T: 100% U/T: 100% U/T: 95% U/T: 95% U/T: 100% U/T: 100% 1440m 1440m 1440m 1440m 960m 1440m 510m 1440m 2.4hrs 2.4hrs 1.2hrs 6-10m 5-10m 10-30m 20-45m 30-120m 5-15m 5-30m 1-5m CSM: 20/21 Sept. 06 Key: I = Inventory p = patients m = minutes C/T = Cycle Time S/U = Set-Up Time U/T = Up-Time N = Nurse D = Doctor T = Technologist A = Assistant C = Clerk P = Pharmacist Other symbols used are as shown in Learning to See by Rother & Shook (2003) Figure 5: Value Stream Map as at 20/21 September 2006 29
5.2. Hypothesis 2 H 0 : VSM can be used as a tool to assist improvement initiatives in enhanced patient flow and improved patient service level Stage(s) Applied in Action Research Cycle Data Analysis Action Planning Method of Hypothesis Testing Qualitative Data Feedback on current state VSM was given to role players during Data Analysis and Action Planning stages prior to Implementation of Experiments 1 and 2. This data, in conjunction with VSM, formed a basis for improvement initiatives Source Material Exhibit 2, Appendix Table 4: Testing of Hypothesis 2 5.2.1. Testing of Hypothesis During AR cycle 1 it was necessary to conduct Data Analysis of findings during VSM undertaken on 20/21 September 2006 (Figure 5) prior to Action Planning through the implementation of the philosophies and principles of TPS/Lean. Data Analysis of VSM was initially undertaken with the research supervisor on 12 October 2006 (Exhibit 2, Appendix). At this juncture, given the similar cycle times and relative stability 23 of the processes within the Triage section of the A&E Unit it was decided that Triage would provide an appropriate starting point in testing Hypothesis 3 (Section 5.3) through the implementation of TPS/Lean principles and philosophies. 23 The capability to produce consistent results over time, Instability is the result of variability in your process (Liker & Meier, 2006: 56)
On 16 October 2006 a second workshop was held at G.F. Jooste Hospital. At this workshop the key role players within the Triage section Triage Doctor, Triage Nurses and Medical Records Clerk were given a refresher course on VSM (Exhibit 2, Appendix). Data Feedback on findings during VSM on 20/21 September 2006 was given to role players, and Data Analysis undertaken through the use of the current state VSM. The current state VSM formed the basis of the Action Planning stage during which improvement initiatives of patient throughput times in the Triage section were planned (Section 5.3). During VSM on 20/21 October 2006 it was recognised that the three Triage processes of assessment by the Triage Nurse, creation of a patient folder with the Medical Records Clerk, and assessment by the Triage Doctor were similar in cycle times and relatively stable. Hence, in conjunction with the key role players, focus was placed on efforts to reduce waiting time equated to inventory 24 on the VSM between the various processes, through implementation of TPS/Lean principles and philosophies, to facilitate improved patient lead time in the Triage section of the A&E Unit. 5.2.2. Conclusion Cannot reject H 0 : VSM can be used as a tool to assist improvement initiatives in enhanced patient flow and improved patient service level 24 excess raw material, WIP (Work-In-Progress), or finished goods causing longer lead times, obsolescence, damaged goods, transportation and storage costs, and delay (Liker & Meier, 2006: 36) 31
5.3. Hypothesis 3 H 0 : Application of the principles, philosophies and tools of TPS/Lean thinking can contribute to improved patient flow and patient service level, as measured by reduced time in the system Stage(s) Applied in Action Research Cycle Implementation Evaluation Method of Hypothesis Testing Qualitative & Quantitative Implementation of TPS/Lean was conducted during Experiments 1, 2 & 3. Evaluation was measured quantitatively (patient throughput time) and qualitatively (perceptions of role players participating staff & patients) Source Material Quantitative (Section 6.3) Qualitative (Section 5.3 & Exhibit 2, Appendix) Table 5: Testing of Hypothesis 3 5.3.1. Testing of Hypothesis Hypothesis 3 was tested during the Implementation and Evaluation stages of the AR cycle (Coughlan & Coghlan, 2002) by means of experiments. It is within the context of the three experiments conducted during this research study that findings during AR methodology implementation are expanded upon further within this sub-section: 32
Experiment 1: (Conducted on 16 October 2006) Problem Hypothesis Expected Outcome Waiting time By reducing Improved (10-45 waiting time patient lead minutes between the time in Triage between various (targeted 10% Triage Nurse Triage improvement & Medical processes on range of Records, and patient lead lead time) will 30-60 minutes time can be enhance between significantly service Medical reduced, delivery to Records and thereby patients triage enhancing Doctor) service between the delivery Triage processes retards patient throughput in Type of Waste Waiting Excessive Inventory (patients waiting between steps is equated to Inventory) TPS/Lean Principle/Philosophy Make Value flow (Flow principle) Specify value from a customer perspective Involve & empower employees the system, thereby contributing to suboptimal service delivery Table 6: Outline of Guidelines for Experiment 1 Findings: The experiment was Implemented over a 90 minute period. A time limit was imposed due to the nature of service delivery within the A&E Unit, which provides emergency medical services to patients. Five patients were included in the sample. Unfortunately, a control group could not be arranged for the day of the experiment as other patients were not willing to participate in the experiment. Results were therefore controlled against findings from VSM on 20/21 September 2006 (Figure 5). 33
During Experiment 1 focus was placed on eliminating the waiting times between the various processes within Triage, thereby eliminating much of the non-value added time that patients experience within Triage. Participating staff Triage Nurse, Medical Records Clerk and Triage Doctor were instructed to send patients to the next process Just-In-Time (JIT), thereby avoiding the formation of queues. As the existing sequence of these processes meant that they were not visually connected, the kanban 25 for availability was to be given manually by the researcher who was located at a strategic position which allowed visualisation of all three Triage processes. Findings from Experiment 1 are shown in Table 7: Pt. No. Waiting Time Triage Nurse Waiting Time Medical Records Waiting Time Triage Doctor Throughput TIME 1. Not Measured 8 min. 0 min 8 min. 2 min. 10 min. 28 min. 2. Not Measured 10min. 0min. 8 min. 0 min. 8 min. 28 min. 3. Not Measured 6 min. 0 min. 7 min. 1 min. 7 min. 21 min. 4. Not Measured 6 min. 0 min. 7 min. 2 min. 8 min. 23 min. 5. Not Measured 5 min 0 min. 5 min. 10 min. 11 min. 31 min. Table 7: Findings during Experiment 1 Limitations of Experiment 1 included: Takt Time 26 (calculated at approximately 7.2 minutes during peak periods) was not imposed. The intention hereof was deliberate as the researcher sought to test the stability of process cycle times under conditions of flow. Furthermore, the period of hypothesis testing was outside of peak patient flow. Process times were found to be stable during the experiment when compared to baseline process cycle times Waiting time prior to the first process of assessment by the Triage Nurse was not measured. This is in keeping with the intention to test the concept of flow. The consequence is that patient lead time, as illustrated in Table 7, is not an accurate reflection of throughput from entering the Triage system, rather a reflection of lead time through the various Triage processes 25 Japanese word meaning signal or sign : the kanban is simply the communication method for the customer to say, I am ready for more (Liker & Meier, 2006: 95) 26 The rate of customer demand (Rother & Shook, 2003: 44; Liker & Meier, 2006: 38) 34