Abstraction

The paper presents a comprehensive research survey on the Simulation Model for Emergency Department. The survey was conducted in the Emergency Department ( ED ) at The Ohio State Medical Center. A simulation theoretical account was developed utilizing Arena ( 7. 0 ) to pattern the procedure flow of patients in order to analyse the ED System public presentation. A Statistical Design of Experiments survey was performed to analyze the significance/ non significance degrees in order to better ED public presentation.

Cardinal Wordss: Emergency Department. Simulation. Work Flow.

1 – INTRODUCTION

The Emergency Department ( ED ) at Ohio State Medical Center specializes in the intervention of critically sick and earnestly injured patients and possess a comprehensive array of the most up to day of the month diagnostic and intervention services coupled with a extremely trained and experient wellness attention staff.

It was observed that the overall clip patients spent in the Emergency Department of the infirmary was significantly higher ( 500 proceedingss ) than the benchmarked clip for the patient length of stay ( LOS ) . This has an inauspicious impact on patient throughput in the ED. If the patient length of stay in the ED is high. so the waiting times for new incoming patients in the ED is high and the in-process patient volumes who have received partial medical service in the ED is besides high. These factors contribute to a important diminution in the ED system public presentation.

The aim of our research survey was to analyse and measure ED system public presentation. The research survey was conducted in several stages. The first stage involved building of a elaborate flow chart of the “As Is” ED system. This enabled us to understand the work flow of patients. doctors and nurses in the ED system. The 2nd stage involved building of a elaborate simulation theoretical account utilizing Arena ( 7. 0 ) .

The simulation theoretical account calculates the public presentation rating steps such as entire patient length of stay. entire patient throughput. waiting times in the system. resource use. etc. The fake values are compared with the benchmarked values and the values observed when the work trying surveies were conducted. The simulation theoretical account helped us to understand the ED system kineticss and was an priceless tool for measuring the fluctuations in the public presentation steps.

We propose Thin methodological analysiss to optimise the public presentation of the Emergency Department. The thin direction based methodological analysiss were embedded in the simulation theoretical account in to obtain the ED system public presentation steps. The fake steps of public presentation were tabulated and a statistical Design Of Experiments ( Analysis of Variance ) was performed in order to obtain important and non- important factors in the survey.

2 LITERATURE REVIEW

Samaha and Armel ( 2003 ) present a simulation theoretical account and a complete analysis of operations in the Emergency Department of Cooper University Hospital. which is a 554-bed accredited installation. Miller and Ferrin ( 2003 ) simulated a big infirmary in South East USA and proposed Six Sigma-based procedure betterment thoughts for decrease of patient length of stay. .

Garcia et Al. ( 1995 ) analyzed the flow of patients at the Mercy Hospital with the nonsubjective being to minimise the waiting times of patients. The above documents describe a simulation theoretical account of the Emergency Department but do non depict equal policies and recommendations to better the public presentation of the Emergency Department. A through statistical analysis is non conducted to place important and non- important factors impacting the ED public presentation.

Centeno et Al. ( 2003 ) observed in their survey at the Baptist Health South Florida that one of the important operating costs in the Emergency suites is the staffing costs. Baesler et Al. ( 2003 ) have developed a simulation theoretical account for ciphering the maximal possible demand increase in an Emergency room of a private infirmary in Chile. The simulation theoretical account was used to make a curve that analyzes and predicts the patient length of stay in the system.

Baesler et Al. . ( 2003 ) performed a statistical Design of Experiments analysis which considered 4 factors: # of doctors. # of paramedics. # of receptionists and nature of exigency room. The above documents describe the simulation theoretical account to mensurate the operating costs in the ED but fail to propose the operations procedure betterment steps. The Analysis of Variance consequences of the analysis explains the important factors but the factors are non incorporated in the simulation theoretical account to analyze the improved response steps.

Bard et Al. ( 2005 ) discuss the jobs. hospital direction faces several times in a twenty-four hours as the demand for wellness attention services departs from the planned agenda. Harrison et Al. ( 2005 ) present a stochastic version for the Harrison Millard multi- phase theoretical account of the flow of patients through a hospital division in order to pattern right non merely the norm but besides the variableness in tenancy degrees.

Bard and Purnomo ( 2005 ) present two theoretical accounts to work out the midterm penchant scheduling job based on displacement position preparation. The above documents describe a additive whole number programming preparation to optimally work out the staff planning job in the infirmary. The math-based theoretical accounts are nevertheless able to optimally calculate solutions to little size jobs and neglect to work out big size jobs particularly when the infirmary staff size is big.

Akcali et Al. ( 2006 ) present a web flow attack to optimising infirmary bed capacity determinations. Basu Ghosh and Cruz George ( 2006 ) present a Physician Requirements Planning theoretical account in response to high demands for doctors in infirmaries. The Physician Requirement Planning theoretical account is an unreal intelligence based package system used for ciphering the figure of doctors and physician rolls based on the physician agendas. The package system is basically restricted to physician planning but does non capture the ED operations be aftering and direction facets.

Kevin Leonard ( 2004 ) studied the function of patients in planing wellness information systems and the instance of using simulation techniques to plan a patient record interface. Baker ( 2002 ) studied the sensitiveness analysis for wellness attention theoretical accounts utilizing statistical methods. Their research determined which parts of the theoretical account caused greatest uncertainness in the prognostic theoretical accounts and is a determination support tool for the modeller. assisting them to polish the theoretical account further or collect extra informations. Beguin and Simar ( 2004 ) analyze disbursals linked to hospital corsets and a methodological analysis to cipher outliers.

Jones et Al. ( 2002 ) depict a prediction theoretical account that forecasts the day-to-day figure of occupied beds due to Emergency Admissions in the infirmary. They discovered that the figure of occupied beds is related to Emergency Admissions. Utley et Al. ( 2003 ) address the inquiry of what degree of capacity is required to run a system if cancellations of engaged patients are kept to a low degree. Karnon et Al. ( 1998 ) discourse the suitableness of patterning techniques for economic ratings of wellness attention plans in general.

Rohleder et Al. ( 2007 ) study on the usage of simulation patterning for redesigning venesection and specimen aggregation centres at a medical diagnostic research lab. The aim of their research was to cut down mean waiting times and their variableness.

Channouf et Al. ( 2007 ) develop and measure clip series theoretical accounts of call volume to Emergency Medical Service in a Canadian metropolis. Denton et Al. ( 2007 ) worked on the job of sequencing and scheduling surgeries under uncertainness. Littig et Al. ( 2007 ) developed an analytical theoretical account for anticipation of short term infirmary tenancy.

3. EMERGENCY DEPARTMENT ANALYSIS

We analyzed the current province of the Emergency Department at OSU-MedCtr and constructed a Flow Process Chart of the system. The Flow Process Chart is explained by Figure 11 and Figure 12. The procedure that was charted is presented in two stages: Phase 1 describes patient medical procedure till the lab operations are performed and Phase 2 is from the point of patient lab operations to patient temperament from ED to the infirmary floor or place. [ movie ]

Figure 1. Emergency Department Phase 1

[ movie ]

Figure 2. Emergency Department Phase 2

The procedure starts with the reaching of the patient at the entryway of the Emergency Department and subsequent patient preregistration. checking of critical marks. and patient triage. The procedures in the ED system vary depending on the medical status of the patient. Medical intervention in the Emergency Department involves medical scrutiny by ED doctors. medical scrutiny by forte advisers. intercessions provided by nurses and medicines. The figure of doctors and forte adviser visits is dependent on nature of the medical status of the patient. While waiting ( i. e. . clip hold ) for the patient to react in some instances. the following medical intercession is delayed until the reaching of forte advisers for medical scrutiny.

The medical intervention of the patients besides involves disposal of trials such as lab. X rays. CT. MRI. etc. On norm it was observed that the figure of physician ( ED physician visits ) per patient varied from 3 to 4. While medicines provided require a Physician order. other intercessions may or may non necessitate the doctor. The nurse: patient ratio for ague attention presently in the system was observed to be 1: 3 while that for the fast path was observed to be 1:7. However it was observed that the ratio alterations due to changing staff handiness and staffing capacity every bit good as high variableness in the patient demand. The alterations to the nurse: patient ratio was initiated by the charge nurse.

In our research survey we propose a variable “cycle time” ( besides defined as length of stay ) . The motive behind presenting this factor was that patients with high sharp-sightedness may hold high or low length of stay whereas patient with low sharp-sightedness may hold high or low length of stay. Therefore we have four different instances. depending on patient high and low sharp-sightedness degrees and patient high and low length of corsets.

It was observed that the nurse: patient ratio was different for all possible scenarios depending on patient’s sharp-sightedness degree and rhythm clip in the system. The expected patient length of stay at the oncoming would help the charge nurse in optimising the nurse: patient assignment by rapid dynamic updating of nurse: patient ratio. The rapid dynamic updating of nurse: patient ratio would guarantee complete patient satisfaction and optimise the system public presentation steps.

It is noted that if the nurse: patient ratio is kept a invariable for the full continuance of the twenty-four hours. high quality and high service degree of patient attention can non be achieved. The patient demands fluctuate with a high grade of fluctuation at different clip intervals of the twenty-four hours and for different yearss of the hebdomad. Hence if the nurse: patient ratio is kept as changeless. the nurse: patient assignment would non take into history the sharp-sightedness degrees of the entrance patients.

This would take to low service degrees of patient attention. patient dissatisfaction and increase in patient length of stay. Hence the dynamic updating of nurse: patient ratio degrees are carried by the charge nurse topic to high variableness in patient volumes and sharp-sightedness degrees. In the OSU- MedCtr ED installation. a white board is deployed to enter the patient IDs. nurse IDs and nurse: patient assignment. The white board is dynamically updated by the charge nurse.

Patient temperament could either be patient floor admit or patient discharge. This temperament determination is made by an ED doctor in audience with the forte adviser. The temperament determination in the current system is non taken preemptively by the doctors or the forte advisers but is merely taken in the ulterior phases of the class of intervention in the Emergency Department. It was observed that in the instance of patient admits to hospital floors. the scrutiny by forte advisers was observed to be of frequent happening. The patient admittances on the floor were delayed on history of the floor beds non being ready for the admittance of new patients.

As a consequence it was observed that the patients had to remain in the Emergency Department for a important sum of clip although their medical intervention in the Emergency Department was complete. Hence this resulted in a important sum of hold for admittance of new geting patients in the Emergency Department. which is a major concern from the point of position of criticalness of patients and their medical status and a loss of gross to the infirmary due to possible loss of patients. We performed clip surveies and collected informations from the patient record sheets. pertinent to the above the flow chart in Figure 11 and Figure 12.

A simulation theoretical account that was built in Arena 7. 0. The three yearss were sampled from a month’s period as we observed that the patient volumes in the ED stayed changeless for each hebdomad but varied on different yearss of the hebdomad. We assumed that the patient reachings are exponentially distributed with a mean of 15 proceedingss. Data aggregation for service times of medical procedure activities was conducted and distributions computed to cipher mean and standard divergences. We simulated the system for a period of one twenty-four hours. The one-day ( 1440 proceedingss ) was the clip for which the Arena Simulation theoretical account was run.

4. SIMULATION MODEL OF EMERGENCY DEPARTMENT

The input informations for the simulation theoretical account was follows: The waiting times of the triage follow a unvarying distribution between ( 5- 10 min ) . The mean waiting times of the bed assignment follows a unvarying distribution between ( 15 – 20 min ) . Since one nurse attends on mean 3 to 4 patients at any given clip. the mean waiting times due to intervention of other patients follows a unvarying distribution from ( 10 – 20 min ) . The mean waiting times for the physician to get is 5 proceedingss for ague attention and 15 proceedingss for fast path. [ pic ]

Figure 3. Simulation Model

Figure 3 explains the Simulation Model of Patient Process Flow.

The Emergency Department staff has a inclination to batch trial consequences and the mean waiting times due to batching of trial consequences is 32 min. Since Lab is a constriction resource in the Hospital and receives petitions from all inpatient units. surgery section. ICU and the Emergency Department. the waiting times for the reaching of lab trial consequences follows a unvarying distribution from ( 60 – 90 min ) . The waiting times of the forte adviser are on an mean 35 proceedingss whereas the waiting times of the patients for acquiring to the infirmary floor bed follows a unvarying distribution from ( 60-75min ) . The mean # of people in the ED at any given clip is on mean 17.

This is normally referred to as the “work in process” stock list of the ED ( this merely includes people being served plus the people in the waiting room ) . Throughput is defined as the entire figure of patients served by the Emergency Department in a twenty-four hours from the clip of patient entry at the preregistration to the clip of patient temperament. Cycle clip is defined as the length of clip from the patient entry at the preregistration to the patient temperament ( either discharged place or sent to the infirmary floor ) .

The rhythm clip is besides defined as the patient length of stay in the Emergency Department. The mean patient throughput is 132. The mean length of stay of patients discharged place is equal to 55 proceedingss. The mean length of stay of patients traveling to hospital floor bed is equal to 504 proceedingss. The % use of ED nurse. Triage Nurse and ED Physician are close to 100 % . This implies that the ED system is overutilized. This motivated us to develop thin methodological analysis based recommendations for optimising the public presentation of the ED system.

We observed that the important waiting times in the Emergency Department were waiting clip for the infirmary floor bed to be ready. waiting clip for the Emergency Department bed to be ready. waiting clip for lab trial consequences to arrive and waiting clip for the reaching of forte adviser.

These waiting times result in constrictions in the system and are cardinal factors lending to high patient length of stay in the ED. The waiting line waiting times for the fast path physicians and nurses was observed to be zero. The mean waiting times for ague attention doctors was observed to be on mean 10 proceedingss. These clip holds in the system motivated us to develop fresh thin direction based methodological analysiss. The nonsubjective behind this attack was clearly to better ED throughput. cut down patient LOS and patient waiting times.

Thin methodological analysiss were developed for Emergency Department operations are listed below. a ) Execution of Triage Short Form V. Regular Form Triage – The construct of Triage Short Form stemmed from the demand to cut down the clip taken for triage in order to cut down the entire length of stay of patients. B ) Execution of Visual Display for Dynamic Nurse: Patient Ratio – The construct of Visual Display stemmed from the demand to cut down the clip taken to make the nurse intercession. cut down the hold for delay for nurse and therefore minimise the entire patient length of stay.

degree Celsius ) Execution of Preemptive Disposition Decision Making by ED

Physician – The construct of Preemptive Disposition Decision Making stemmed from the demand to cut down the clip taken by the ED Physician to go to to the patient and therefore minimise the patient entire length of stay and guaranting good patient attention and satisfaction. The above methodological analysiss were besides treated as Factors ( Independent Variables ) for the Design of Experiments. The Response Variables in the experiment were Entire Patient Throughput in a twenty-four hours. Entire Patient LOS for Fast Track Patients & A ; Acute Care Patients Total Patient Length of Stay for Acute Care Patients. The Design of Experiments involves the execution of the ANOVA process ( t trial ) in order to measure the significance / non significance degrees of the factors.

Each Factor is tested at 2 degrees: High Level ( + ) and Low Level ( – ) . Triage Short Form Strategy involved unhinging the clip taken to make the triage operation at high and low degree ( i. e. . high and low times to execute triage operation ) . Ocular Display for Nurse to Patient Operation involved distressing clip hold for carry oning Nurse Intervention at high clip values and low clip values. Preemptive Disposition Decision Making Strategy involved distressing clip hold to carry on ED Physician scrutiny at high and low clip values.

Sensitivity Analysis of the Simulation Model was carried out as follows: Each factor is embedded in the Simulation Model one at a clip and the response variables are evaluated. All five response variables are ab initio considered and Multi Variate Analysis of Variance ( MANOVA ) is conducted for happening important factors. After obtaining response values for single chief effects. response values with regard to two manner interactions and three manner interactions are obtained. Thus we obtain a shaping contrast tabular array for a 2k factorial design.

The degree of significance is decided if the P value is less than 0. 1 and undistinguished if the P value is more than 0. 1. R Square value is observed to be low equal to 17. 58 % . For the 2nd response variable. The R Square value is observed to be 86. 79 % which indicates the theoretical account tantrum is good. For other response variables none of chief effects were observed to be important and the R Square values were observed to be low. therefore bespeaking that the independent variables do non adequately predict the response variables.

5. CONCLUSIONS & A ; FUTURE RESEARCH

We analyzed the Emergency Department at The Ohio State University Medical Center. The work flow in the Emergency Department was modeled utilizing the simulation and with computational consequences were presented. The simulation theoretical account considered the several system public presentation rating steps such as patient throughput. patient length of stay. patient waiting times etc. which provide several penetrations in our analysis.

We identified the independent variables ( factors ) based on Lean Thinking attacks which could potentially impact the ED system. Next we conducted an Analysis of Variance process for placing the important and non- important factors in the survey. These factors were embedded into the simulation theoretical account to measure the new ED system public presentation rating steps. Future work could affect develop a Stochastic Mixed Integer Programming Model for analysis of Emergency Department. The MIP Model could potentially be solved with ILOG CPLEX Solver.

5. Reference

Akcali. E. . Cote. J. M. . Lin. C. A Network Flow Approach To Optimizing Hospital Bed Capacity Decisions. Health Care Management Science. 9. 391-404. 2006.

Baker. Rose. D. Sensitivity Analysis of Health Care Models Fitted To Data By Statistical Methods. Health Care Management Science. 5. 275-281. 2002.

Baesler. F. F. . Jahnsen. E. H. Emergency Department I: the usage of simulation and design of experiments in gauging maximal seating capacity. Proceedings of 2003 Winter Simulation Conference.

Bard. J. F. . Promos. W. H. Short Term Nurse Scheduling In Response to Daily Fluctuations In Supply & A ; Demand. Health Care Management Science. 8. 315-324. 2005.

Bard. J. F. . Purnomo. W. H. Incremental Changes in the WorkForce to Accommodate Changes in Demand. Health Care Management Science. 9. 71-85. 2006. Beguin. C. and Simar. L. Analysis Of the Expenses Linked to Hospital Stays: How To Detect Outliers. Health Care Management Science. 7. 89-96. 2004. Centeno. A. M. . Giachetti. R. . Linn. R. . Ismail. A. Emergency Department II. a
simulation based ip tool for scheduling ER staff. Proceedings of 2003 Winter Simulation Conference.

Carter. W. M. . Lapierre. D. S. Scheduling Emergency Room Physicians. Health Care Management Science. 4. 347-360. 2001.

Harrison. W. G. . Shafer. A. . Mackay. M. Modeling Variability In Hospital Bed Occupancy. Health Care Management Science. 8. 325-334. 2005.

Leonard J. K. The function of patients in planing the wellness information systems. The instance of using simulation techniques to plan patient electronic record interface. Health Care Management Science. 7. 275- 284. 2004.

Harper. P. A model For Operational Modeling Of Hospital Resources. Health Care Management Science. 5. 3. 165-173.

Karnon. J. . and Brown. J. Choosing a Decision Model For Economic Evaluation: A Case Study And Review. Health Care Management Science. 1. 133-140. 1998.

Miller. . J. S. . SzmannskY Starks. W. D. . Ismail. M. A. Emergency Departments II. Imitating Six Sigma Improvement Ideas for infirmary Emergency Department. Proceedings of 2003 Winter Simulation Conference

Mahapatra. S. . Koelling. C. P. . Patvivatsiri. L. . Fraticelli. B. . Grove. E. Pairing Emergency badness index – 5 degree triage informations with computing machine aided system design to better Emergency Department. Proceedings of Winter Simulation Conference 2003.

Gandjour. A. . Weyler. E. J. Cost Effectiveness of Referrals to High Volume Hospitals. An analysis based on Probabilistic Markov Model for Hip Fracture Surgeries. Health Care Management Science. 5. 3. 2002.

Puig. J. J. . Jaume. S. . Martinez. G. E. Why do Patients prefer Hospital Emergency Visits. A Nested Multinomial Logit Analysis For Patient Initiated
Contacts. Health Care Management Science. 1. 1. 39-52.

Rohleder. T. R. . Klassen. K. J. Rolling Horizon Appointment Scheduling. A Simulation Study. Health Care Management Science. 5. 3. 201-209. 2002.

Samaha. S. . Armel. S. . Starks. W. D. . Emergency Departments I. The usage of Simulation to cut down patient length of stay in Emergency Department. Proceedings of 2003 Winter Simulation Conference.

Utley. M. . Gallivan. S. . Treasure. T. . Valencia. O. Analytical Methods For Calculating Capacity Required To Operate An Effective Booked Admissions Policy For Elective Inpatient Services. Health Care Management Science ( 6 ) . 97-104. 2003.