Abstract
Objective
To determine the interaction and waiting times of patients in Barbados Queen Elizabeth Hospital Accident and Emergency Department (A&E), identify their determinants, and compare them to international benchmarks.
Methods
Trained research assistants monitored all patients attending A&E during a one week period in 2003. The time in and out of each step in the process of care was recorded along with demographic data, diagnosis, and acuity. Time elapsed was calculated for various steps in care and compared to international benchmarks. Determinants of timely care were identified.
Results
Of 882 eligible A&E patients, 675 (77%) had accurate data and were entered into the study. Interaction times were short, with median total interaction time 13 (IQR 9–21) minutes. Waiting times were long ranging from median 6.5 (IQR 2–22) minutes for registration to 213 (IQR 154–316) minutes for lab results. Of concern was a median wait of 10 (IQR 2–46) minutes for triage and 178 (IQR 105–305) minutes to be seen by a doctor. Mean total length of stay was 377 (SD 261) minutes compared to US benchmark of 90 minutes. All other waiting times were at least twice US benchmarks. Paediatrics cases and children aged 0–11 years had the shortest waiting times and length of stay, whereas medicine patients and those over 49 years had the longest. Those with highest acuity had the shortest waiting times and length of stay.
Conclusions
The A&E could improve patient care processes by shortening waiting times, especially for laboratory results, triage, and seeing a doctor, particularly for older medicine patients.
Keywords: emergency medicine, quality care, timely care, benchmarking
The Institute of Medicine's Committee on Quality of Healthcare in America 2001 report recommends that health care should be delivered by systems that are designed to provide care that is safe, effective, patient centred, timely, efficient, and equitable.1 Thus, to be considered quality care, care delivery must be timely. This is particularly true in the emergency department (ED). Numerous ED patient surveys have shown that timeliness is a very important contributor to patient satisfaction.2,3,4,5,6,7,8,9,10,11,12,13 As patient satisfaction in the ambulatory setting is correlated with other important outcomes, including higher compliance, decreased utilisation of medical services, and less malpractice litigation,1,14 achieving timely delivery of service in the ED has significant implications for population health.
Timeliness of care, as measured by time studies, has been studied in many healthcare settings, including ED settings.15,16,17,18,19,20 In order to determine acceptable waiting times, international benchmarks are consulted. For ED services, benchmarks used are those developed in the US13,21,22—the only published ones available. They were developed at a cross section of district, teaching, and tertiary hospitals, during 1997–2000.
Barbados is a moderately developed island nation of 275 000 people located in the Eastern Caribbean. The Queen Elizabeth Hospital (QEH), the only state funded acute care facility on the island, is a 600 bed university affiliated hospital located in Barbados' urban core. The Accident and Emergency (A&E) department of QEH has 45 000 patient visits per year. Facilities include 11 acute treatment beds (including two for resuscitation and one for paediatrics), a general holding/observation area with three beds, a two bed long stay area, and a dedicated area for treatment of up to nine asthmatic patients. Usually three doctors, one fully trained and two postgraduate trainees, staff the facility each shift (absentees may reduce this). A patient survey has documented patient dissatisfaction with waiting times in A&E.23
The goal of this study was to determine the time spent by patients in the QEH A&E in and between interactions, compare them to international benchmarks for timely care, and identify the determinants of timely care in order to make proposals for change that might improve the timeliness and hence the quality of health care delivered.
Methods
Ethical approval for the study was obtained from the Institutional Review Board of the University of the West Indies.
A time sheet data collection instrument was developed based on the patient flow through the A&E. Six research assistants (RA) were trained in the study methods and maintenance of patient confidentiality.
A 24 hour pilot study was conducted starting at 0800 on a day chosen randomly from the month of June 2003. The RAs initiated a time sheet for every patient that came into the A&E. Demographic data and triage category were obtained from the patient chart. Triage categories were assigned by a nurse: category 1—patients who must be seen immediately; category 2A—patients who can wait up to 30 minutes to be seen (suspicious headache, chest pain, open fractures, etc); category 2B—patients who can wait up to six hours (diabetes, abdominal pain, serious lacerations); category 3—patients who may be seen elsewhere (uncomplicated urinary tract infection, gastroenteritis, minor lacerations). The time of arrival to A&E was noted and subsequently, throughout the patient's movement through A&E, the “time in” and “time out” at every step was noted and entered by the RA. Personnel working in A&E were informed of the study and the fact that all data concerning both the patient and the caregivers were anonymous. Neither the patients nor the staff were involved in the data collection process.
When the pilot ended, the data pertaining to those patients whose visit was still incomplete were discarded. Because over 95% of the sheets produced were correctly completed and no data errors were detected, it was decided to proceed with the actual study without changing the methodology. A full week of data was collected by starting at 0800 on a randomly selected Thursday morning in November 2003 and ending at 0800 on the subsequent Wednesday morning and adding the data from the pilot study (which was a Wednesday).
The interaction time at each station and the wait time between each station was determined and the following variables calculated: length of stay (LOS)—time from arrival to final disposition: discharged, admitted, or left without treatment (LWT) (defined as leaving before being brought into a cubicle); total interaction time (the time the patient was under care and treatment); total waiting time; for patients who were referred, waiting time between referral and being seen by specialist.
Using Statistical Package for the Social Sciences (SPSS version 11 for Windows), the various times were tested for normality of distribution. Since distributions were not normal, non‐parametric tests (Mann Whitney U test for binary variables and the Kruskal‐Wallis test for variables with multiple choices) were used to compare the median times of various groups within the population (p<0.05 was considered significant).
Hospital data on A&E visits for the year 2003 were obtained for comparison with the study week.
Results
Of 882 eligible patients, 743 (84.2%) had data collected. Of these, 68 (9.2%) were discarded because of data errors and 675 (76.5% of those eligible) were used for the analysis although not all had complete data. When the study week was compared to an average week in 2003, a significantly higher proportion of asthmatics and category 1 patients (28% v 24% and 32% v 28%, respectively, p = 0.04 for both) were seen. (Asthmatic patients are triaged as category 1 in A&E.)
The characteristics of study patients are given in table 1. Most (68%) of the patients came to A&E between the hours of 9 am–9 pm. About 10% of the daily census of patients arrived during the night (midnight–6 am).
Table 1 Characteristics of study patients (n = 675).
| Characteristic | Number | Proportion (%) |
|---|---|---|
| Age: 0–11 yrs | 145 | 22.7 |
| 12–20 yrs | 101 | 15.8 |
| 21–49 yrs | 229 | 35.8 |
| ⩾50 yrs | 164 | 25.7 |
| Missing data | 36 | |
| Gender: Male | 321 | 50.1 |
| Female | 319 | 49.9 |
| Missing data | 35 | |
| Specialty: Medicine | 264 | 42.1 |
| Paediatrics | 162 | 25.8 |
| Orthopaedics | 71 | 11.3 |
| Surgery | 70 | 11.2 |
| O&G | 31 | 4.9 |
| EENT | 29 | 4.6 |
| Missing data | 48 | |
| Acuity: Emergency (1) | 228 | 34.0 |
| Urgent (2A) | 400 | 59.6 |
| Non‐urgent (2B) | 43 | 6.4 |
| See elsewhere (3) | 0 | 0 |
| Missing data | 4 | |
| Had vitals done | 568 | 84.1 |
| Had laboratory test done | 142 | 21.0 |
| Had x ray done | 133 | 19.7 |
| Had CT scan done | 24 | 3.6 |
| Had referral to specialist | 152 | 22.5 |
| Seen by specialist | 52 | 7.7 |
| Admitted to hospital | 111 | 16.4 |
CT, computed tomography; EENT, eye, ear, nose, and throat; O&G, obstetrics and gynaecology.
The time required for the various patient interactions and the waiting times between those interactions are listed in table 2.
Table 2 Descriptive statistics of various waiting and interaction times in minutes.
| n | Mean | SD | Median | IQR | Minimum | Maximum | |
|---|---|---|---|---|---|---|---|
| Interaction times | |||||||
| Registration time | 656 | 1.8 | 1.4 | 1.0 | 1–2 | 1 | 12 |
| Triage time | 618 | 1.7 | 1.4 | 1.0 | 1–2 | 1 | 17 |
| Vitals time | 675 | 1.8 | 1.9 | 1.0 | 1–2 | 1 | 28 |
| Exam time by EP | 609 | 7.6 | 4.4 | 6.0 | 5–10 | 1 | 36 |
| Admission clerking time | 94 | 6.1 | 2.8 | 5.0 | 4–8 | 1 | 18 |
| Total clerking time | 660 | 2.7 | 2.7 | 2.0 | 1–3 | 1 | 19 |
| Total interaction time | 675 | 13.8 | 67.9 | 13.0 | 9–21 | 1 | 532 |
| Waiting times | |||||||
| Wait for registration | 662 | 18.4 | 29.8 | 6.5 | 2–21.5 | 0 | 239 |
| Wait for triage | 675 | 39.5 | 68.5 | 10.0 | 2–46 | 0 | 660 |
| Wait for cubicle | 582 | 92.5 | 139.0 | 30.0 | 2–131 | 0 | 780 |
| Wait in cubicle | 570 | 138.7 | 132.9 | 107.0 | 49–192 | 0 | 1140 |
| Wait to see EP | 622 | 234.1 | 192.2 | 178.0 | 105–305 | 0 | 1318 |
| Wait for treatment | 339 | 133.1 | 199.7 | 35.0 | 8–205 | 0 | 1299 |
| Wait to 1st review | 162 | 169.2 | 129.9 | 135.0 | 81–236 | 5 | 790 |
| Wait for specialist | 43 | 48.2 | 76.7 | 19.0 | 4–55 | 0 | 390 |
| Wait for lab results | 46 | 236.4 | 104.9 | 213.0 | 154–316 | 62 | 486 |
| Wait for X‐Ray results | 86 | 75.8 | 77 | 52 | 28–94 | 8 | 413 |
| Wait for CT results | 16 | 92.3 | 48.1 | 76.0 | 52–121 | 39 | 210 |
| Wait after admission | 103 | 169.7 | 172.9 | 87.0 | 43–295 | 4 | 775 |
| Wait after discharge | 501 | 15.7 | 73.5 | 3.0 | 0–8 | 0 | 950 |
| Total time in A&E (LOS) | 657 | 377 | 261.3 | 302 | 180–545 | 2 | 1397 |
A&E, accident and emergency; CT, computed tomography; EP, emergency physician; IQR, interquartile range; LOS, length of stay; SD, standard deviation
Non‐parametric analysis of associations between several important times and patient characteristics are seen in table 3.
Table 3 Median times in minutes by specialty of diagnosis, age, and acuity.
| Wait to cubicle median (IQR) n = 582 | Wait in cubicle median (IQR) n = 570 | Wait after admission median (IQR) n = 103 | Total time in A&E (LOS) median (IQR) n = 657 | |
|---|---|---|---|---|
| Overall | 30 (2–131) | 107 (49–192) | 87 (43–295) | 303 (180–545) |
| Specialty of Diagnosis | ||||
| Medicine | 18 (0–130) | 130 (68–228) | 155 (75–338) | 394 (214–652) |
| Paediatrics | 5 (0–29) | 119 (77–187) | 30 (10–52) | 225 (141–340) |
| Surgery | 100 (37–203) | 76 (16–178) | 122 (55–228) | 345 (245–555) |
| O & G | 79 (44–227) | 82 (15–126) | 40 (9–88) | 395 (156–602) |
| Ortho | 80 (20–232) | 88 (25–200) | 99 (48–214) | 354 (230–593) |
| EENT | 74 (26–170) | 65 (16–97) | 49 (17–49) | 235 (104–353) |
| Age group | ||||
| 0–11 yrs | 5 (1–116) | 110 (67–206)* | 45 (15–63) | 235 (150–348) |
| 12–20 yrs | 7 (0–92) | 103 (44–180)* | 30 (20–75) | 244 (147–425) |
| 21–49 yrs | 45 (5–132) | 106 (51–188)* | 95 (33–294) | 310 (202–585) |
| ⩾50 yrs | 60 (9–189) | 103 (31–241)* | 100 (60–325) | 448 (280–675) |
| Acuity | ||||
| Emergency | 1 (0–5) | 141 (90–200) | 55 (10–363)* | 195 (141–279) |
| Urgent | 87 (30–197) | 92 (28–193) | 95 (45–296)* | 430 (265–642) |
| Non‐urgent | 60 (9–189) | 103 (31–241) | 100 (60–325)* | 448 (280–675) |
*No significant association found.
EENT, eye, ear, nose, and throat; IQR, interquartile range; LOS, length of stay; O&G, obstetrics and gynaecology; ortho, orthopaedics.
All associations p<0.05 by Kruskal‐Wallis test unless indicated.
There were no statistically significant associations between any of the waiting times and gender. In general, more acutely ill, younger, paediatric patients had lower median waiting times to reach a cubicle and to be admitted resulting in shorter LOS. However, these same patients waited as long or longer to be seen after they reached the cubicle.
There was significant association between interventions and patient LOS (table 4). When the relation between acuity and interventions was examined, category 2A patients were significantly more likely to have laboratory tests, x ray examination, and referrals than either category 1 or category 2B patients (p<0.001 for all).
Table 4 Association between various procedures and total stay in A&E (LOS) (n = 657).
| Intervention | n | Median (min) | Minimum (min) | Maximum (min) | IQR | |
|---|---|---|---|---|---|---|
| IV access* | Yes | 131 | 515 | 42 | 1397 | 340–662 |
| No | 526 | 261 | 2 | 1274 | 162–455 | |
| Laboratory testing* | Yes | 135 | 515 | 42 | 1397 | 347–654 |
| No | 522 | 260 | 2 | 1274 | 161–446 | |
| x Ray* | Yes | 130 | 389 | 70 | 1397 | 242–625 |
| No | 527 | 283 | 2 | 1350 | 165–516 | |
| CT* | Yes | 24 | 600 | 225 | 1274 | 438–890 |
| No | 633 | 295 | 2 | 1397 | 175–525 | |
| Referral* | Yes | 146 | 448 | 29 | 1397 | 282–638 |
| No | 511 | 265 | 2 | 1274 | 163–485 | |
| Admission* | Yes | 97 | 475 | 30 | 1350 | 287–638 |
| No | 560 | 281 | 2 | 1397 | 166–494 | |
| Overall | 657 | 302 | 2 | 1397 | 180–545 | |
*p<0.001 by Mann‐Whitney U test
CT, computed tomography; IQR, interquartile range; IV, intravenous; LOS, length of stay; min, minutes.
Discussion
During the patient care process at A&E, the average patient waited for long periods of time for each interaction. However, the actual interaction times were short. Of the total median LOS of 302 minutes, 13 minutes (4.3%) was interaction time and the rest was waiting time. Only six minutes (2% of total visit time) was spent with the emergency physician—the core service the patient was seeking. The remaining seven minutes of patient interaction time was spent at registration, clerking, triage, and other such activity. These activities are unlikely to be considered valuable by the patient.
Comparing QEH A&E times to the international benchmarks (table 5) shows major scope for improvement. The only area where A&E performs better is in the registration process, which takes 1–2 minutes compared to up to 10 minutes in the US probably because only demographic and clinical information is solicited, not the detailed payment related data required in the US. The only other area where A&E performs close to current or worthy interim goal times is the x ray turn around time (TAT). All remaining times are at least twice as long as benchmark times.
Table 5 Mean A&E waiting times compared to international benchmarks.
| Mean A&E time (min) | Mean US time21 or worthy goal22 (min) | Recommended time (min) | ||
|---|---|---|---|---|
| Registration | 1.8 | >1021 | 5–1021 | |
| Wait to triage | 39.5 | – | 024 | |
| Wait to reach cubicle | 92.5 | 1522 | 1022 | |
| Wait in cubicle | 138.7 | 1522 | 1022 | |
| Door to EP | 234.1 | 2222 | 1422 | |
| Wait before treatment | 133.1 | 6021 | 20–3021 | |
| Lab TAT | 236 | 6021 | 30–4021 | |
| x Ray TAT | 75.8 | 6021 | 20–3021 | |
| Wait after admission | 170 | 4522 | 2422 | |
| Wait after discharge | 15.7 | – | 1513 | |
| Total LOS | Admitted | 486 | 240+21 | 120–24021 |
| Discharged | 357 | 15021 | 60–9021 | |
| Overall | 377 | 11522 | 9022 | |
A&E, accident and emergency; EP, emergency physician; LOS, length of stay; min, minutes; TAT, turn around time.
Wuerz et al24 have shown that triage assessment using a three level classification is usually inconsistent. Inefficient triage in the ED has been linked with prolonged waiting times and resultant patient dissatisfaction.25 Therefore, if waiting times are to be reduced, safe and effective triage is mandatory.
The average patient in A&E waits a mean of 40 (median 10) minutes to be triaged. This is potentially dangerous and negates the basic tenet of triage. In addition, there were a large proportion of patients triaged in category 2A and no patients assigned to category 3, reducing the number of categories to 3. This may reflect inadequate triaging and could lead to a compromised level of care for those who are seriously ill. Clearly the triage process at A&E must be improved.
The youngest patients and those with paediatric diagnoses had the shortest wait to reach a cubicle, whereas those ⩾50 years old and those with surgical diagnoses waited the longest period of time. Since paediatric patients can be assigned to the same cubicles as adults, they are being given preference by nursing staff selecting the next patient to be brought to a cubicle. The reasons for this need to be explored.
As expected, the most urgent cases (category 1) had the shortest wait for a cubicle, but, surprisingly, the category 2B patients were placed in a cubicle more quickly than the category 2A patients. The reasons for this need to be elucidated but may include incorrect triage.
Paediatrics and medicine patients and patients triaged to category 1 waited longest in the cubicle. Surgery and category 2B waited for the shortest duration in the cubicle. The latter is contrary to the expectation that seriously ill patients are seen quicker. Possible reasons include inconsistency of triage and the fact that some category 1 patients—for example, asthmatics, those needing oral rehydration, those in sickle cell crisis (mainly medicine and paediatrics patients)—are treated immediately on arrival by nurses following a protocol. The EP sees these patients only on completion of their treatment. Therefore, these patients wait longer to see the EP.
Patients who have been admitted prior to their laboratory and/or x ray results arriving in A&E would not have a laboratory and/or x ray TAT. This, along with missing data, accounts for the fact that most patients having these tests did not have TAT values. The mean laboratory TAT in the A&E of 236 minutes is far above the benchmark recommendation of 30 minutes and the typical laboratory TAT in the US of 60 minutes. Attention to reducing the TAT would significantly improve overall A&E LOS. x Ray TAT was much less than lab TAT and much closer to benchmark values. Hence, x ray TAT cannot be counted as a serious problem in A&E probably because there is a radiology suite in the A&E.
The shortest wait for admission was for paediatric patients and adults with obstetrics and gynaecology and eye, ear, nose, and throat diagnoses. All fell within what might be expected at a US hospital. Older adults and those with medical, surgical, and orthopaedic diagnoses waited for the longest periods probably because of a recognised shortage of adult inpatient beds, especially monitored beds, to which many medicine patients are admitted. This leads to prolonged boarding of these patients in the department and increased congestion in the A&E.
A large variation in the waiting time after discharge was noted in this study and some were very long. Because the EP performs all discharge formalities, including discharge instructions, letters, and prescriptions, the discharged patient requiring this paperwork must wait until the EP is free to do it, contributing to congestion and LOS.
Younger patients were seen more quickly and had a shorter disposition time; thus, reducing their LOS. But the system seems to fail for older patients, who spend an inordinately long period in A&E. The main contributors to this longer LOS include waiting for specialist review, laboratory results, inpatient beds, or waiting after discharge.
The vast majority of category 1 and 2B patients had LOS ⩽ median time. For category 1 cases, this is due to these patients being seen earlier and their disposition arrived at earlier (either because of the urgency of the case or because there was not any doubt about admissibility). The category 2B patients do not need extensive investigations as part of their A&E visit, which significantly reduces the LOS. In contrast, the longer LOS of category 2A patients is due to their need for more extensive investigative workup before disposition can be decided.
Limitations
A truly representative sample would have been obtained by randomly sampling days of the week taken over the entire year. However, analysis for one day at a time would necessarily have had to ignore patients at the end of each 24 hour block whose visit is not completed. This would introduce a significant bias against longer stay patients. A continuous study overcomes this problem but exposes the study sample to bias if unusual health conditions are affecting the population of Barbados at the time such as asthma. We conducted a one day study and combined it with a continuous six day study resulting in some of the disadvantages of both types—for example, the study week, when compared to an average week in 2003, was found to be non‐representative because of an increased number of asthmatic patients visiting A&E. In addition, a number of patients were lost to the study because their visit was not complete at the end of one of the two study periods.
It was not possible to capture all the activity at all stations and, as a result, 24% of cases were lost. Patients who are most likely to be missed are those with extremely short or long stay in the A&E, leading to a non‐representative sample.
The staff were aware of the ongoing study, hence it is likely that there may have been some enhanced performance (Hawthorne effect26).
The actual departure time of the LWT patients could not be captured. The time recorded was the time when the patient was called to come to a cubicle and found to have left A&E. The study reflects the actual time these patients would have waited if they had decided to stay in the department for treatment.
Conclusions
This study has shown that A&E is not meeting recommended benchmarks for timeliness in EDs or even the timeliness of US hospitals where studies have been done. A major portion of the patients' stay in A&E was due to the waiting times at various stages, particularly waiting for laboratory results, waiting to reach a cubicle, waiting in the cubicle to see the EP, and waiting in A&E after admission. All these waiting times have to be improved considerably to enhance the service provided by A&E.
Abbreviations
A&E - accident and emergency
ED - emergency department
LOS - length of stay
LWT - left without treatment
QEH - Queen Elizabeth Hospital
RA - research assistant
TAT - turn around time
Footnotes
Funding: this study was funded by an unrestricted grant from the University of the West Indies, which played no role in the study design, collection, analysis, and interpretation of data or in the writing of the report and decision to submit the paper for publication.
Competing interests: none declared
References
- 1.Sitzia J, Wood N. Patient satisfaction: a review of issues and concepts. Soc Sci Med 1997451829–1843. [DOI] [PubMed] [Google Scholar]
- 2.Schwartz L R, Overton D T. The management of patient complaints and dissatisfaction. Emerg Med Clin North Am 199210557–572. [PubMed] [Google Scholar]
- 3.Rhee K J, Bird J. Perceptions and satisfaction with ED care. J Emerg Med 199614679–683. [DOI] [PubMed] [Google Scholar]
- 4.Bjorvell H, Stieg J. Patients' perceptions of the health care received in an emergency department. Ann Emerg Med 199120734–738. [DOI] [PubMed] [Google Scholar]
- 5.Costello K J. Evaluating consumer perceptions of emergency medical services: an exploratory study. J Ambul Care Mark 19871(2)9–21. [DOI] [PubMed] [Google Scholar]
- 6.Burstein J, Fleisher G R. Complaints and compliments in the pediatric emergency department. Pediatr Emerg Care 19917(3)138–140. [DOI] [PubMed] [Google Scholar]
- 7.Carey R G, Goby M J, Posavac E J.et al Waiting time and patient satisfaction with emergency room care. Emerg Health Serv Q 19821(4)37–40. [DOI] [PubMed] [Google Scholar]
- 8.McMillan J R, Younger M S, Dewine L C. Satisfaction with hospital emergency departments as a function of patient triage. Health Care Manage Rev 198611(3)21–27. [DOI] [PubMed] [Google Scholar]
- 9.Thompson D A, Yarnold P R, Williams D R.et al Effects of actual waiting time, perceived waiting time, information delivery and expressive quality on patient satisfaction in the Emergency department. Ann Emerg Med 199628657–665. [DOI] [PubMed] [Google Scholar]
- 10.Booth A J, Harrison C J, Gardener G J.et al Waiting time and patient satisfaction in accident and emergency department. Arch Emerg Med 19929162–168. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Bursh B, Beezy J, Shaw R. Emergency department satisfaction: What matters most? Ann Emerg Med 199322586–591. [DOI] [PubMed] [Google Scholar]
- 12.Boudreaux E D, Ary R D, Mandry C V.et al Determinants of patient satisfaction in a large municipal Emergency department. Am J Emerg Med 200018394–400. [DOI] [PubMed] [Google Scholar]
- 13.Kapriel M S. Benchmarking facilitates process improvement in the emergency department. Healthc Financ Manage 200054(4)54–59. [PubMed] [Google Scholar]
- 14.Hickson G B, Clayton E W, Entman S S.et al Obstetricians' prior malpractice experience and patients' satisfaction with care. JAMA 19942721583–1587. [PubMed] [Google Scholar]
- 15.Reiber N H. Survey of emergency room usage gives guidelines for improvement. Hosp Topics 19654369–73. [DOI] [PubMed] [Google Scholar]
- 16.Goss M E, Reed J I, Reader G G. Time spent by patients in emergency room: survey at the New York Hospital. N Y State J Med 1971711243–1246. [PubMed] [Google Scholar]
- 17.Cue F, Inglish R. Improving the operations of the emergency department. Hospitals 197852110–119. [PubMed] [Google Scholar]
- 18.Fineberg D A, Stewart M M. Analysis of patient flow in the emergency room. M Sinai J Med 197744551–559. [PubMed] [Google Scholar]
- 19.Fries B E, Gutkin C E, Ginsberg A S. Emergency room utilization: data reconstruction using a deterministic simulation model. Comput Biomed Res 197710153–163. [DOI] [PubMed] [Google Scholar]
- 20.Heckerling P S. Time study of an emergency room. Identification of sources of patient delay. IMJ Ill Med J 1984166(6)437–440. [PubMed] [Google Scholar]
- 21. Data Benchmarks—How does your ED measure up? Cost Reengineering Report Ann Arbor MI: Chi Systems Division, 1997
- 22. Clinical Initiative Centre: The clockwork, ed. Vol I, II and III. Washington DC: The Advisory Board, 1999
- 23.Cruickshank A J. Patient Perception of the services of the accident and emergency department of the Queen Elizabeth hospital, Barbados [Thesis] Barbados: School of Clinical Medicine and Research, University of West Indies 2000
- 24.Wuerz R, Fernandez C M B, Alarcon J. Inconsistency of emergency room triage. Ann Emerg Med 199832431–435. [DOI] [PubMed] [Google Scholar]
- 25.Derlet R W, Nishio D A. Refusing care to patients who present to an emergency department. Ann Emerg Med 199019262–267. [DOI] [PubMed] [Google Scholar]
- 26.Mayo E. The human problems of an industrial civilization. New York: MacMillan, 1933