CT processes imitated: the development of a simulation model of CT processes at Campbelltown Hospital to improve the ability to anticipate on the impact of changes and to evaluate current practice

CT processes imitated

The development of a simulation model of CT processes at Campbelltown Hospital to improve the ability to anticipate on the impact of changes and to evaluate current practice

Bachelor thesis

Marten de Bruin

1

CT processes imitated

The development of a simulation model of CT processes at Campbelltown Hospital to improve the ability to anticipate on the impact of changes and to evaluate current practice

Bachelor thesis

Author: M.B. (Marten) de Bruin

Student number: s0041580

Date: August 2010

Study program: Bachelor Industrial Engineering and Management

University of Twente

School of Management and Governance

Supervisor

Dr.ir. E.W. Hans, Associate professor

School of Management and Governance

Dep. Operational Methods for Production and Logistics

Project Visual Optimisation of Patient Flow in Hospital Emergency Departments Centre for Industry and Innovation Studies Research Group(CInIS)

University of Western Sydney Locked Bag 1797

Penrith South DC NSW 1797

Australia

2

Management Summary

Hospital emergency departments (EDs) within Australia are under increasing pressure to deliver quality because of a growing demand for their services. The increasing levels of ED activity can result in major problems because the capacity of hospital EDs is limited. Campbelltown Hospital, a public hospital delivering health care services in the south west area of Sydney, is experiencing the increasing levels of ED activity over the last couple of years and is confronted with recurrent access blocks. The diagnostic services delivered by the ID are of high importance to the ED. Stakeholders experience diagnostic services as a barrier for patient flow inside the ED. The Centre for Industry and Innovation (CInIS) from the University of Western Sydney was invited to do research on this problem.

The project undertaken by CInIS is called the Visual Optimisation of Patient Flow in Hospital Emergency Departments. The aim of this project is to optimise patient flow in the ED of Campbelltown Hospital via the analysis of work processes using interactive computing simulation. Previous research within this project has given insight in the processes around Ultrasound (US), and a simulation tool was developed for the US processes to simulate improvement strategies. The next step is to do the same for the other ID services. This specific study focuses on the CT processes at Campbelltown Hospital.

The goal of this study is to provide Campbelltown Hospital with a versatile and comprehensive simulation tool of CT processes that improves the ID manager's ability to anticipate on the impact of changes, as well as to evaluate the effectiveness of the current practices. To achieve this goal, current CT processes were mapped and described using qualitative data. These data were gathered by observation on working days and semi structured interviews with CT radiographers, administrative staff and the manager of the ID. Next, quantitative data was collected over a period of 13 weeks to analyse the current performance of CT processes, and to specify model parameters and input probability parameters for the simulation model. The data was combined and compared with existing data from ward orderly books, CT logbooks and imaging request forms. Finally, a computer model was constructed and validated.

In conclusion, we partly succeeded in providing Campbelltown Hospital with a versatile and comprehensive simulation tool of CT processes that improves ID manager's ability to anticipate on the impact of changes, as well as to evaluate the effectiveness of the current practices. Changes in operating procedures can easily be adapted in the computer model and therefore it becomes easy to experiment with different configurations. On the other hand, we failed in introducing the factors that result in the current waiting times, which makes it impossible to evaluate the effectiveness of all of the current practices.

We recommend that upfront processes of CT, especially (re)scheduling and transportation, are extensively analysed in order to create insights that can be used to improve the simulation model and to create alternative configurations which can be compared. The current model shows us that the available CT capacity is well enough to meet the current demand. Waiting times are mainly a result of the current system of planning and scheduling.

As demonstrated within the simulation model, the introduction of a walk-in principle at the ID is a possible solution to lower waiting times substantively. Boundary conditions for the introduction of such a system therefore have to be explored. When integrated in the model, their influence can also be quantified. Maybe even less rigorous interventions in the current system of planning and scheduling can also have a major impact on the current waiting times. But if this system remains undefined, no alternative configuration can be created in the simulation model and no conclusions can be drawn about the effectiveness of possible improvements in this system.

At the start of this study it was known that the model created during this study will eventually be combined with simulation models from the other diagnostics services delivered by the ID. During the integration of the different models it is important to include shared departmental facilities, like patient transportation by wardsmen and the availability of nurses. Since more variables are introduced for which different configurations can be simulated, this will improve the ability to anticipate on the impact of changes, as well as to evaluate the effectiveness of the current practices.

3

Table of contents

Management Summary ... 2

Preface ... 5

Chapter 1: Introduction ... 6

1.1 Context ... 6

1.1.1 Australian Health Care System ... 6

1.1.2 Sydney South West Area Health Service ... 7

1.1.3 Campbelltown Hospital ... 7

1.2 Problem description ... 9

1.3 Research goal & research questions ... 9

1.4 Methodology ... 10

Chapter 2: Current situation ... 11

2.1 System layout and operating procedures ... 11

2.1.1 Patient flow process ... 11

2.1.2 Available resources ... 13

2.1.3 Lay out of the ID ... 14

2.2 Planning and control ... 14

2.3 Performance ... 15

2.3.1 Number of CT exams done ... 15

2.3.2 Patient type frequencies ... 16

2.3.3 Exam type frequencies ... 16

2.3.4 Number of CT requests... 17

2.3.5 Waiting time ... 18

2.3.6 Summary... 19

Chapter 3: Simulation model ... 20

3.1 Conceptual model ... 20

3.1.1 Scope of the model... 20

3.1.2 Specification of model parameters ... 22

3.1.3 Input probability distributions ... 22

3.1.4 Remaining assumptions... 26

3.1.5 Performance measures ... 27

3.1.6 Is the conceptual model valid? ... 28

3.2 Construction of a computer program ... 28

3.2.1 Software to be used ... 28

3.2.2 Verification of the model content ... 28

3.2.3 Description of the programmed model ... 28

4

3.3 Model validation... 30

3.3.1 Number of CT requests... 30

3.3.2 Arrival times of CT requests ... 30

3.3.3 Number of CT exams done ... 31

3.3.4 Patient type frequencies ... 32

3.3.5 Exam type frequencies ... 32

3.3.6 Waiting time ... 33

3.3.7 Conclusion ... 33

Chapter 4: Conclusions ... 34

Chapter 5: Recommendations ... 36

References ... 37

Appendices ... 38

Appendix 1: Imaging request and patient flow process ... 38

Appendix 2: Resources Imaging Department ... 42

Appendix 3: Map of the imaging Department ... 43

Appendix 4: CT exams per patient type ... 44

Appendix 5: CT codes ... 45

Appendix 6: Interarrival times of CT requests ... 46

Appendix 7: Scanning time data ... 48

Appendix 8: CT Finishing time ... 61

5

Preface

With this thesis I will complete my bachelors degree in Industrial Engineering and Management at the University of Twente. My interest in the field of optimization of healthcare processes opened the door for a trip to Australia, in which research could be combined with a cultural exploration of the continent.

The Centre for Industry and Innovation Studies (CInIS), a research group of the University of Western Sydney, has provided me the opportunity to assist them on the project Visual Optimisation of Patient Flow in Hospital Emergency Departments. I thank the team members for their support and trust during the 10 weeks that I was part of the team. The Wednesdays at Paramatta were always inspirable.

During my graduation project I have held office at Campbelltown campus. I have shared workplaces with PhD students, who have created an ambiance of performance, creativeness and fun. Without them my stay would not have been so productive and spectacular at the same time. I thank Kathy, Geoffrey, Cassandra, Ryan and Nazlee for making this trip to Australia an incredible experience. During working hours and after working hours, you guys showed me Australian hospitality and I loved it.

My research comprehended interaction with the personnel of the Imaging Department at Campbelltown Hospital. If they would not have cooperated in such a positive way, I would have left with nothing. I thank all of them for their valuable time and proactive role in this research.

A special word goes out to Anneke. Next to supervising me abroad, she was able to make me feel at home. The family diners for which I was invited many times at her place, were always easy going. The places she and her family took me too were amazing. Sadly, during the writing of this report her husband John passed away. I wish Anneke, Kim, Chris and Michael a lot of strength in this difficult time. John was an amazing person and will be deeply missed.

Finally I thank Erwin for being my supervisor from the University of Twente. Your comments on the thesis and constructive feedback were very helpful. And also your personal interest, patience and motivating words were of major importance to me.

This leaves me with nothing but to wish you a pleasant reading.

Marten de Bruin

Veenendaal, August 2010

6

Chapter 1: Introduction

Hospital emergency departments (EDs) within Australia are under increasing pressure to deliver quality because of a growing demand for their services. The increasing levels of ED activity can result in major problems because the capacity of hospital EDs is limited (Booz Allen Hamilton, 2007). The limited capacity is evidenced by access blocks (ACEM, 2004; Cameron & Campbell, 2003). An access block occurs when a patient remains in an ED for over eight hours (Paoloni & Fowler, 2008).

Campbelltown Hospital, a public hospital delivering health care services in the south west area of Sydney, is experiencing the increasing levels of ED activity over the last couple of years and is confronted with recurrent access blocks. The Centre for Industry and Innovation Studies (CInIS), a research group of the University of Western Sydney, was invited to do research on this problem.

The project undertaken by CInIS is called the Visual Optimisation of Patient Flow in Hospital Emergency Departments. The aim of this project is to optimise patient flow in the ED of Campbelltown Hospital via the analysis of work processes using interactive computing simulation. Previous research within this project has given insight in the processes around Ultrasound (US), and a simulation tool was developed for the US processes to simulate improvement strategies (CInIS, 2009). The next step is to do the same for the other ID services, like X-ray and CT. This specific study focuses on the CT processes at Campbelltown Hospital.

This chapter is an introduction to the report and will start with a description of the research context (1.1). In section 1.2 we elaborate on the problem, leading to the research goal and research questions that will guide this study (1.3).

1.1 Context

This section starts with a general introduction to the Australian health care system to create some understanding about the way things are organized. Via the Sydney South West Area Health Service, one of the most populous Area Health Services in the state New South Wales and in which Campbelltown Hospital is operating, we will come to a description of Campbelltown Hospital and the departments that are relevant for this research.

1.1.1 Australian Health Care System

Australia, the smallest continent of the world, has a total population of around 21.5 million people (WHO, 2009). The life expectancy at birth is 79.0 years for men and 83.7 years for women. Most of the population is concentrated in large cities along the eastern side and the south-eastern corner of the continent; 88,6 percent of the population is living in urban areas. Like in many countries Australia’s population is ageing, with the number of people aged 65 years or more projected to almost triple by 2051, an increase from 13.2% to 22.3%

of the total population.

The Australian health care system is a partnership between the federal government, state and territory governments (WHO, 2009). Responsibility and funding for health is shared between them, and health care is provided by both public and private sectors. The federal government of Australia is the major funder of health services and responsible for national policies and outcomes. The state and territory government are mainly responsible for providing public hospital services, community and public health services and the regulation of private hospitals.

The aim of the Australian health care system is to give universal access to health care (WHO, 2009). This is done by what is known as ‘Medicare’. But individuals still have a choice through substantial private sector involvement in delivery and financing. Medicare is a compulsory insurance system financed by progressive income tax and an income-related Medicare levy. Medicare provides access to free treatment as a public (Medicare) patient in a public hospital, and free or subsidised treatment by medical practitioners including general practitioners, specialists, participating optometrists or dentists (for specified services only).

The Australian Government funds a system of private health insurance rebates that subsidize the cost of premiums to private health insurance (WHO, 2009). Every Australian can decide to be treated as a private patient in a public hospital to have a choice of doctor. For many procedures the private hospitals provide an alternative to the public hospital system.

7 1.1.2 Sydney South West Area Health Service

Campbelltown Hospital, a public hospital, is part of the Sydney South West Area Health Service (SSWAHS) and this is currently the most populous and ethnically diverse area health service in New South Wales (SSWAHS, 2009a). New South Wales is one of the 6 states of Australia. The SSWAHS covers a land area of 6.380 square kilometres and in 2006 had an estimated residential population of 1.340.378 residents. The population of the SSWAHS is projected to reach almost 1.5 million people by 2016, an increase of 11 percent over the next 10 years. The region is also characterised by high levels of unemployment and encompasses 9 suburbs that are in the 30 lowest socio-economic suburbs in NSW, and in the 15 most disadvantaged suburbs in metropolitan Sydney.

Serving a highly populated community, the SSWAHS emergency departments have collectively experienced significant increases in demand (SSWAHS, 2009a). From 2006-07 to 2007-08, the SSWAHS witnessed:

An increase of ED presentations from 310,822 in 2006-07 to 327,945 in 2007-08; an increase of 6 percent (or 17 percent over the last two years)

A total of 79.352 ED admissions; an increase of 1 percent (or 10 percent over the last two years) The highest number of ambulance presentations (110.408); an increase of 6 percent (or 22 percent over the last two years)

Despite the increase in demand, the percentage of emergency patients transferred to an inpatient bed within 8 hours has been maintained at 75 per cent in 2007-08 (SSWAHS, 2009a). The target was set at 80 per cent. This performance indicator is called the Emergency Access Performance (EAP) and is considered to be very informative since Paolini & Fowler (2008) showed that access block occur when patients remain in an ED for over eight hours.

1.1.3 Campbelltown Hospital

Campbelltown Hospital is a major metropolitan hospital providing a diverse range of services including intensive care, cardiology, maternity, gynaecology, paediatrics, palliative care, respiratory and stroke medicine, surgery and emergency medicine and broad aged care services (SSWAHS, 2009a). Campbelltown Hospital is a part of the Macarthur Health Service which includes Camden Hospital and the Queen Victorian Memorial Home (SSWAHS, 2009b). The Macarthur area includes Campbelltown, Camden and Wollondilly Local Government Authorities and incorporates both metropolitan and rural settings. It covers 3,070 square kilometres and has a population approaching 250,000 residents. Campbelltown Hospital has approximately 389 beds and has approximately 26,000 inpatient admissions per year. Emergency and Imaging departments work across both Campbelltown and Camden Hospitals with staff rotating across both sites.

Within this paragraph, we will take a closer look at the Emergency Department and Imaging Department of Campbelltown Hospital, which is the subject of this research.

Emergency Department

The Emergency Department of Campbelltown Hospital, being one of the bigger SSWAHS emergency departments, has experienced an increase of ED presentations from 44.276 in 2006-07 (SSWAHS, 2008) to 47.210 in 2007-08 (SSWAHS, 2009a). This increase of 6,6 percent is a little above the average of the SSWAHS emergency departments together.

A more detailed look at the performance of the Campbelltown ED, derived from the “2009 Campbelltown Hospital Experience Based Co-design Project (SSWAHS, 2009b), shows us that within 2007-08 they witnessed:

A relatively stable level of ED attendances between 3,500 to 4,300 per month

50 to 70 percent of emergency patients were transferred to an inpatient bed within 8 hours Admission of approximately 25 percent of ED attendances to an inpatient unit

A fairly consistent level of ‘Did Not Wait’ in the ED at around 5 percent

Activities in the ED involve examining a presenting patient, run diagnostics tests, commence initial treatment and determine whether the patient is a likely admission. The diagnostic services delivered by the Imaging Department are thereby of high importance to the ED.

8 Imaging Department

The Imaging Department (ID) is a service department to the other departments within the hospital, including the Emergency Department. Medical imaging is the technique and process used to create images of the human body (or parts and function of it) for clinical purposes (medical procedures seeking to reveal, diagnose or examine disease) or medical science (including the study of normal anatomy and physiology). In the clinical context, medical imaging is generally referred to as radiology. The medical practitioner responsible for interpreting the images is a radiologist. The radiographer is responsible for acquiring the medical images of diagnostic quality.

The ID at Campbelltown Hospital offers the following three services:

X-ray - 3x

Ultrasound (US) - 3x

Computed Tomography (CT) - 1x

Table 1 gives an overview of the number of the specific ID services provided. Mobile X-rays are the X-rays performed with a mobile X-ray machine, like the ones performed within operating theatres. The category

‘Other’ comprises interventional procedures in which minimally invasive procedures are performed using image guidance, fluoroscopy exams, and exams performed within the operating theatres.

year Xray CT US Xray (mobile) Other

2005/2006 23038 4472 4023 5923 1456

2006/2007 24093 5942 4263 6548 1750

2007/2008 25988 6528 4352 7603 1740

2008 Jul to Dec 13882 3515 2275 3823 897

Table 1: ID services provided

(source: Co-design Medical Imaging Data until 31-12-2008, N=152.111)

The proportion of referrals to the Imaging Department from the Emergency Department within Campbelltown Hospital have been increasing to a point where they are now over 60% of all Radiology referrals. Table 2 shows the rising demand for ID services over the last couple of years.

# exams # patients # exams # patients % of total demand

2005/2006 38912 32940 18387 14645 47

2006/2007 42596 36827 21918 18091 52

2007/2008 46211 39196 25967 21839 56

2008 Jul to Dec 24392 20677 15269 12923 63

total demand for ID services ED demand for ID services year

Table 2: Demand for ID services, including the specific demand by the ED (source: Co-design Medical Imaging Data until 31-12-2008, N=152.111)

9

1.2 Problem description

As can be derived from section 1.1.3, Campbelltown Hospital has experienced an increase of ED presentations over the last couple of years. As a consequence, it is confronted with recurrent access blocks. It is clear that the Emergency Access Performance level of Campbelltown Hospital between 50 and 70 percent is worrisome, and well below the SSWAHS average of 75 percent. It is a source of frustration among staff, as well as a cause of dissatisfaction among patients.

The diagnostic services delivered by the ID are of high importance to the ED. Stakeholders experience diagnostic services as a barrier for patient flow inside the ED. The proportion of referrals to the Imaging Department from the Emergency Department within Campbelltown Hospital have been increasing to a point where they are now over 60% of all Radiology referrals.

The improvement of the patient flow between the two departments can be used as a weapon to battle with the recurrent access blocks.

1.3 Research goal & research questions

The goal of this study is:

To provide Campbelltown Hospital with a versatile and comprehensive simulation tool of CT processes that improves the ID manager's ability to anticipate on the impact of changes, as well as to evaluate the effectiveness of the current practices.

We achieve this goal by addressing the following research questions:

1. What is the description of the CT process, its control and its current performance?

An in-depth understanding of the CT process in the ID is needed to develop a simulation tool. For that matter, an analysis of the current situation of the ID will be made to give a clear insight of “how things work”. Section 2.1 will reveal the way in which space, equipment, staff, and patients are assembled to maximize patient flow. In section 2.2 the planning and control of the CT process is discussed. The current performance of the CT process is described in section 2.3.

2. What are the existing measurements points within the process, are they appropriate and if not, what are the appropriate new ones?

As a result of the process analyses key data elements (activity, material and resource) and input/output parameters associated with various functions and events are identified. From the analyses performed in section 2.1 it becomes clear what the existing measurement points are and which data elements need to be collected.

3. How to develop a valid simulation model, which is versatile and comprehensive?

The approach taken within this study to develop a valid simulation is based on the steps described by Law (2007). Based on the CT process information provided in chapter 2, a conceptual model is defined in section 3.1. This conceptual model is then used to construct a computer program, which is explained in section 3.2. Model validation will be performed in section 3.3. A valid model is both accurate and able to meet the objectives of the simulation project for which it is being used. The purpose of validation is to guarantee the correct degree of accuracy by checking that the overall behavior of the model is representative for the real situation.

10

1.4 Methodology

To gain an in-depth understanding of the processes within the CT department, patient journeys were mapped using flowcharts. This step included the observation of document flow and the patient journey from three access points into the CT department: access as an outpatient, access as an inpatient, and access via the Emergency Department. Qualitative data was gathered by observation on working days and semi structured interviews with CT radiographers, administrative staff and the manager of the ID between March and May 2009.

Next, quantitative data was collected to analyse the current performance of CT processes and to specify model parameters and input probability parameters for the simulation model. This data was collected by radiographers inside the CT room over a period of 13 weeks (between 17 March and 15 June 2009). Explorative analysis were performed during this period and after 9 weeks of data characteristics were not changing significantly. Another 4 weeks were added to be sure of the representativeness of the data. The collected data was combined and compared with existing data from ward orderly books, CT logbooks and imaging request forms. The ward orderly book records the time into department, the time out of CT room and the time out of department of the in- and emergency patients. The CT logbook is an official document that is used to register exam specific information for each patient. Imaging request forms contain information about the patient and the requested exam.

Timestamps of patients collected from this documentation included:

Request time of CT exam (request form) Booking time (request form)

Scheduled time (ward orderly book)

Arrival time of patient in the waiting room (ward orderly book) Arrival time of patient in CT room (data collection sheet)

Stop time CT scan - patient leaves CT table (data collection sheet) Stop time post-processing CT images (data collection sheet)

Departure time of patient out of the CT room - entering waiting room (ward orderly book) Departure time of patient out of the department (ward orderly book)

To verify and validate the primary data collected, it was cross-checked with CT radiographers, wardsmen, administrative staff and the manager of the imaging department on several occasions. In addition, the mapped process were also confirmed by all CT staff.

To prepare the data for analysis, the primary and secondary datasets were coded, cleaned, and integrated using Microsoft Excel. Following this, the collected data were cleaned by excluding illegible, unclear or illogical records (for instance, improbable scanning times, as confirmed by the radiographers), and completing missing data by comparing the different data sources. The resulting dataset consists of 1395 cases, representing the total number of CT scan done between 17 March and 15 June 2009.

An analysis of the dataset has been performed in order to find patient type frequencies (in-/out-/emergency patients), exam frequencies, and probability distributions for the interarrival time between CT requests and contact times.

After the data collection and analysis was finished, a simulation model of the current situation was built using the commercial software Plant Simulation. The model was validated by comparing the model with reality.

11

Chapter 2: Current situation

Following the description of the research environment and introduction of the methodology used to create a simulation model for the CT processes at the ID of Campbelltown Hospital, this chapter will provide an overview of the system layout and operating procedures (2.1), the planning and control of CT exams (2.2) and the performance of the existing system (2.3).

2.1 System layout and operating procedures

An in-depth understanding of the processes in the ID is needed to develop a simulation tool. For that matter, an analysis of the current situation of the ID will be made to give a clear insight of “how things work” at the ID The results of the analysis consists of two parts: a description of the patient flow process and a description of the available resources inside the ID.

2.1.1 Patient flow process

The patient flow process was mapped using Event-driven Process Change (EPC). The EPC methodology incorporates functions and events as key process elements, which are connected with logical operators (AND,OR and XOR) to form business process models (Curan, Keller & Ladd, 1998; Sandoe, Corbitt & Boykin, 2001). Business process models, in the form of EPC diagrams, are connected with process paths to form process cycles across relevant departments of the organization. The EPC diagram of the patient flow process is presented in appendix 1 and consists of 4 consecutive pages. This section will disclose the content of the EPC diagram.

Booking and scheduling of CT requests

The process starts with an incoming request for CT imaging from outside the department. The request comes in via the intranet application of the Imaging Department (PowerChart) and will automatically be printed out at the central workplace of the imaging department near the X-ray workstations. CT requests from outside the hospital arrive via the fax machine.

The request form is brought to or collected by a receptionist for prioritization and booking. In most cases a timestamp will be placed on the request form by a receptionist to mark the time that the form is available to her. During normal working hours all CT exam request forms are then vetted by a registrar/radiologist except for the non contrast brain exams (XB). These are given to the CT area to be done as soon as possible because of the medical priority of these exams. During afterhours CT requests are to be approved by the registrar/radiologist except for non contrast CT exams.

The booking will be made by a receptionist based on the requested type of scan (with or without contrast enhancement), priority and resource availability. As a result of this a confirmation form is created and send to the referring doctor, and an appointment time is available in the scheduling function of PowerChart. However, in certain trauma cases, the paperwork is done later and the patient immediately goes to the preparation phase if needed.

In practice, the initial planning for the CT scanner will be subject to changes. Reasons for rescheduling CT exams, which were distilled from interviews with CT radiographers and a receptionist, are:

cancellation of an exam

an arrival of a new emergency patient for whom a CT request is made a previous CT scan is completed earlier than the scheduled time next scheduled CT exam is delayed by the current CT exam

a scheduled CT exam is postponed because a patient is not ready for his/her scan

the priority of an exam has changed, requiring that the scheduled CT exam has to be carried out earlier

12 Preparation and transportation

From the most current schedule available, the wardsmen have to pick up the patient at their ward before their appointment time and transport them to the imaging department. Wardsmen only have to focus on inpatients and emergency patients because outpatients arrive by means of own transportation.

Inpatients and emergency patients

Inpatients and emergency patient are prepared for their CT exam at the ward by nurses, including transport.

When contrast enhancement is needed during the exam a patient can be required to have an intravenous access line. Other CT studies require oral contrast, which is administered via a drink. The necessary information for the ward/ED is provided by the wardsmen of the ID. Patients must have the oral contrast administered one hour before the procedure, which is the optimum time for the scanning procedure. IV contrast enhancement studies do not have time restrictions. The patients just have to have an intravenous access line before transportation to the ID takes place. The only exception is the XCHO exam. When this exam is planned, a contrast preparation should be administered intravenously 45 minutes before the scan can start.

Using the Pick-up Form, which was created by the receptionist in an earlier stage, a wardsperson will leave the imaging department in order to collect the patient. The wardsperson will transport the patient from the ward to the waiting room at the imaging department with help of a wheelchair, bed or just by walking, depending on the physical capabilities of the patient.

A log named the ‘ward orderly book’ is used by the wardsmen to record the time that inpatients and emergency patients arrive at the waiting room at the department. It is also used to record the time that these patients leave the CT room (and enter the waiting room again) after finishing their CT scan, and the time that they actually leave the department. In the ward orderly book there is also a timestamp available about the (in some cases updated) appointment time.

Outpatients

Outpatients will arrive at the outpatient waiting room by means of own transportation. In this case there is no involvement needed from wardsmen, and therefore no registration takes place in the ‘ward orderly book’ for these type of patients. When an outpatient requires preparation for a contrast study (intravenous access line, drink, etc.) this will be done by a nurse at the department after the patient’s arrival.

CT exam

A nurse will collect the patient from the waiting room and transport the patient to the CT room when the scanner is free. The patient will be placed and positioned on the CT table by the nurse and CT radiographer, depending on the requested CT scan. In case of a contrast study the contrast infusion pump will also be connected to the patient’s intravenous access line. The request form which contains the necessary information about the exam, is available to the CT radiographer.

When the patient is ready the actual scan is made. Depending on the radiographer’s opinion about the quality of the images and in some cases the opinion of the referring doctor, it is decided if the scan is successful or not.

If not, it can happen that the scan has to be redone at the same time, or another appointment has to be made.

If the scan is done the nurse and radiographer will help the patient to get back into their bed/ wheelchair if necessary, and transport him/her back to the waiting room. Wardsmen are responsible for taking the patient from the waiting room back to his/her ward

The nurse will make the CT room ready for the next patient. This involves changing the linen sheet on the CT table, removing used contrast media and installing new if necessary.

Image reconstruction and reporting the findings

After the scan has been performed, the images from the CT exam have to be reconstructed by the CT radiographer before they can be judged by a radiologist. The reconstruction, also called post processing, is performed within the control room of the CT scan. After the post processing phase, the images are uploaded to

13

the intranet of the Imaging Department and a radiologist is then able to make a judgment about them. The radiologist uses a dictaphone in order to make up the report. Next the dictaphone will be given to a typist who will finish the report. The final check on the typed report is performed by the radiologist again and then the report will be made available to the referring doctor.

2.1.2 Available resources Employees

The staff of the ID consist of radiologists, radiographers, nurses, wards person, receptionists and two departmental managers. In the previous section some of their tasks were already disclosed. In this section we are mainly interested in the staff that is directly involved in the processes inside and outside the CT room.

Booking, scheduling and reporting procedures will not be included in the simulation model.

CT radiographers

CT radiographers are responsible for acquiring the medical images of diagnostic quality. Their task is to place and position patients on the CT table when they enter the CT room, acquire the images, and reconstruct the images before they can be judged by a radiologist.

From Monday until Friday the CT room is staffed between 8.30 and 21.00 hrs. This time span is covered by two radiographers; a shift from 8:30 to 16:30 and a shift from 13:00 to 21:00 hrs. The radiographer on the first shift receives a 45 minute lunch break when the second shift begins. During afterhours and the weekend there is no special dedication of a radiographer to the CT room, but there is always a radiographer present inside the ID to perform CT exams.

Nurses

Nurses support the clinical processes in the ID. Their tasks involve the preparation of outpatients for contrast enhancement if needed. This is done via an intravenous access line or providing the patient with a drink containing oral contrast, depending on the requested exam. Inpatients and emergency patients are already prepared at the wards, so a formal check on these patients should normally be enough. Next to this nurses arrange the transport of patients from the waiting room to the CT room. Inside the CT room they assist the CT radiographer with placing and position the patient on the CT table. After the scan has been performed nurses transport the patient back to the waiting room and they will make the CT room ready for the next patient. This involves changing the linen sheet on the CT table, and cleaning and preparing the contrast infusion pump. 24 hours a day, 7 days a week nurses are available at the ID.

Wardsmen

Wardsmen, also called ward orderlies, are responsible for the transport of inpatients and emergency patients.

The transport of these patients takes place between the wards/emergency department of the hospital and the Imaging Department. Their primary task is to make sure that the patients are in time before they have an appointment. The Imaging department of the Campbelltown hospital employs 5 wards person, which do not need medical education. Wards persons have to be in the department 7 days a week. Normal working hours are from 9:00 to 17:00 hrs. In addition to those hours, 1 WP is in the ID from 8:00 and 1 (other) WP will be there until 20:00. Transportation during the afterhours is done by other staff members of the hospital. Because this is not their primary task this may result in delay.

Rooms

The ID uses several spaces to deal with patients that will receive a CT scan. Dedicated waiting room capacity is created for 2 CT beds and 13 seats are available in the general waiting room (appendix 2). Next to the CT room, space is available for certain clinical procedures that have to be performed by nurses inside the department, like inserting the intravenous access line for contrast enhancement if needed. There is one CT scanner available at the ID inside a dedicated room. All the material needed to perform a CT scan (including contrast media) are present in this room. The CT room is in direct contact with the control room; the working area for the CT radiographers. Inside the control there is one workstation available for controlling the scanner. Post-processing of the images can be done on this workstation after a scan has been performed. A second workstation is available for post-processing only. When the first station is busy controlling the scanner, the second station can simultaneously be used to post-process images from previous exams.

14

The CT room is available 24 hours a day, 7 days a week. Maintenance of the CT scanner is performed twice a year and will consist of a half day work each time.

2.1.3 Lay out of the ID

The ID is located at the ground floor of Campbelltown Hospital and the emergency department is next door. A map of the department is included in appendix 3. It is very difficult to keep good visual control at the ID. This is mainly caused by the architectonically lay-out of the ID. This causes a poor view on patients in the seated waiting room, current patients, colleagues, etc.

2.2 Planning and control

The planning and control of CT exams is entirely dedicated to the personnel working inside the ID. With respect to the planning and control of CT exams, necessary decisions are made by different people inside the department. An important distinction needs to be made between clinical and logistical decision making. Within this section both types will be discussed.

Clinical decision making

During normal working hours an incoming request for a CT exam, except for the non contrast brain exam, is vetted by a radiologist/registrar. The radiologist/registrar determines if the requested CT exam is justified from a clinical point of view, and he/she allocates a priority level to the requested exam. During afterhours only requests for CT exams involving contrast enhancement are to be approved by the registrar/radiologist. Non contrast brain exams (XB) are never vetted by a radiologist. These exams have to be done as soon as possible and therefore take priority over all requested exams.

Logistical decision making

After the approval and prioritization of a CT request by a radiographer, a receptionist at the ID is responsible for scheduling. Every booking they make is based on the requested type of scan (with or without contrast enhancement), the priority of the exam, and the availability of resources. Every booking will result in a appointment time, and all these appointment times together will result in the CT schedule. When the schedule is made, wards person are responsible for the transport of in- and emergency patients to the ID, prior to their appointment time.

In practice, the schedule for the CT scanner will be subject to changes during the day. Reasons for rescheduling CT exams, which where distilled from interviews with CT radiographers and a receptionist, are:

cancellation of an exam

an arrival of a new emergency patient for whom a CT request is made a previous CT scan is completed earlier than the scheduled time next scheduled CT exam is delayed by the current CT exam

a scheduled CT exam is postponed because a patient is not ready for his/her scan

the priority of an exam has changed, requiring that the scheduled CT exam has to be carried out earlier The radiographer(s) and wards person on duty will consult together how the rescheduling takes place during the day. The consultation normally takes place inside the CT control room.

15

2.3 Performance

This section will extent the current analysis of CT processes, available resources, and planning and control at the ID, and focus on the overall performance of this subsystem. First we will present the number of CT exams done (2.3.1). Second we will give information about patient type frequencies (2.3.2) and exam type frequencies (2.3.3). Next the number of CT requests per day will be presented (2.3.4). And to create insight in the current patient flow, the waiting times for CT exams will be analysed (2.3.5).

2.3.1 Number of CT exams done

The average number of CT exams done on a weekly basis is shown in figure 1. As can be seen from the figure the number of CT exams done differs considerably between normal working days (Monday until Friday) and the days of the weekend. This is explained by the fact that during the weekends no CT exams are scheduled, but immediately requested and necessary CT exams can be done. From figure 1 we can also see that the number of exams on Mondays and Wednesdays is less than the other days of the week. The difference is partly explained by the fact that on Wednesdays mornings procedures (XFNA exams) have been scheduled. These procedures take more time than standard CT exams.

As described earlier (2.1.2) the CT room is available for examining patients 24 hours a day, 7 days a week. From Monday until Friday the CT room is always staffed between 8.30 and 21.00 hrs. Therefore these periods of time are defined as normal working hours. The remaining periods, including the weekend, are defined as after hours. During afterhours exams are performed, but there is no special dedication of a CT radiographer to the room. Table 3 represents the number of scans done daily during normal working hours and after hours. A total of 1159 cases was used for analysis purposes; within 236 of the 1395 cases included in the dataset, information about the starting times of the exams was missing. The table shows that around 90% of the scans performed on Monday until Friday are done during normal working hours. When the weekend is also taken into account, 76,5% of the total number of exams is performed during the hours that a special dedication of radiographers to the CT room exist.

Table 3: Number of exams done during normal working hours and after hours (source: CT data 17 March until 25 May 2009, N=1159)

Normal hours Total

08:30-21:00hr

Monday 153 91,1% 15 8,9% 168

Tuesday 200 92,2% 17 7,8% 217

Wednesday 163 89,6% 19 10,4% 182

Thursday 192 93,2% 14 6,8% 206

Friday 179 93,2% 13 6,8% 192

Saturday 0 0,0% 91 100,0% 91

Sunday 0 0,0% 103 100,0% 103

Total 887 272 1159

% After hours % Figure 1: Average number of CT scans

(source: CT data 17 March until 25 May 2009, N=1395)

0,00 5,00 10,00 15,00 20,00 25,00

Number of CT scans per day

Average

16 2.3.2 Patient type frequencies

CT exams are performed on three categories of patients, which are: emergency patients, inpatients and outpatients. The data indicated that 58 percent of the CT exams was performed on ED patients were, while 36 percent was performed on inpatients and 6 percent on outpatients. So the largest part of the CT exams (58%) is delivered to patients that are unplanned and usually urgent.

Table 4: Patient type frequencies

(source: CT data 17 March until 25 May 2009, N=1383)

2.3.3 Exam type frequencies

Patients receive different types of exams because of the differences in clinical purpose. The total frequencies of the CT exams performed at Campbelltown Hospital are presented in table 5 and the frequencies of CT exams per patient type are given in table 6. Exams performed less than 20 times were combined in the category other.

The reason why this was done will explained later on. The extensive list of CT exams done is presented in appendix 4.

Exam type Frequency Percentage Cumulative

XB 579 42,08% 42,08%

XAPC 249 18,10% 60,17%

XSPA 111 8,07% 68,24%

XAP 84 6,10% 74,35%

XNPC 45 3,27% 77,62%

XCAC 35 2,54% 80,16%

XBC 30 2,18% 82,34%

XEX 24 1,74% 84,08%

XFNA 21 1,53% 85,61%

XCA 21 1,53% 87,14%

OtherEM 78 5,67% 92,81%

OtherIN 82 5,96% 98,76%

OtherOUT 17 1,24% 100,00%

Total 1376 100,00%

Table 5: Frequencies of all CT exams

(source: CT data 17 March until 15 June 2009, N=1376)

Exam type Percentage

XB 56,95%

XAPC 13,27%

XSPA 7,63%

XAP 7,63%

XBC 2,00%

XEX 1,50%

XNPC 0,63%

XCAC 0,50%

XCA 0,13%

Other 9,76%

Emergency patient

Exam type Percentage

XAPC 27,33%

XB 24,09%

XSPA 9,11%

XAP 4,66%

XCAC 4,66%

XNPC 4,05%

XCA 3,44%

XBC 2,83%

XEX 2,02%

XFNA 1,21%

Other 16,60%

Inpatient

Exam type Percentage

XNPC 24,10%

XFNA 18,07%

XAPC 9,64%

XCAC 9,64%

XB 6,02%

XSPA 6,02%

XCA 3,61%

XEX 2,41%

Other 20,48%

Outpatient

Table 6: Frequencies of CT exams per patient type (source: CT data 17 March until 15 June 2009, N=1376)

As can be seen from the tables above the exams are coded. A list explaining the different codes being used is presented in appendix 5. The exam codes ending with a ‘c’ are the CT exams that need contrast enhancement.

Contrast can be administered in two different ways: intravenously (IV) during the exam or by consuming a drink containing the contrast media.

Patient type Number Frequency

IN 498 36,01%

OUT 83 6,00%

EM 802 57,99%

1383 100,00%

17 2.3.4 Number of CT requests

Section 2.3.1 provided us already with general information about the number of CT exams done on a daily basis. To get a better idea about the demand for CT exams, the number of CT requests per day over a period of 10 weeks (17 March until 25 May 2009)from the dataset was analysed. Within 104 of the 1142 cases included, information about the CT request date was missing. This was caused by the unavailability of some CT request forms within the Imaging Department. Predictive replacement was used to salvage the dataset because of the large number of missing data and the possible influence this could have on the outcome of the analyses.

Missing request dates were replaced by the dates on which the actual exam took place; a reliable estimator considering the departmental activities. Table 7 gives an overview of the number of CT requests per day across the 10 weeks of data.

Table 7: Number of CT request per day across 10 weeks (source: CT data 17 March until 25 May 2009, N=1142)

A Chi-square test was performed to test the assumption that CT requests arrive at the Imaging Department at a constant rate across the week. The results are shown in table 8.

Table 8: Results Chi-square test : CT requests per day (source: CT data 17 March until 25 May 2009, N=1142)

The observed frequency for each row is the total number of CT requests per day over the 10 weeks of data. The expected value for each row is equal to the sum of the observed frequencies (1142) divided by the number of days (7). The residual is equal to the observed frequency minus the expected value. The obtained chi-square statistic equals 53.602. This is computed by squaring the residual for each day, dividing by its expected value, and summing across all days. ‘Asymp. Sig.’ is the estimated probability of obtaining a chi-square value greater than or equal to 56.602 if CT requests arrive evenly across the week. The significance value of ,000 suggests that the number of CT requests really does differ by day of the week.

A follow up analysis needs to show if on weekdays only (Monday through Friday) CT requests arrive at a constant daily rate. The results of the corresponding chi-square test are shown in table 9. It shows that there is no statistically significant difference between the number of CT requests on weekdays only.

18

Table 9: Results Chi-square test : CT requests per day (source: CT data 17 March until 25 May 2009, N=1142)

An chi–square test on the CT requests on Saturday and Sunday (table 10), shows that there is also no statistically significant difference between the number of CT requests on the days of the weekend (at level α=0,05).

Table 10: Results Chi-square test : CT requests per day (source: CT data 17 March until 25 May 2009, N=1142)

We conclude that a distinction needs to made between the number of CT requests on weekdays (Monday through Friday) and the weekend. Table 7 shows that an average of 18,38 CT requests per day will arrive at the Imaging Department on Monday to Friday. On Saturday and Sunday this is 11,15 CT requests per day.

2.3.5 Waiting time

Within this study waiting time is defined as the time between the request time and the time that the patient arrives in the CT room. This is a definition in line with the general definition of waiting time by the NSW department of Health (NSW Health, n.d.).

Within the dataset a total of 880 of the 1395 cases were available for analysis, because in 494 cases one or more time/date stamps were missing and 21 cases had negative waiting times. Negative waiting times can exist because paperwork is done later in case of high priority exams. Analysis of the gathered data showed that the average waiting time for a CT scan is 15 hour and 13 minutes, disregarding exam- and patient type. The median however, which is less affected by outliers, is 2 hours and 47 minutes. Table 11 presents the descriptive statistics of the waiting time for CT exams.

N Minimum Maximum Mean Median Std. Deviation Skewness Kurtosis

Waiting time 880 0:06 314:05 15:13 2:47 30:27 4,31 27,01

Descriptive Statistics

Table 11: Descriptive statistics of overall CT waiting time data (displayed in hours) (source: CT data 17 March until 15 June 2009, N=880)

When a distinction is made between the different patient types, this results in the descriptive statistics as presented in table 14. It shows that the mean waiting time for emergency patients is 3 hour and 10 minutes, while the median is 1 hour and 20 minutes. The average waiting times for in- and outpatients are considerably larger. An important thing to be aware of is that the statistics for the outpatients are only based on only 6 observations because most of the missing time-/date stamps were from this category of patients.

19

Waiting times N Minimum Maximum Mean Median Std. Deviation Skewness Kurtosis

EM 503 0:06 124:05 3:09 1:20 9:21 9,68 112,21

IN 368 0:14 213:49 29:20 21:06 33:58 2,24 5,99

OUT 6 17:50 314:05 157:28 119:09 125:35 0,53 -1,83

Total 877

Descriptive Statistics

Table 12: Descriptive statistics of CT waiting time data per patient type (displayed in hours) (source: CT data 17 March until 15 June 2009, N=877)

The waiting time per exam type is presented in table 13. It shows that the XFNA exam (fine needle aspiration) has by far the longest waiting time (mean: 159:03hrs, median: 141:54hrs) because a certain medical specialist needs to be present for this type of procedure. Currently, XFNA exams can only be done on Wednesday morning. From the dataset it was derived that no ED patients received this scan (see appendix 4). So the priority of this exam can be considered to be low.

The XCAC,XNPC and XAPC exams from table 13 also have notable waiting times. A part of this can be explained by the fact that these exams are confronted with ‘compulsory’ waiting time. XCAC, XNPC and XAPC are oral contrast studies, and the contrast has to be administered to the patient an hour before the exam starts.

One of the shortest waiting times is for XB exams (mean: 8:15hrs, median: 1:23hrs). The XB exam is the most frequent exam being performed inside the CT room (also on ED patients). The XB exam also has the highest priority and is a non contrast study.

Waiting times N Minimum Maximum Mean Median Std. Deviation Skewness Kurtosis

XFNA 7 23:16 314:05 159:03 141:54 115:36 0,43 -1,24

XCAC 17 11:05 213:49 52:21 24:26 53:37 2,14 4,63

XNPC 25 0:50 92:24 31:56 25:32 26:08 0,92 0,22

XCA 13 2:58 165:34 36:10 21:30 41:04 2,98 9,74

XAPC 176 0:19 162:23 17:29 5:49 24:58 2,66 8,59

XAP 53 0:11 66:10 6:27 2:14 11:17 3,46 14,89

XBC 21 0:22 90:22 18:48 5:00 24:54 1,69 2,56

XEX 17 0:24 75:08 13:51 5:01 22:24 2,29 4,40

XSPA 65 0:23 75:28 8:59 3:31 13:21 2,81 9,79

XB 376 0:06 188:30 8:15 1:23 22:12 4,64 24,94

OtherEM 44 0:16 124:05 5:20 1:03 18:49 6,13 39,14

OtherIN 58 0:25 147:04 32:55 23:18 33:38 1,74 3,03

OtherOUT 1 94:19

Total 873

Descriptive Statistics

Table 13: Descriptive statistics of CT waiting time data per exam type (displayed in hours) (source: CT data 17 March until 15 June 2009, N=873)

2.3.6 Summary

In this section we have analysed the overall performance of the CT processes. The data indicates that an average of 18,38 CT requests per day will arrive at the Imaging Department on Monday to Friday. On Saturday and Sunday this is 11,15 CT requests per day. Of these requested CT exams, 58 percent is performed on ED patients, while 36 percent is performed on inpatients and 6 percent on outpatients. 76,5% of the exams is done during normal working hours, which are on Monday until Friday between 08:30 and 21:00 hrs. The remaining 23,5% is done during the afterhours, including the weekend. The mean waiting time for emergency patients is 3 hour and 10 minutes while the median, which is less affected by outliers, is 1 hour and 20 minutes. The average waiting times for in- and outpatients are considerably larger. Finally we have seen that one exam (XFNA) has significant higher waiting times than the other exams. And the most frequent exam (XB) has one of the shortest waiting times.

The next chapter focuses on the creation of a simulation tool of CT processes that improves ID manager's ability to anticipate on the impact of changes on the overall performance of this subsystem.

20

Chapter 3: Simulation model

In simulation we use a computer to evaluate a model numerically. Building a simulation model of the CT processes at the ID of Campbelltown hospital requires a systematic approach. The approach taken within this study is based on the steps described by Law (2007), and is presented in figure 1.

Within this chapter we will focus on the conceptual model (3.1) and the construction of the computer program (3.2). The programmed model will be validated in section 3.3.

3.1 Conceptual model

When defining the conceptual model, different steps can be distinguished according to Law (2007):

A. Collect information on the system layout and operating procedures;

B. Collect data on the performance of the existing system for validation purposes;

C. Collect data to specify model parameters and input probability distributions;

D. Delineate the above information and data in a written assumptions document

The results of the first two steps (A and B) were already disclosed in chapter 2 and they will be used as input for the conceptual model. This section will start with a description of the scope of the model (3.1.1). Then the model parameters (3.1.2) and input probability distributions (3.1.3) will be specified. Next the remaining assumptions will be presented (3.1.4) and the performance measures used for validation (3.1.5).

3.1.1 Scope of the model

The model includes three types of patients, for which an CT is requested by means of a request form:

emergency patients, inpatients and outpatients. The simulation starts at the moment of arrival of the CT request and ends when the patients leave the ID at the exit or are transferred back to the ward/ED by a wards person. Reporting procedures on the results of the CT exam will not be included in the simulation model.

Level of detail

The following level of detail will be included in the model:

Space:

o Reception;

when outpatients arrive at the ID they will have to subscribe at the reception desk first.

Outpatients can be examined only on working days between 09:00 and 17:00 hrs. During these times the reception desk is always staffed.

o Waiting room (2x);

when outpatients have subscribed themselves at the reception they will have to wait in the outpatient waiting-room until a nurse will collect them. In- and emergency-patients will arrive a their own waiting room by means of transport provided by a wards person.

o Clinical preparation area inside ID;

outpatients requiring preparation for a contrast study (intravenous access line, drink, etc.) will be collected by a nurse from the waiting room and brought to the clinical preparation area.

Figure 2: Modelling steps

21

Patients are transported from the waiting room/preparation area to the CT room by a nurse.

One patient can enter the CT room at a time. There is 1 CT room available 24 hours a day, 7 days a week. Maintenance of the CT scanner, which is performed twice a year and will consist of a half day work each time, will not be part of the model.

Workforce:

o CT radiographers; From Monday to Friday CT radiographers are available as follows:

08:30- 13:00 hrs. - 1 CT radiographer 13:45 -16:30 hrs. - 2 CT radiographers 16:30 -21:00 hrs. - 1 CT radiographer

During afterhours and the weekend there is no special dedication of a radiographer to the CT room, but there is always one present inside the ID to perform CT exams.

o Receptionist; From Monday to Friday the reception desk is always staffed between 09:00 and 17:00 hrs by two receptionists.

o Nurses; 24 hours a day, 7 days a week nurses are available to support all diagnostic services delivered at the ID. Therefore It is assumed that there is always a nurse available to support CT processes.

o Wardsmen; Normal working hours for wardsmen are from 9:00 to 17:00 hrs. In addition to those hours, 1 WP is in the ID from 8:00 hrs. and 1 (other) WP will be there until 20:00 hrs. Transportation during the afterhours is done by other staff members of the hospital. In the model it is assumed that transport is always available when needed.

Patients:

The model includes three types of patients: emergency patients, inpatients and outpatients.

Queues:

Emergency patients have priority over inpatients, and inpatients have priority over outpatients. In all cases non-contrast brain exams (XB) are done as soon as possible, disregarding patient type. In the real-life situation sometimes patients get preferential treatment based on their medical status. In the model this will be considered an exception and will not occur in the model.

Paths:

The model will show the paths that the patients will follow when they are moving

Excluding infrequent events

There are a number of events that only take place at infrequent intervals. When such an event occurs some alternative operating procedure is needed. As these events are not part of normal operating practice, they will not be included in the simulation model. Some examples of these events are:

- Sickness of personnel.

- CT breakdown, which can have a impact on the available CT capacity.

- Sometimes in-/or emergency patients are (re)prepared for contrast studies inside the clinical preparation area inside the ID, instead of at their ward.

- It can happen that the radiographer(s) and wards person on duty will consult together how rescheduling takes place during the day because of the reasons mentioned in section 2.2. During this rescheduling the normal queuing discipline can be overruled.

22 3.1.2 Specification of model parameters

Profiles

CT exams are performed on three categories of patients, which are: emergency patients (58%), inpatients (36%) and outpatients (6%). The corresponding frequencies of CT exams per patient type are presented in table 11.

Exam type Percentage

XB 56,95%

XAPC 13,27%

XSPA 7,63%

XAP 7,63%

XBC 2,00%

XEX 1,50%

XNPC 0,63%

XCAC 0,50%

XCA 0,13%

Other 9,76%

Emergency patient

Exam type Percentage

XAPC 27,33%

XB 24,09%

XSPA 9,11%

XAP 4,66%

XCAC 4,66%

XNPC 4,05%

XCA 3,44%

XBC 2,83%

XEX 2,02%

XFNA 1,21%

Other 16,60%

Inpatient

Exam type Percentage

XNPC 24,10%

XFNA 18,07%

XAPC 9,64%

XCAC 9,64%

XB 6,02%

XSPA 6,02%

XCA 3,61%

XEX 2,41%

Other 20,48%

Outpatient

Table 14: Frequencies of CT exams per patient type (source: CT data 17 March until 15 June 2009, N=1376) Preparation for contrast enhancement

The exam codes ending with a ‘c’ are the CT exams that need contrast enhancement (orally or intravenously).

As described in section 2.1 inpatients and emergency are prepared for contrast enhancement at their wards, and outpatients are prepared at the ID in the designated area.

Studies that need oral contrast are:

XAPC XNPC XCAC XAC XCBC

Patients must have the oral contrast administered one hour before the procedure, which is the optimum time for the scanning procedure. IV contrast enhancement studies do not have time restrictions. These patients only need to have an intravenous access line before transportation to the ID takes place. It is assumed in the model that patients are prepared immediately after their CT exam is requested.

3.1.3 Input probability distributions

To carry out a simulation using random inputs it is necessary to determine the probability distributions for these inputs and therefore it is important to collect data on these input variables of interest (Law, 2007).

Within this study probability distributions were determined for:

- The interarrival times of CT requests during a day - The scanning time

- The finishing time

The interarrival times of CT requests during a day is an important variable of interest because CT requests for patients arrive at the ID at random points in time.

Scanning time is defined as the time needed to perform the CT exam. The scanning time includes the time that is needed to place and position the patient on the CT table. The scanning time ends when the patient leaves the CT table.

Finishing time includes the time that is needed to transport the patient outside the CT room, and to clean and prepare the room for the next patient (change linen sheet etc.).