Dealing with overcrowding, a high per centage of sickness, and a high level of stress among staff are universal problems in Emergency Departments (ED). In creating a new ED, we chose to deal with overcrowding as our major challenge. By using new IT technologies in planning standardised patient tracks as well as focusing on control of logistics, we managed to deal with major overcrowding situations. We brought down the per centage of sickness from 11 per cent to an average of 6 per cent in 2014. In addition, all patients are seen by a consultant 41 minutes after arrival and a final diagnosis and further plan for the patients is completed within 3 hours and 52 minutes.
In January 2012 a new Emergency Department (ED) opened at Odense University Hospital in Denmark —the largest ED in the country. The department should handle small casualties and traumas on level 1 and include facilities for evaluation of patients before admittance and a community service provided by general practitioners. The estimates were 350 patients per day with a day-to-day fluctuation of +/- 30 per cent.
The creation of a new set-up for emergency service was a unique chance to rethink organisation in a building that was made for that exact purpose. We decided that the most important issue to deal with the design of the set-up should be overcrowding and we identified several areas which we found important to work with.
This article describes some of the initiatives we took, massively and still supported by our Board of Directors, and how we dealt with them. Nonetheless, it is very important to emphasize the urgent need to work with attitudes and culture among the staff. We engaged a lot of energy in dealing with these issues and even displaced members in our staff to place the right people in the right positions. We have gone a long way and there is still room for improvements.
The majority of patients in a Danish ED will have been seen by a skilled physician before arriving. During the day and at ordinary times, patients who contact General Physician (GP), if required are redirected to the ED asking for acute evaluation. Outside opening hours the patients in our region have to call a unique phone number where 80 per cent mainly get in contact with a GP. The GP can ask the patients to consult their own GP in daytime, can ask the patients to come to a consultation by a GP situated in our ED, or can call the ED for an acute evaluation.
Apart from this, patients are arriving after calling our general emergency number 112. However, the number of patients arriving at the ED this way accounts for less than 20 patients per day, out of an average of 370 patients.
The described system gives our ED a unique case mix where only patients in actual need of ED-services are getting evaluated in the department. In numbers, it means that around 50 per cent of all patients having a consultation outside opening hours are seen by GPs instead of in the ED.
The new building includes a total of 10.000 m2, including 4.000 m2 at the ground floor and 4.000 m2 at the top floor. The ground floor is the diagnostic area with 4 trauma rooms, radiology, a small casualty-area, 15 consultation rooms, and an area for community service. The top floor houses up to 48 admitted patients who are expected to stay for less than 48 hours (on average, 18 hours).
The organisation of the ED in Odense
In Denmark (DK) emergency medicine is not yet recognised as a specialty. Odense University Hospital is the largest hospital in DK and our emergency department is the largest ED in DK. The hospital is represented by all recognised specialties in the country and all specialties are on duty–either by attendance at the hospital or on house call.
The department is staffed by nurses, secretaries, and a minute number of doctors (12 consultants and 8 junior doctors). The rest of the staff in our ED come from other departments to solve dedicated work (e.g. radiologist, lab physician, and cleaning) or to take care of the patients belonging to their specialty (e.g. surgeons, neurologist, and orthopaedics).
Every patient is allocated to a certain specialty upon arrival and the appointed specialist is in charge of the patient throughout the patient’s stay in the ED.
The department has a large scientific unit headed by the first professor in emergency medicine in DK. The agenda of the scientific work is to create scientific results that can be applied to our emergency set-up without too much delay.
In the former ED set-up, the youngest doctors were in front. The consultants and other doctors with major knowledge only saw those patients that the junior doctor could not handle.
We changed that scenario and made the consultants the front doctors to all new arrivals in the department. The process was not without discussions on whether we were using the right competences at the right level in patient treatment. Nevertheless, our Board of Directors decided, upon our recommendation, that we should have the highest level of competencies in front to spot patients who need specialist services.
Our experience is that placing a specialist in front was right. A case mix of older patients with chronic diseases and complications calls for higher expertise from the beginning as the symptoms of a severe illness is often disguised and overlooked (e.g. sepsis).
One of the results of having specialists in front is that, the number of patients finishing treatment in the diagnostic area is increased from 60 to 73 per cent, decreasing the number of patients being admitted by almost 5.000 over the last three years.
In preparing the patient flow through the diagnostic area, we described three different patient tracks: A fast track for small casualties (20 minutes), a track for the community service (6 minutes), and a track for patients to be evaluated for admittance (up to 240 minutes).
Each of the three tracks consists of a varied number of processes the patient has to pass through until diagnosis can be confirmed and further plans about treatment and observation of the next 8 – 12 hours can be made. Each process is defined by what, who, and the number of minutes available. An example could be blood sample, lab technician, and 5 minutes.
Combining the patient tracks with the number of patients arriving in an hour through a whole year, we had a mathematical model that could be simulated.
From the simulation, it was possible to staff the unit during 24 hours and simulate variations during the week, incorporate incidents with extraordinary incoming patients and so forth.
One of the biggest challenges in organising the patient flow was to standardise the patient’s way through the diagnostic area without losing track and always knowing if the patient was kept in flow to meet the time criteria.
For that purpose we use a special logistic board where all patients in the department are presented with logistic data. In the diagnostic area the focus is on getting the patient through to a diagnosis within four hours. The continuous surveillance is managed by a “flow master” (a senior consultant) who is the logistic manager and the leader of the doctors in the area. Every 2 – 3 hours (Figure 5) during duty he/she has a 5 minute briefing with the rest of the staff to ensure that requirements to keep the patients in flow are met. Every 4 hours the flow master meets with the nursing-coordinators trying to look 4 hours ahead of time – the purpose is to prevent overcrowding and stress.
Every time the logistic boards are used, the user leaves a time-sample to be used in our follow-up on the organisational set-up. As managers of the department, this information gives us a unique opportunity to adjust to our standardised time-schedules, locate bottlenecks, and make valid plans for changes to overcome overcrowding.
The visitation into emergency departments is traditionally based on a suspected diagnose. We are trying to change that into visitation based on a symptom combined with a triage colour and basic vital parameters. For example: pain in the upper abdomen, triage colour yellow, and with normal basic BP, pulse and temperature.
Over a period of three years the Region of Southern Denmark developed 34 somatic and 5 psychiatric packages, each by considering a major symptom combined with defined blood samples, ECG and suggestions of radiology. Upon arrival the new patient is admitted as per the purview of the packages and the investigations can begin without delay. In addition, a package describes which specialty has the definite responsibility of the patient.
The 34 somatic packages will include 97 per cent of all patients coming to the ED.
In the spring of 2013 we tested more than 17.000 patients to see if the packages they had assigned led to the final diagnosis without additional investigations. In 78 per cent of the cases, the emergency diagnosis was set from the initial investigations. When counting the number of primary diagnoses upon dismissal, we found a total of 2.147 different diagnoses. To us, it proved the value of the packages, and the method saved a lot of time 'from door to primary diagnoses'.
The location of our radiology facilities between the trauma-rooms and the small casualty-area very quickly showed impressive improvement in the time taken to decide on the need of a service from a radiologist to receiving the result.
Within weeks of opening the department, we saw a reduction in time for treatment of patients suspected of cerebral thrombosis. The door-to-needle-time was reduced from 75 minutes on an average, to 18 minutes in our new ED–almost 1 hour of cerebral ischemia saved for those patients who had an effect from thrombolysis.
The general efficiency of the Radiology Department was increased considerably. The time from referral to CTC was received until the investigation was performed decreased from 43.33 to 24.73 minutes. For chest X-ray the decrease was from 39.60 to 15.8 minutes.
Obtaining an initial plan for treatment and observation within 4 hours after arrival is based on two sets of facts. First of all, we were inspired by clinical studies showing that the patient’s mortality increase with 3 per cent for every hour they spend in an ED beyond 4 hours without a diagnosis. Secondly, the simulations we performed (see above) showed that we increased the numbers of patients in the diagnostic area to a critical level if they have stayed more than 4 hours.
By the continuous registrations on our logistic boards, we can verify that the 4-hours-criterion is met with an average of 3 hours and 52 minutes.
In planning a new emergency department, we decided to try to cope with overcrowding as our main challenge. The large numbers of patients (350 per day) with a day-to-day fluctuation of +/- 30 per cent was our challenge.
We recognised a number of options and through standardisation, simulation, and optimised logistics we succeeded in getting along the way. Having the specialist in front, reducing the time for radiology services by more than
70 per cent, creating diagnostic packages that lead to a diagnosis eliminating the need for further investigations for 78 per cent of the patients, and reaching the goal of 4 hours counted from arrival until an initial diagnosis to planning for the next hour(s) is settled and are standards now.
In our belief these results have contributed to reducing the days of sick leave among our staff by approximately 50 per cent and have helped us cope with the major challenge of overcrowding.
Nevertheless, it is important to remember to deal with the cultural changes needed among the staff in the preparations for so many changes and be ready to answer the question:'What good does it do to the patient and me?'