Emergency department: Developing a framework to explain long waiting times.

Emergency department: Developing a

framework to explain long waiting times.

Master Thesis

R.H. Waterlander

S2242222

University of Groningen

Faculty of Economics and Business

Master thesis – Business Administration:

Technology and Operations management

28

_{of February 2014}

2 Contents Abstract ... 3 1. Introduction ... 3 2. Theoretical background ... 5 2.1 Diagnosis process ... 5

2.2 Emergency care waiting time: common causes ... 6

2.3 Previous research findings ... 9

2.4 Diagnosis of waiting time in emergency care ... 10

3. Diagnosis framework ... 11

3.1 Considerations in developing the framework ... 11

3.1.1 Considerations on waiting time causes ... 12

3.1.2 Indicators of causes for waiting time within the ED ... 13

3.1.3 Benefits of pre structuring the framework ... 14

3.2 Framework model design ... 14

3.2.1 Part 1: Orientation ... 14

3.2.2 Part 2: Specification ... 15

3.3.3 Part 3: Modeling and analysis ... 17

3.3.4 Part 4: Results and reflection ... 20

4. Example of Framework application ... 20

4.1 Short case ED description ... 20

4.2 Application of the framework ... 22

4.2.1 Availability of ED staff (nurse, physicians, administrative staff ). ... 22

4.2.2 Shortage of physical plant space within the ED ... 23

4.2.3 Increased medical record documentation requirements ... 24

4.2.4 Waiting for inpatient bed placement ... 25

4.2.5 Language and cultural differences ... 26

5. Discussion and conclusion ... 27

5.1 Framework design ... 27

5.2 Framework application outcomes and recommendations ... 28

5.3 Recommendations for future research ... 28

References ... 30

Appendices: ... 34

Appendix A: Process scheme ... 34

Appendix B: Data analysis... 35

Appendix C: Interviews and coding ... 36

Interview 1: ... 37

Interview 2: ... 41

Interview 3: ... 46

Appendix D: Semi structured, interview Protocol (in Dutch) ... 48

3

Abstract

In this paper, a diagnosis framework is developed for use within an emergency department (ED) context. This framework can be used to stepwise diagnose the causes for waiting times that occur within the ED to create a clear overview for hospital management, so that management knows on which causes should be focused to reduce overall waiting times within the ED. In chapter 3 the framework is explained, in chapter 4 part of the framework will be demonstrated and presented and in chapter 5 both the framework and outcomes are discussed.

1. Introduction

The healthcare context has changed significantly over the past decades. Because of social, economic and political change, hospitals are facing budget cuts and closure of departments. These measures put pressure on capacity of hospitals and especially their EDs, whilst emergency care is one of the most visible parts of a healthcare system, where process failures are widely publicized and subject to public scrutiny and debate (Van der Vaart, Vastag and Wijngaard, 2011). Neergaard (2006) mentions that, demand for emergency care in the United States of America is increasing, but the capacity for hospitals, ambulance services and other emergency workers to provide it, is dropping. This is in accordance to other western healthcare systems in general, which are under increasing pressure to accommodate the growing demand for healthcare resulting from population growth, ageing population, new technology and heightened expectations from a better informed community. (Daly, Campbell & Cameron, 2003). This causes long waiting times for hospital patients, (Solberg, Asplin, Weinick and Magid, 2003). Therefore hospitals in general are focusing on minimizing throughput time (i.e. the time it takes patients, from entering the hospital to exiting the hospital) of patients and improved efficiency (Griffin, Xia, Peng and Keskinocak, 2012).

4 waiting times in EDs (Verbree, 2013). Because of the importance of timeliness in patient treatments in emergency rooms, which are often often life or death situations, it is very beneficial for hospitals to be able to reduce their waiting times without loss of quality of treatments. A suitable framework to quickly and effectively diagnose the causes for waiting times could therefore be a helpful tool for hospitals to get an overview of their performance and improvement potential. Therefore the focus of this master thesis is to develop a framework for diagnosis of causes of waiting times within EDs.

Frameworks are often developed to generally explain or structurally analyze a certain phenomenon. Wagner (1993) who reviewed literature in order to find a suitable domain independent diagnosis concept and strategy to structure the diagnosis process, elaborates on this, by explaining the importance of detecting causes of negative influences on performance (i.e. throughput times) which require both general problem solving and domain specific knowledge. In spite of this importance, a framework to diagnose the causes for waiting times in EDs has not been developed yet. Developing a diagnosing framework to find causes for waiting times within EDs could significantly assist hospitals in decision making on which aspects within the ED should be focused to reduce overall waiting times and therefore reduce throughput time, of patients.

In previous studies of waiting times in EDs, certain performance data was required to find the causes of the waiting times. In literature, also the commonly occurring causes for waiting times in EDs are described. But both these kinds of studies lack the generic approach which should make it easy for hospitals to quickly asses the performance of their EDs. To be able to do a fully guided diagnosis of causes of waiting times, there is need for a design of a framework for diagnosing the causes for waiting times within EDs. The diagnosis framework should be able to distil the problems occurring in the ED which cause extended waiting times, by making use of the quantitative patient-process data, which could be obtained from the hospitals databases.

5 As was mentioned the main goal of this thesis is to develop a diagnosis framework for diagnosing causes of waiting times within EDs. To be able to get started in building this framework some important factors need to be investigated and summarized. The most common causes of long waiting time in EDs and the indicators of each of the causes which are discussed in literature are important input factors for the diagnosis framework. The diagnosis framework will be built up in the form of a stepwise approach to guide the diagnosis of causes for waiting time in the ED.

In Chapter two, the common causes for waiting time will be discussed in the theoretical background. In this chapter the theory behind developing a diagnosis framework is also explained. Here the general problem solving literature is used to determine appropriate process steps and domain-specific literature is used to find the common causes for waiting times in EDs, linked to the performance indicators from which the waiting times can be distilled. In chapter 3 the theoretical framework will be presented. In chapter 4 the framework will be tested and in chapter 5 the framework and research results will be discussed and evaluated.

2. Theoretical background

In this theoretical background we will explain and discuss the diagnosis process. Afterwards the variables that are of influence on the waiting times for patients before and within the ED will be discussed and finally the diagnosis methods which are commonly used for similar processes and which therefore can be of interest in development of a diagnosis framework for finding waiting time causes are explained and discussed.

2.1 Diagnosis process

The diagnosis of problems is very important in companies and this is therefore often mentioned in literature. However, the diagnosis of problems is often highly task or domain specific. To look at a broader perspective, the article of Wagner (1993) is considered. In this article he discusses the separation of problem solving and diagnosis, from their specific domain, to discuss the development of independent problem solving concepts and strategies.

6 domain independent problem solving strategies. Domain knowledge means understanding of objects, their structural and functional properties and their interaction within a certain system (Wagner, 1993). Wagner (1993), distinguishes different types of diagnoses, namely causal diagnosis and situation

understanding. Situation understanding is about getting an understanding of how a system works; with

this diagnosis type an undesirable situation is not present. The relevant diagnosis in this paper however, is causal diagnosis, this diagnosis seeks to determine the cause of an undesirable situation (i.e. waiting times in emergency care). In causal diagnosis it is relevant whether the source of the problem can be treated as a structural or functional problem. A structural problem can be analyzed as, finding the structure which causes the problem, this usually does not require understanding of the underlying malfunctions of the system. For a functional analysis, more detailed study of the situation is necessary; the problem solver needs to

understand the system behavior in order to find the function which causes the problem. Therefore structural analysis requires less expertise than functional analysis.

A causal diagnosis has a certain degree of difficulty, which is determined by several factors:

Size and depth of the system, an unknown desired state, the existence of possible (multiple) causes, a large set of potential causes, symptom similarity, never before observed causes, signal noise and lack of observability (Wagner, 1993)

In executing a diagnosis these factors have to be taken in to account. Because these factors are domain specific, they need to be applied to the context of this paper, which is emergency department waiting times. Although not all emergency departments are equal, there are some factors which are found to influence waiting times, which are encountered in most emergency rooms. These will be discussed in the next paragraph.

2.2 Emergency care waiting time: common causes

In order to be able to develop a diagnosis framework for waiting times in emergency care, it is important to know what are the often occurring causes for waiting times within the EDs. From the article of Paul, Reddy and DeFlitch (2010) in which they reviewed simulation studies investigating emergency department overcrowding, a lot of possible common causes for waiting times can be distilled. Since waiting time tends to go up on average when queue lengths increase (Hopp &

7 Spearman, 2011) in this case due to overcrowding. This review compares results of different kinds of addressing the problems of ED overcrowding. Several studies have attempted to find explanations for overcrowding and waiting times within EDs.

The most common reported causes for Emergency department overcrowding and with this increased waiting time, are found within literature and literature reviews. For this research, only causes within the ED itself will be taken into account and summarized below. For example ambulance routes/diversions etc. which are commonly mentioned in literature are not taken into account.(Derlet, Richards, 2000; Derlet, Richards and Kravitz, 2001; Trzeciak and Rivers, 2003;Hoot and Aronsky, 2008)

1. Patient urgency (Derlet and Richards,2000;Trzeciak and Rivers, 2003; Yoon, 2003; Potisek,

2007; Arkun et al, 2010; Van der Vaart et al., 2011)

With an aging population chronic diseases are more common, also medical technology and medicine allow patients to survive longer. The diseases they have can be obstructive for diagnosis procedures because they make it more difficult to make diagnoses, they often have multiple different medical problems and thus are more time consuming in diagnosis.

2. Hospital bed/room shortage (Derlet and Richards, 2000; Derlet, Richards and Kravitz, 2001,

Hoot and Aronsky,2008)

In different ED situations, different shortages can be a cause for overcrowding. EDs either have hospital beds for inpatients, or need capacity outside of the ED (beds) to take patients in. In the latter case, patients have to wait in an ED room to be admitted a hospital bed in a different department.

3. Number of patients, (Derlet and Richards, 2000; Ekelund et al, 2011; Waldrop, 2009; Van der

Heide, 2006; Hoot and Aronsky, 2008)

Increased number of patients, arriving within the EDs have a negative effect on throughput time and increase waiting. Large numbers of patients are clogging the ED forming queues which increase waiting times and therefore overall treatment time of patients which require emergency care. 4. Avoiding inpatient hospital admissions by “intensive therapy” and elongated treatment

times in the ED. (Derlet and Richards, 2000)

8

5. Laboratory or Radiology delays (Kyriacou, 1999; Yoon, 2003; Miro et al., 2003; Storrow et al.,

2008; Yen and Gorelick, 2007; Holland, 2005; Scalise, 2006; Derlet and Richards, 2000; Van der Vaart, 2011; Nanninga, 2005)

Advanced technology and increased standards of care, result in more ED patients which need radiology (X-ray, CT scans, nuclear scans etc.) or laboratory (blood) tests. In some hospitals (primarily teaching hospitals) these services have been notoriously slow.

6. Specialty consultant delays (Yen and Gorelick, 2003; Derlet and Richards, 2000)

Consultations by specialists are needed to provide definitive treatment in the ED or for admittance to different departments within the hospital. On-call specialists often have very busy schedules in their own departments within the hospital and this causes delays in responding to the ED.

7. Availibility of ED staff. (Derlet and Richards, 2000; Fitzpatrick et al, 2006; Laskowski et al.,

2009; Sinreich, ; Ekelund, 2011; Kyriacou et al., 1999; Yoon et al, 2003; Waldrop, 2009; Van der Vaart et al, 2011; Van der Heide, 2006; Miro et al, 2003; Verbree, 2013)

Many ED functions next to treatments of patients, rely on tasks like telecommunication, requisition of paperwork, and processing done by clerical support staff. Inadequate numbers of staff like administrative staff, nurses or physicians could cause lengthened treatment times. With budget cuts and cost reductions, hospitals have reduced the number of staff thus leading to delays in start of the treatment, patient admissions, requisitions for laboratory/radiology admissions etc.

8. Waiting for inpatient bed placement (Nanninga, 2005; Waldrop, 2009; Verbree, 2013) After patient treatment in the ED, some patients require a transfer to another department for monitoring the patient’s condition or continue treatments. In between the moments that the treatment in the ED is finished and the patient is transferred, there is a moment in which the patient occupies a treatment room in the ED whilst the patient is waiting to be transferred to a different department.

9. Shortage of physical plant space within the ED (Derlet and Richards, 2000; Drummond, 2002) As the length of stay increases for patients, a larger ED becomes necessary to facilitate the same or a higher number of patients.

9

EDs within areas with high proportion of immigrants may have immigrant patients coming which do not know the proper healthcare routes, have cultural issues (i.e. do not want female/male doctors) or do not understand the language. This might prolong diagnosis and treatment time. 11. Increased medical record documentation requirements (Derlet and Richards, 2000; Wu,

Chaudhry, Wang, Maglione, Moica, Roth, Morton and Shekelle, 2006)

Many years ago, ED charts had brief notes to maximize physicians time to provide hands-on care to patients with emergency medical conditions. Notes gradually increased in length and complexity to meet demands of insurance companies who pay the bill.. Emergency physicians spend more time with the “chart” than with the patient

2.3 Previous research findings

Although these mentioned causes are indicated as being the most common causes for overcrowding, depending on the layout or structure of the specific emergency department, other causes might be present or only one or a few of the above presented causes are present in the emergency department. As became visible from a previous study on waiting times in the specific case hospital, (Dijk, 2013) the important factors to reduce waiting time of patients are to be responsive, have clear overview of patients in the system at any given situation and have flexible capacity for peak hours, to prevent long queues from forming. Other findings were lack of physical capacity in rooms as well as personnel and number of patients and patient priority (Verbree, 2013) which were found to be of negative influence on waiting times of patients. These factors can be linked back partially to the most common causes of overcrowding as shown in the list above. Some however cannot be linked back to the most common causes but are difficult to measure or present as a cause.

10

2.4 Diagnosis of waiting time in emergency care

Diagnosis of the causes of emergency care waiting times is often done by analyzing large quantities of data and extracting and drawing conclusions from these data. The data which was used to draw these conclusions however can possibly be generalized. By this it is meant that indicators from the data, which can be linked to general causes, can be used to provide similar information in EDs in which have similar data available. Long return times of lab results visible in time data measurements show that laboratory response time is a possible cause for overall long waiting times. By focusing in this manner on indicators of causes, relevant data which is actually required to make an analysis could be gathered in a structured way. So that data sets take less time to be formed or can be made clearer and more understandable. If it is clear which indicators (effects) belong to which causes of increasing waiting times, the data can be analyzed to test if the indicators are present and a cause can be identified. By testing each specific indicator, it can be concluded it is (part of) the cause of the waiting time occurring. A second way to diagnose causes of emergency care waiting is to do expert interviews, interview different members of the ED staff and find out if there are common causes mentioned by staff and if these correlate with expectations. If experts or staff members mention new causes of waiting times, which have not yet been identified from the data, it could indicate that there is a gap within the data collection. There is no data available which shows this cause being present in the system. With the data available from the expert interviews, specific indicators can be tested for presence by gathering the relevant data.

A third way to diagnose causes for waiting times is to do simulation studies. From the article of Paul, Reddy and DeFlitch (2010) it becomes clear that a lot of simulation studies have already been done on the emergency department theme. Within simulation studies of course input data is needed, this might be gathered from the former two named sources for diagnosis, time/patient data and staff/expert interviews. With these data inputs estimates can be made of performance and patient mix. The simulation model can be run numerous times and the bottlenecks of the system will become visible. Although a simulation model would be very helpful this is very time consuming and is not considered in this master thesis.

11

3. Diagnosis framework

The focus of this research lies on creating a framework for analyzing the causes of waiting times within the ED of hospitals in the Netherlands or other countries with similar healthcare systems. Because of the absence of a structured approach to assess waiting times within EDs in current literature, the aim is to provide a framework to quickly assess the causes for waiting times occurring within EDs. For diagnosing the causes of long waiting times for patients a framework will be developed according to the development method of Wagner (1993). The framework will be constructed with knowledge from literature and previous research. The framework will consist of a stepwise approach to find and address causes of waiting time within the EDs. To validate the framework, the causes for waiting times from a single case study in the Medical Centre of Leeuwarden, which will be distilled from observations, measurements and interviews, will be compared with the outcome of the framework. In paragraph 3.1 the considerations which are important in this specific framework are discussed. In paragraph 3.2 the basis of the framework will be described. Following the framework will be more elaborately explained in 3.3.

3.1 Considerations in developing the framework

When developing a diagnosis framework for causes of waiting times in an emergency department of a hospital, many different aspects are to be considered. For example in comparison to a framework diagnosing the waiting times in a production facility environment there are a lot of differences. A few major considerations will be discussed and explained below

Demand

As also mentioned in previous research (Verbree, 2013), the demand of an emergency department is not comparable to a demand in a production environment. In a production facility, the orders which go into production can be managed and a choice can be made when to launch a new order into a system. In an emergency department, demand occurs equally random as demand in a production environment, but the main difference is that the patients (orders) need to be carried out immediately. The patients cannot be planned into a production schedule.

Capacity (planning)

12 the market which is the constraint (Hopp and Spearman, 2010). However the performance of the emergency department is very much dependent on peak loads, the demand is not equally spread out over the day and over the week (Verbree, 2013) and again the demand cannot be planned in to a production schedule, the patients need treatment straight away. The key to a balanced ED with acceptable waiting time, is to get a proper match of capacity and demand; the capacity should not exceed the demand too much to reduce inefficiencies (in most importantly budget), but it should also not be equal or below the demand because of possible increasing waiting times occurring and in the worst case scenario this can lead to patient mortalities.

Cycle time

A third difficult aspect of the emergency department is the divergence in treatment time (in a production environment equal to cycle time) of a patient. Because of the wide range of medical conditions, complexity and complications of treatments and differences in experience and expertise of staff, there is a lot of variation in treatment times. Especially for example in teaching hospitals in which a lot of unexperienced doctors are working next to the experienced ones. This also makes it difficult to point out causes for long treatments or waiting times, and variation in treatment times, for patients because the causes for this can be very diverse.

3.1.1 Considerations on waiting time causes

Contradictions

13

Categorization of causes

Treatment process

The first categorization which will be used in the framework is that of the process components, the process components. The process will be divided in three components:

Pre-treatment, which encompasses registration, waiting room and triage.

Treatment, which encompasses all activities in the treatment room(s) by nurses physicians and

in-between waiting times.

Post-treatment, which encompasses all activities after treatment, waiting for inpatient bed placement,

i.e. waiting for a pickup or other delays before leaving the ED.

By making this distinct division within the treatment process, will allow for a more specified search for waiting time causes. Usually the part of the treatment process in which the waiting times occur is known, by using this global approach of splitting up the patients process through the ED, the focus can remain on the process steps where most time is spent waiting by the patient. This categorization will help in speeding up the search process and will result in a better directed walk through the framework.

Capacity

From analyzing the most common causes of waiting time found in literature, almost all causes can be described as a sort of capacity, for example: shortage of personnel, or shortage of waiting rooms or beds, delay in specialist consults. As time is a buffer for capacity, and a hospital not having an inventory, it is logical that capacity and time are the only two parameters that can influence the throughput times. In one way or another all these causes can be led back to capacity shortages. Waiting patients, are always waiting for: A treatment room, a nurse, a physician, a specialist or radiology or lab results. Shortage in personnel capacity, shortage in space (rooms) capacity etc. . Therefore, capacity is considered a very important theme and an essential parameter to the ED performance. Capacity factors taken into account are: Personnel capacity (doctors, nurses, specialists etc.), space capacity (waiting room, treatment rooms, etc.), third party capacity (laboratory, radiology, etc.). With a categorization of these factors, a more detailed search is possible to find the causes of waiting times within the ED.

3.1.2 Indicators of causes for waiting time within the ED

14 earlier research or discussion with ED staff. To focus the search for causes within the framework, a distinction is made between sources for the indicators. Indicators which can for example be time measurements from a data set, need to be harvested from a certain source, in this case the database for data registration. The sources available in the ED are: Data registration, within the ED, where patient personal and process data is registered to a certain degree. A second source interviews, experienced personnel of the ED can point out the presence of certain indicators. And thirdly

observation is a way to find the existence of certain indicators, when for example the interviews with

personnel could be biased, because the indicator has a link with personnel performance, observations can be used to rule out the bias of personal interest, in outcome of the research, of the ED personnel.

3.1.3 Benefits of pre structuring the framework

By doing a structured search for indicators of causes, the indicators will be categorized in three groups of sources, as explained above they will be categorized in: Data registration, interviews and observations. If within the ED a certain area is suspected of causing excessive waiting times, the diagnosis framework can be applied in a more efficient way, by first zooming in on the potential causes which are of importance in this area and then try to find the indicators which belong to these factors to find structural evidence for the cause to be present in the ED either by using data registration, interviews or observations to find this indicator.

3.2 Framework model design

The general steps of Wagner (1993) which should compose the diagnosis framework in any context are described in the original article. The steps described by Wagner (1993) will be followed as a general guideline to build up the diagnosis framework in a structured way. To develop the framework, the choice is made to take in account only the operational process variables. Human emotions, worker satisfaction and other soft performance variables will not be taken into account in this framework. This choice is not because of the lack of importance but it is too broad of a subject to take into consideration in this master thesis all at once.

3.2.1 Part 1: Orientation

Step 1: Finding the signal that waiting time is too high.

15

Step 2: Description of the system: How does the emergency department work, what does the system look like.

The second step of Wagner (1993) is to get a good clear overview of the system which has to be submitted to the diagnosis. Therefore the processes etcetera that take place within the emergency department should be mapped and put in a structured layout. After the system description is complete, it is easier to search for deviations from the desired state, and make sure that every part of the process is carefully assessed for performance.

3.2.2 Part 2: Specification

First all potential problem causes are looked up in literature, these can be found in the theoretical background. The potential problem causes can be expanded by doing a quick survey with specialist staff (management/administrative staff/physicians) in the department, they can be shown the list and asked if any potential problems occurring in the department are missing, so that the list can be expanded. Hereby noticing that the list presented below is not finite, neither complete, to extend the current diagnosis framework, future research from practice and literature can extend this list to make the framework more complete.

16

Process steps:

General

Pre-treatment

Treatment

Post-treatment

Causes:

-Shortage of physical plant space within the ED - Increased medical record documentation requirements -Language and cultural differences -Patient urgency

-Hospital bed /room shortage

-Avoiding inpatient hospital admissions by “intensive therapy” and elongated treatment times -Laboratory or Radiology delays -Specialty consultant delays

-Waiting for inpatient bed placement

Table 1: Causes and indicators divided by process step

Within these “process steps” the causes are further diversified in different categories to further specify the search. Until a certain level some causes can be dismissed purely on prior knowledge of the system and its behavior. The problems which cannot be dismissed should be investigated by trying to find their linking indicators from data sources which can be obtained within the ED namely: observations, patient data or interviews. The causes can be identified by finding indicators. The indicators can be found by using one or more of the data sources which can be obtained within the ED. The causes and their identifiers are divided to the proper data source in the table 2:

CAUSE INDICATOR: DATA SOURCE:

1. AVAILABILITY OF ED STAFF (NURSE, PHYSICIANS, ADMINISTRATIVE/SUPPORT STAFF ).

1. Nurses/physicians/administrative staff is very busy/overworked, no breaks, running multiple tasks at the same time, making mistakes etc.

Observations/interviews

2. SHORTAGE OF PHYSICAL PLANT SPACE WITHIN THE ED

2. Waiting room/hallways full with patients occurring often.

Observations/interviews

3. INCREASED MEDICAL RECORD DOCUMENTATION

REQUIREMENTS

3. Doctors/nurses spend a long time documenting patient status and treatment proceedings

Observations/interviews

4. LANGUAGE AND CULTURAL DIFFERENCES

4. Treatments of people with other nationalities than native, influence average treatment times that day.

Data analysis

5. HOSPITAL BED /ROOM SHORTAGE

5. High bed/room utilization (<90%) often occurring

17

6. PATIENT URGENCY 6. People with high urgency have longer processing times and this increases waiting times for other patients.

Data analysis

7. SPECIALTY CONSULTANT DELAYS

7. When specialist consultants are involved treatment tends to take longer on average that day

Data analysis

8. LABORATORY OR RADIOLOGY DELAYS

8. When laboratory/radiology is involved more often, daily average treatment times increase.

Data analysis

9. AVOIDING INPATIENT HOSPITAL ADMISSIONS BY “INTENSIVE THERAPY” AND ELONGATED TREATMENT TIMES

9. Physicians in ED treat more different illnesses and send less people on for inpatient placements

Observations/interviews

10. WAITING FOR INPATIENT BED PLACEMENT

10. Patients needing bed placement in another department have long time between ready for departure, and departure

Data analysis

Table 2: Elaboration causes, indicators and sources

3.3.3 Part 3: Modeling and analysis

For the first step first for specifying the direction of the search for causes, a very global approach is used to define in which part of the process the waiting times occur. This can be seen in figure 2.

18 When it is identified in which part of the system the waiting times occur, causes can be searched in this process step more specifically. As can be seen in the model, if it is not known in which process parts the delays start, a back to forth search is done to find the causes. This back to forth search is done, to find the sources of the waiting time causes as close to the end of the process as possible. Because essentially, the goal is to find the

bottleneck/system constraint, which you in any process would prefer at the beginning of the process to ensure a more smooth throughput with less accumulation of patients within the treatment process (Goldratt, Cox and Withford, 1992). To zoom in on the analysis of a process component, the treatment process step will be highlighted in the following paragraph, the other components can be found in appendix A.

The second step is made by showing in the model in figure 3, that two routes chosen in

analyzing the data. The branches of the route can either be data analysis or observation/interviews, these routes, can be used to find different causes. In the model the Indicators of the causes are used to point out the presence of a certain cause.

When this indicator can be found within the interviews/observations respectively in the data analysis, the cause for waiting time can be pointed out. In the next paragraph, the process of data analysis, observations and interviews will be more elaborately explained and shown.

Data analysis

For the data analysis part, data registration in the hospital is of course needed. Every hospital ED may have different kind of data registrations or different kinds of software. Globally however some main data features are always recorded. This data is most likely to encompass patients personal information, and some medical/ treatment/requests and time data which will be added along the treatment process. Some features that could be recorded are: Nationality, gender, type of forwarding, type of arrival, moment of arrival, triage code, description of the patient’s problem, time of triage, time of treatment, time of departure etc. If this data is available, comparisons can be made to exclude or find the presence of certain indicators which can lead to finding the causes of waiting times within the ED.

19 As an example, the process of testing the presence of one of the indicators within the dataset is explained:

As an example for data analysis, the indicator: 7. When specialist consultants are involved, treatment tends to take longer. Will be examined.

The data set is used to find similar patients within the data set. Patients with specialist consult are selected and a testing group will be selected according to similar medical conditions. So if a for example person with internal problems, enters an ED and is helped by the ED physician, he will be compared with a person with similar injuries (rea: same specialism and same urgency code), which is seen by a specialist consultant. There will be a paired comparison between at least 30 patient pairs to ensure statistical significance. The treatment times of both patients are compared and tested for significant differences according to statistical analysis.

Then the results can be compared and conclusions can be drawn from the outcomes.

Interviews/observations:

For the interviews/observations part, an interview or observation protocol is needed. Which things should be observed or which questions should be asked in the interview depend on the indicators which need to be examined. If an interview is held, ideally all possible indicators within the process step should be examined at once, therefore the interview protocol will encompass questions to identify all possible indicators. Same goes for the observations, if observations are done in one part of the treatment process, as many indicators as possible should be tested in the same observation round, this could save a lot of time because observation is a time consuming process of gathering data. For observation an example will now be given. For the interview protocol a question list is provided in appendix D.

Example: When trying to observe if nurses/physicians/administrative staff do not have enough personnel capacity, observations can be done to check the activities of the personnel, examples can be: Is there time for coffee breaks, do they have time to take lunch breaks, is there always enough time for communication on important patient statuses. These observations which can be done, can then also be verified by asking the personnel themselves about these activities.

20

3.3.4 Part 4: Results and reflection

When the analysis of the different indicators is complete, the results can be distilled. If indicators are found within either the data analysis or the interviews and observations, they can be linked to the likely causes which could cause waiting time or when detailed data is missing, it can point to causes of lengthy treatment times within the ED. When the cause or causes have been identified, the ED management can then try to find solutions for the cause or causes which have been found.

After using the framework, the performance needs to be reflected upon. Which components could be added to the framework, and which essential steps have been missed in this research. This reflection should not in particular, solely be done by the researcher applying the framework, but it can be discussed in cooperation with management and hospital experts to try to improve and extend the framework and make it more robust.

4. Example of Framework application

Because of the fact, that there is limited time to conduct this master thesis research, only a part of the framework will be tested in this thesis. This part is chosen because when the framework should be applied on an ED in which there is no idea in which area waiting time is caused, the analysis should start at this process part, as explained in the previous chapter.

4.1 Short case ED description

21

22

4.2 Application of the framework

To test the framework, the data was first gathered as is explained in the previous paragraph. Because of time limitations in this research only the general component, and post-treatment component of the framework will be analyzed. The potential causes will be walked through step by step and the data gathering/selection which was needed is explained per potential cause analysis.

4.2.1 Availability of ED staff (nurse, physicians, administrative staff ).

One of the major causes for waiting time in EDs is in literature said to be shortage of ED staff. In this case this could be nurses, physicians or administrative staff. As can be seen in the framework there can be two ways to identify if there is a shortage of a kind of staff members, observations or interviews. In this case, because of limited time there was no time to do long periods of observation on a large number of days. Therefore in this case only interviews are held with each of the staff groups which could have insufficient capacity to process patients in the ED. Each of the staff groups will is investigated according to the semi structured interview which is analyzed. The transcription and coding of the interviews can be found in appendix C.

Physicians

By interviewing an ED physician it became clear, that they do not seem to suffer from lack of capacity. The ED physicians do normally have a supervising, coordinating role in the ED and try to do the most important and first diagnoses of the patients, next the doctor assistants or specialist assistants which are available at the ED, see and aid the patients and do further patient treatments, in correspondence with the specialists (by phone), or call them in to do the treatment if they themselves cannot manage it. For these assistants there is a very fluctuating capacity, but this does not always mean that treatment is slower or faster when there are more or fewer assistants available. Because it is a teaching hospital, there are a lot of intern doctor assistants which take on average longer than experienced personnel. However overall physicians, or doctor assistants do not seem to have lack of capacity in the current situation.

Nurses

23 interviewed nurse also mentioned that it was discussed to have an extra shift from 11:00 - 19:00 this was not possible because of Union agreements on working times and break times. Next to the fact that there are extra nurses available in the most busy periods of the day, the unit leaders (coordinating staff) also have a nursing background and can step in when required or the department is starting to get (over) crowded, to help speed up the processing of patients and reduce the waiting time for new patients. So here is some extra flexible capacity available, but even with this extra capacity there is still no time for nursing staff to take a lunch break for example, they always have a quick meal in their department.

An emerging problem in the last (few) year(s) however was an increase of patients turning up after office hours, after 19:00, during the nights and in the weekends. As the interviewee said: “Maybe because of the current economic situation that people fear losing their jobs and do not dear to take days off.. they come to the ED outside of office hours.” At these times there are no extra shifts planned and the unit leaders are not there during these times. The interviewee said that sometimes they could get some extra personnel from the Intensive Care but that was not always the case and they do sometimes struggle in these times to help all the patients in time.

Administrative staff

When analyzing the situation of the administrative staff, in person of the secretary of the ED, it seems that there is sufficient capacity to cope with current demand. The secretary which was interviewed, mentioned that there is always a secretary at the desk 24/7. They do not experience special busy days, as opposed to especially the nurses, which do experience those busy days. The secretary however does mention not having time for coffee/breaks, it was not a problem, because between busy moments there was always time to get a drink etc. During busy times, 11:00 – 15:00, there was no extra secretary personnel available in the ED, but there was also no need for extra personnel according to the secretary.

4.2.2 Shortage of physical plant space within the ED

24

“It is quite often the fact that patients are waiting on chairs in the hallways to get in to a treatment room, or they have to wait inside the ambulance for a room to become available. This happens certainly once a week especially between 11.00 and 15.00. In the weekends also the waiting room sometimes is full, because we share the same waiting room with the general practitioners (GP) post which is located down the hall also inside the hospital”

This confirms the fact that treatment rooms are in shortage during some times of the week, as was also found in the previous case study of Verbree (2013), but also physical space in the waiting room is sometimes in lack of capacity to hold all the patients. This could be a problem for patients which are in a bad health status, but it does not automatically lead to elongated waiting times or longer treatment times. Of course if there are more patients at one time the waiting time increases. However staff state:

“Even if for example a treatment room is added, we just need different policy, patients that shouldn’t be here should go directly to the departments or should see a GP first. Also it would be better if patients could be sent on directly after treatment even if they have to be taken in to a specialist department for bed placement. “

Therefore just adding rooms or space is probably not the solution to any of the waiting time problems, this will be further discussed in chapter 4.3.2.

4.2.3 Increased medical record documentation requirements

When considering the interviews done with different staff members it becomes apparent that there is a very clear increase in medical documentation requirements. Also the workload of medical record documentation compared to other tasks is very high. The ED physician mentions at least 50% of the time spent on a patient is medical record documentation (Building a treatment protocol, making medicine lists, documenting progress, status and actions done to medicate the patients, etc.). Also the nurse mentions similar experience according to the increase in medical documentation over the past years, for them it is especially screenings, new government regulations etc. they also spend more time behind the computer screen than with the patient. Both parties mentioned, say the ICT facilities and hospital software are not assisting and supportive to the personnel.

Two mentioned comments from the interviews are:

25

dramatic, every day the system needs a complete shutdown and we have to lay down our work and wait for this to happen.”

The third staff member interviewed was the secretary. The secretary apparently has less problems with administrative tasks and increase in these tasks over the year. The work for the secretary staff seems manageable still from a workload perspective and also no negative mentioning were made about the computer systems and ICT facilities. Overall the increase of medical record documentation could be part of increased waiting time, because it takes a lot of capacity away from treatment personnel, nurses and physicians. If somehow this process could be sped up, this would most likely be beneficial to the patient flow.

4.2.4 Waiting for inpatient bed placement

From literature, one of the most common causes for waiting time in an ED, is waiting for inpatient bed placement. By analyzing the data, it became clear that of all the patients visiting the ED in the months October and November of 2013, about 50% of the patients need further treatment or bed placement and therefore need to be picked up from the ED and transferred to another department.

These patients waiting for a pickup are occupying valuable space in the ED, keeping rooms occupied which otherwise could be used for treatment of a new patient, blocking the patient flows. As was found in the previous research done at the same ED in the MCL (Verbree, 2013), there is a high room utilization. Therefore rooms being blocked by waiting patients for inpatient placements or further treatment, put extra pressure on room utilization and is a major problem and is clogging the patient flow. To verify this, data of all patients which destination is the “clinic” and have the “ready for departure” and “actual departure” times will be analyzed and time between ready for departure and actual departure will be calculated. Then the excess occupation time of these patients can be calculated and their average waiting times (post-treatment) are calculated. If these results are significantly large, an analysis will be done to see if a larger number of clinical patients in a day, which need to wait for hospital admission and pickups, has an elongating effect on overall average treatment time that day.

26 This seems a lot of lost time, if compared to the average treatment time of these patients. The average waiting time for inpatient bed placement is 18.7% of the total treatment time of these patients. Because the need for inpatient bed placement is about 50% of the total patient volume entering the ED, this seems as an important factor for delays for other patients which are waiting to be taken into a treatment room, as well as the patients themselves which are waiting inside the treatment room for inpatient bed placement.

Some interesting additions to this came up in the interviews with different staff members. The nurse which was interviewed stated that, during lunch breaks for example, patients often cannot be picked up by other departments staff because of lack of staff to do the patient placement during the lunch breaks of the other internal hospital departments. These breaks (12:00 – 14:00) happen to just be in the middle of the most busy time in the ED, which is from 11:00 – 15:00 she said. This might contribute to stagnation in the patient flow, which causes extra waiting time for patients arriving at the ED during or after this period. Secondly a problem was that patients always need to be picked up and cannot be brought to the specialty departments, therefore the ED is always dependent on the speed of pickups by these departments to be able to use its treatment rooms again. As an example the nurse mentioned: “In the UMCG (university medical Centre Groningen) there is a special logistics team which takes care of patient transfers between the ED and the other departments in the hospital to take care of patient admissions to the hospital from the ED.”

4.2.5 Language and cultural differences

When considering the data, it was not diverse enough to conclude anything about language or cultural differences. The problem only thing mentioned in the interviews is that maybe cultural differences or lack of knowledge of the medical care system in this country, make for a larger percentage of foreign origin people to come to the ED instead of going to a general practitioner first. However because of the lack of occurrences in the dataset it no conclusions or statements can be made on the increase of average waiting or treatment time of patients when there is also an increase in patients who do not speak native language or are from a different culture.

50%

75%

90%

Waiting

times

(post-treatment) for inpatient bed

placement

0:22:00

0:33:00

0:46:00

27

5. Discussion and conclusion

First the framework design will be discussed and assessed, in the second paragraph the application of the post-treatment framework step is analyzed to see if it comes up with sufficient evidence of causes for waiting time in the ED. Finally some recommendations with regard to the next step in optimizing and extending the framework and ED potential improvements which are interesting for future research will be given.

5.1 Framework design

The framework in its current form is built up as a stepwise approach in which, step by step, indicators for causes are trying to be found, to point out a certain cause for waiting time. The approach is built up by using Wagner’s (1993) approach in developing a diagnosis framework. Although in this thesis only part of the framework is tested, the aim is to give some overall thoughts about the framework. The framework part, which was applied in chapter 4, has shown that there are interesting outcomes in the area of potential causes to waiting time in the ED. Whilst testing the framework, it becomes apparent that data is an important factor. If the data registration is more detailed and process steps are documented in detail and time based, the search for causes which is the frameworks main purpose will be more efficient and gives better detailed outcomes. However if data registration is not that rich, or does not provide the time measurements needed, a switch can be made to apply more qualitative data sources like interviews and discussions with staff or experts in the ED, this makes the framework more resilient. As is also done for a large part in applying the framework, since the quantitative data was not available in the form which was required to do the proper analysis, qualitative data was used to gain as much insight as possible with interesting results.

28

5.2 Framework application outcomes and recommendations

Through analysis of the post-treatment and other indicators for causes occurring in general, some interesting findings have been made in the partial application of the framework. In general it seemed unpractical using the data supplied from the internal data registration software for a lot of quantitative data analyses, because it lacked a lot of important data entry points like lab requests and patient progress etcetera, but using the qualitative data gathered from the interviews, some important factors that influence waiting time could still be defined. In the future when applying this framework this should be considered, for the post-treatment process step however the quantitative data which was provided, in combination with the qualitative data from the interviews was sufficient for analysis. It seems that the most important factors for potential influence in waiting time are inpatient bed placement, which takes about 18,7% of the total treatment time of patients. A connection is highly likely to be present with the overcrowding of the ED in general and inpatient bed placement. Patients waiting for treatment in the hallways, on ambulance beds, or standing in the waiting room, because of overcrowding, occurs at least once a week. This occurs because of the lack of available treatment rooms. This has influence on the waiting time for patients which arrive at the ED in these periods. It seems that if at first the logistics for pickup of patients after treatment and inpatient bed placement is key, and should be considered if patient waiting time needs to be reduced. Another point mentioned in interviews which is an interesting point, is ICT and software support, in general whilst not even questions were asked directly about the software or ICT support, the interviewees mentioned the faultiness and lack of user-friendliness of the software systems. From the interviews it also became clear that although staff is generally very busy in the periods between 11:00 and 15:00 they do not particularly face lack of capacity in treating patients timely, they see the biggest bottleneck being the treatment rooms which are generally occupied by patients for a long time. Although it might be the case that during the evening and night shifts there is slight lack of capacity amongst nursing staff because of shifting patient behavior mentioned in interviews, the patients are more likely to come to the ED outside of office hours and during the evening and night. During these times there is no backup capacity from unit leaders (which are only available at office hours) which could jump in to create extra nursing capacity, this could be taken in consideration in making new staff shift schedules.

5.3 Recommendations for future research

29 As for recommendations to the ED which was subject to the testing of the framework, future research is recommended to find user friendly ICT support for data registration, and automation. Further investigation is also needed in the pickup strategy and policy of the patients which have finished their treatments in the ED but need inpatient bed placement. Ideally this policy needs to be established in cooperation with the other departments within the hospital. A few examples that could be investigated could be: A push instead off pull strategy, where patients are always brought to the department instead of picked up, or a transition department where patients can be placed by ED staff and picked up by specialist department staff, so that they do not occupy ED treatment rooms whilst waiting for a pickup.

30

References

Baesler, F. F., Jahnsen, H. E., & Bío-bío, U. (2003). The use of simlation and design of experiments for estimating maximum capacity in an emergency room, 1903–1906.

Bullard, M. J., Villa-Roel, C., Guo, X., Holroyd, B. R., Innes, G., Schull, M. J., Rowe, B. H. (2012). The role of a rapid assessment zone/pod on reducing overcrowding in emergency departments: a systematic review. Emergency medicine journal : EMJ, 29(5), 372–8.

Carrasquillo, O., Orav, E. J., Brennan, T. A., & Burstin, H. R. (1999). Impact of language barriers on patient satisfaction in an emergency department. Journal of General Internal Medicine, 14(2),

82-87.

Chan, T. C., Killeen, J. P., Kelly, D., & Guss, D. a. (2005). Impact of rapid entry and accelerated care at triage on reducing emergency department patient wait times, lengths of stay, and rate of left without being seen. Annals of emergency medicine, 46(6), 491–7.

Christ, M., Grossmann, F., Winter, D., Bingisser, R., and Platz, E., 2010. Modern Triage in the Emergency Department. Deutsches Ärzteblatt International, 107 (50), 892-898.

Cochran, J. K., & Roche, K. T. (2009). A multi-class queuing network analysis methodology for improving hospital emergency department performance. Computers & Operations Research, 36(5), 1497– 1512.

Cooke, M. W., & Jinks, S. (1999). Does the Manchester triage system detect the critically ill? Journal of

accident & emergency medicine, 16(3), 179–81. Retrieved from http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1343329&tool=pmcentrez&render type=abstract

Cronin, J. . (2003). The introduction of the Manchester triage scale to an emergency department in the Republic of Ireland. Accident and Emergency Nursing, 11(2), 121–125.

Daly, S., Campbell, D. A. & Cameron, P. A. (2003) "Systematic Review."

31 Derlet, Robert W., Richards, John R., Kravitz, R. L. (2001). Frequent Overcrowding in U.S. Emergency

Departments, 8(2), 151–155.

Dijk, T (2013). “How can the throughput time of patients in the Emergency Department be improved?”. Master Thesis. University of Groningen, NL

Drummond, A. J. (2002). No room at the inn: overcrowding in Ontario's emergency departments.

CJEM, 4(2), 91-7

Ekelund, U., Kurland, L., Eklund, F., Torkki, P., Letterstal, A., Lindmarker, P., and Castrén, M. (2011). Patient throughput times and inflow patterns in Swedish emergency departments. A basis for answer, A National Swedish Emergency Registry. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine, 19 (37)

Fatovich, D., & Hirsch, R. (2003). Entry overload, emergency department overcrowding, and ambulance bypass. Emergency Medicine Journal, 406–409.

Finamore, S. R., & Turris, S. a. (2009). Shortening the wait: a strategy to reduce waiting times in the emergency department. Journal of emergency nursing: JEN : official publication of the Emergency

Department Nurses Association, 35(6), 509–14.

Fitzpatrick, M.K., Reilly, P.M., Laborde, A., Braslow, B., Pryor, J.P., Blount, A., Gaskell, S., Boris, R., McMaster, J., Ellis, J., Fontenot, A., Telford, G., Schwab, W. (2006). Maintaining Patient Throughput on an Evolving Trauma/Emergency Surgery Service. Journal of trauma. 60 (3),

281-288

Goldratt, E. M., Cox, J., & Whitford, D. (1992). The goal: a process of ongoing improvement (Vol. 2). Great Barrington, MA: North River Press.

Griffin, J., Xia, S., Peng., and Keskinocak, P., 2012. Improving patient flow in an obstetric unit. Health Care Management Science, 15 (1), 1-14.

Hoot, N. R., & Aronsky, D. (2008). Systematic review of emergency department crowding: causes, effects, and solutions. Annals of emergency medicine, 52(2), 126–36.

Hopp, W. J., Spearman, M. L., (2011) Factory Physics . Waveland Pr Inc; 3 edition

Kennedy, J., Rhodes, K., Walls, C. a., & Asplin, B. R. (2004). Access to emergency care. Annals of

32 Kyriacou, D. N., Ricketts, V., Dyne, P. L., McCollough, M. D., & Talan, D. a. (1999). A 5-year time study analysis of emergency department patient care efficiency. Annals of emergency medicine, 34(3), 326–35.

Laskowski, M., McLeod R.D., Friesen, M.R., Podaima, B.W., Alfa A.S. (2009). Models of Emergency Departments for Reducing Patient Waiting Times. PLOS Medicine, 4 (7), 1-11

Manchester Triage Group, 1997. In: Mackway-Jones, K. (Ed.),Emergency Triage. BMJ Publishing Group, London.

Medeiros, D. J., Swenson, E & Deflitch, C. (2008). IMPROVING PATIENT FLOW IN A HOSPITAL EMERGENCY DEPARTMENT, 1526–1531.

Miro, O., Sanchez, M., Espinosa, G., Coll-Vinent. B., Bragulat, E., Mila, J. (2003). Analysis of patient flow in the emergency department and the effect of extensive reorganization.

Emergency Medicine Journal, 20, 143-148

Nanninga, S., (2005). Wachttijden voor Spoedpatient, ”. Master Thesis. University of Groningen, NL

Neergaard, L.,(2006) Probe says U.S. emergency care in trouble. Accessed on September 18th_{, 2013 at}

http://www.alipac.us/f12/probe-says-u-s-emergency-care-trouble-28556/

Paul, S. a., Reddy, M. C., & DeFlitch, C. J. (2010). A Systematic Review of Simulation Studies Investigating Emergency Department Overcrowding. Simulation, 86(8-9), 559–571.

Soepenberg, G.D., Land, M.J., and Gaalman, G.J.C., 2008. The order progress diagram: A supportive tool for diagnosing delivery reliability performance in make-to-order companies. International Journal of Production Economics, 112 (1), 495–503

Soepenberg, G. D., Land, M. J., & Gaalman, G. J. C. (2012). A framework for diagnosing the delivery reliability performance of make-to-order companies. International Journal of Production

Research, 50(19), 5491–5507.

Solberg, L.I., Asplin, B.R., Weinick, R.M., and Magid, D.J., 2003. Emergency Department crowding: Consensus development of potential measures. Annals of Emergency Medicine, 42 (6), 824-834

Steinbrook, Robert. (1996). The role of the emergency department. The New England journal of

33 Storrow, A.B., Zhou, C., Gaddis, G., Han, H.H., Miller, K., Klubert., D., Laidig, A., Aronsky, D.

(2008). Decreasing Lab Turnaround Time Improves Emergency Department Throughput and Decreases Emergency Medical Services Diversion: A Simulation Model. Academic Emergency Medicine 15, 1130–1135

Trzeciak, S. (2003). Emergency department overcrowding in the United States: an emerging threat to patient safety and public health. Emergency Medicine Journal, 20(5), 402– 405.

Yen, K., Gorelick, M. H. (2007). Strategies to Improve Flow in the Pediatric Emergency Department.

Pediatric Emergency Care. 23 (10), 745-752

Yoon, P., Steiner, I., and Reinhardt, G. (2003). Analysis of factors influencing length of stay in the Emergency Department. Canadian Journal of Emergency Medicine, 5 (3), 155-161.

Van der Heide, R.F., (2006). Planning van de semi-acute patiënt in een diagnostisch traject op de Spoedeisende hulp. Master Thesis. University of Groningen, NL

Van der Vaart, T., Vastag, G., & Wijngaard, J. (2011). Facets of operational performance in an emergency room (ER). International Journal of Production Economics, 133(1), 201–211.

Verbree, J.M. (2013). Diagnosing and improving the patient throughput time in emergency departments. Master Thesis. University of Groningen, NL

Wagner, C. (1993). Problem solving and diagnosis. Omega, 21(6), 645–656.

Waldrop, R.D., (2009). Don’t be put out by throughput in the Emergency Department.

Physician Executive Journal, 35 (3), 38-41.

Windle, J., Mackway-Jones, K., 2003. Don’t throw triage out with the bathwater. Emergency Medical Journal 20, 119–120.

34

Appendices:

Appendix A: Process scheme

35

Appendix B: Data analysis

MAX_KLAAR_VV_DAT UM MAX_VERTREK_DAT UM Waitin g time for pickup Total treatme nt time TTT 47 15-10-2013 13:23 15-10-2013 14:56 01:33 03:21 93 100% 46 11-10-2013 14:57 11-10-2013 16:13 01:16 04:17 76 45 9-10-2013 12:02 9-10-2013 13:13 01:11 02:32 71 44 29-10-2013 14:10 29-10-2013 15:12 01:02 03:05 62 43 6-10-2013 14:27 6-10-2013 15:17 00:50 03:30 50 42 2-11-2013 14:13 2-11-2013 15:02 00:49 02:15 49 90% 41 12-10-2013 11:14 12-10-2013 12:00 00:46 02:32 46 40 30-10-2013 11:25 30-10-2013 12:10 00:45 01:39 45 39 9-10-2013 14:12 9-10-2013 14:54 00:42 03:12 42 38 6-10-2013 14:22 6-10-2013 15:03 00:41 02:04 41 37 15-10-2013 14:06 15-10-2013 14:44 00:38 02:58 38 36 29-10-2013 10:57 29-10-2013 11:33 00:36 03:00 36 35 8-10-2013 11:29 8-10-2013 12:02 00:33 02:26 33 75% 34 8-10-2013 13:06 8-10-2013 13:38 00:32 02:36 32 33 2-11-2013 13:45 2-11-2013 14:17 00:32 02:13 32 32 6-10-2013 14:13 6-10-2013 14:41 00:28 03:06 28 31 18-10-2013 11:04 18-10-2013 11:32 00:28 01:57 28 30 30-10-2013 12:01 30-10-2013 12:29 00:28 02:37 28 29 15-10-2013 14:54 15-10-2013 15:21 00:27 02:52 27 28 16-10-2013 11:37 16-10-2013 12:03 00:26 01:35 26 27 11-10-2013 13:30 11-10-2013 13:52 00:22 03:10 22 26 15-10-2013 13:24 15-10-2013 13:46 00:22 02:10 22 25 8-10-2013 13:07 8-10-2013 13:28 00:21 02:45 21 24 2-11-2013 14:18 2-11-2013 14:39 00:21 03:44 21 50% 23 6-10-2013 14:10 6-10-2013 14:30 00:20 01:15 20 22 16-10-2013 12:00 16-10-2013 12:20 00:20 01:01 20 21 9-10-2013 14:24 9-10-2013 14:43 00:19 02:29 19 20 18-10-2013 13:33 18-10-2013 13:52 00:19 03:23 19 19 2-11-2013 11:56 2-11-2013 12:15 00:19 01:36 19 18 2-11-2013 13:27 2-11-2013 13:45 00:18 01:35 18 17 8-10-2013 14:16 8-10-2013 14:32 00:16 02:33 16 16 11-10-2013 13:58 11-10-2013 14:14 00:16 01:32 16 15 11-10-2013 14:54 11-10-2013 15:10 00:16 02:10 16 14 12-10-2013 11:28 12-10-2013 11:42 00:14 02:06 14 13 29-10-2013 12:26 29-10-2013 12:40 00:14 03:42 14 12 6-10-2013 12:05 6-10-2013 12:18 00:13 01:38 13 11 16-10-2013 14:55 16-10-2013 15:08 00:13 02:20 13 25%

36

Appendix C: Interviews and coding

Coding scheme interviews:

The interviews are done within the ED with different staff representatives. One ED Physician, a ED nurse and an ED secretary.

The questions used for the interview protocol can be found in the next appendix D unanswered and the specific questions asked to the specific representatives are shown in this appendix. The answers the interviewees have given are coded to be able to analyze and conclude statements from the answers given. The analysis and discussion of the answers and comments of the interviewees is done within the framework application chapter (chapter 4) and conclusion and discussion (chapter 5). Code:

Green Highlight: Positive answer to the question __ Red Highlight: Negative answer to the question __

Gray Highlight : No clear positive or negative answer to the question __

37

Interview 1:

Transcription

Wat is uw functie binnen de SEH? SEH arts

Hoe lang bent u al werkzaam binnen de SEH? 2 jaar binnen de SEH MCL en 5 jaar SEH arts totaal.

Komt het wel eens voor dat er onvoldoende capaciteit is om de patiëntenstroom binnen aanzienbare tijd te behandelen?

Ja, hangt er vanaf wat je aanzienbare tijd noemt natuurlijk, wij gebruiken triage codes en daar moet aan voldaan worden, verschilt wel per patiënt, maar het komt zeker voor.

Zo ja, welke periodes van de week komt dit het meest voor (dagen)? Maandag en vrijdag, wordt gezegd.

**Ervaar je dat zelf ook zo?

Maandag is mijn vrije dag maar vrijdag is denk ik toch wel iets drukker. Persoonlijk ervaar ik dat niet enorm, maar als je de getallen bekijkt is het wel zo

**Met hoeveel spoedartsen zijn jullie op de SEH??

Er is één SEH arts, en daarnaast één assistent chirurgie en één assistent interne.

** Doen de assistenten chirurgie en interne ook volledige patiënt behandeling?

Dat hangt er vanaf, deze overleggen vaak met hun specialist. Wij als SEH arts zien alle categorieën patiënten.. interne ziet cardio en interne, en chirurgie ziet alleen chirurgie.. zij kunnen zelf volledig de patiënt zien in overleg met de dienstdoende chirurg of cardioloog etc. via telefonisch overleg, en soms komt hij nog even kijken.

Kunt u aangeven welke processen/behandelingen volgens u de bottleneck vormen voor deze patiëntenstroom?

Dat is een hele moeilijke vraag, specialisten in consult vooral. Als ze moeten langskomen vooral, de SEH arts kijken maar in andere gevallen waar assistenten niet gelijk de oorzaak kunnen

diagnosticeren moeten verschillende specialisten geconsulteerd worden, dit kost erg veel tijd. Geriatrische patiënten, zijn oudere mensen, die kosten veel tijd. … hebben vaak multi problematiek… etc. … ze zorgen voor stagnatie, afdelingen komen patiënten ophalen…. Logistiek is het niet

optimaal… als de patient klaar is duurt het af en toe nog wel een half uur of drie kwartier voordat de patient gehaald wordt… Dit is het geval wanneer patiënten moeten worden opgenomen door een afdeling. … de patiënten uitstroom is soms lastig…