Article, Critical Care

Admissions of critically ill patients to the ED intensive care unit

a b s t r a c t

Introduction: Many emergency departments (EDs) have established units capable of providing critical care because of increasing need for critical care, called as ED intensive care unit (EDICU). However, prolonged critical care leads to crowding, resulting in poor quality of care and high mortality rates. We aimed to determine which type of critically ill patients play a main role for crowding in the EDICU, and how to manage these patients. Method: Patients aged older than 18 years who presented to the ED and presented for consultation to the ICU were eligible for inclusion in this study. Patients were classified into 4 priority groups by the Society of Critical Care Medicine.

Result: Four hundred Medical patients were enrolled in the study. Sixty-one patients were not admitted to hospital (15.2% of all patients) and were treated in the EDICU. These patients were older (mean age, 66.6 years) and had a higher percentage belonging to the priority 3 group (82.0%–unstable with reduced likelihood of recovery due to chronic illness) in comparison with other ICUs patients (mean age, 60.4 years and 11.9%, respectively) (P b .05). In priority 3 patients, the length of stay was median 120 hours, and also, length of invasive mechanical ventilations duration was median 19 hours in the EDICU.

Conclusions: Emergency department intensive care unit occupancy appears driven by categorized as “reduced benefit” patients, and these units tend to become alternative dumping grounds for palliative care services. Hospitals and health care administrators should take special care to develop policies for improving the management of these patients.

(C) 2014


Critical care has an increasing importance in emergency medicine. In the United States, the annual visits to the emergency department (ED) of critically ill patients between 2001 and 2009 increased by 79%, and the length of stay (LOS) in the ED increased by 32% for these patients during the same period [1]. Similar results that increased annual visit combination with Prolonged LOS were found in other studies [2-7].

Nowadays, many EDs are in change to have capable of units, which provide critical care, called as ED intensive care unit (EDICU) because of the increasing need for critical care. After providing resuscitation and stabilization, critical care of the patients is continued in the EDICU, during the patient’s ED stay [8]. Even these units tend to become alternative dumping grounds, when critically ill patients cannot be admitted to hospital, especially patients presenting emergently have poor survival. Such circumstances can cause crowding and poor health outcomes for all critically ill patients [3,4,7].

Determining priorities of the critically ill patients admitting to the EDICU (and all EDs) is important. If these patients are properly managed and admitted to their appropriate critical care departments (such as intensive care, acute care, or palliative care units), crowding and LOS

* Corresponding author. Emergency Medicine, Hacettepe University Faculty of Medicine, Sihhiye/Ankara/Turkey.

E-mail address: [email protected] (M.A. Aslaner).

reduce significantly. We aimed to answer 2 questions in this study. Firstly, determining which type of critically ill patients play a main role for crowding in the EDICU. And secondly, how to manage these patients to reduce crowding.


This report was based on a prospective observational study conducted at Hacettepe University Faculty of Medicine in Turkey from April 15 to September 14, 2013. The hospital’s ethics committee approved the study, and the need for informed consent was waived. Patients aged older than 18 years who presented to the ED and presented for consultation to the ICU were eligible for inclusion in this study.

There were 7 medical and surgical ICUs, palliative care wards (non- ICU), an acute care unit, and a busy oncology center at the hospital. Treatment of the patients who were not admitted to the hospital was continued at third level ICU of the ED. The terminology EDICU was cre- ated to describe “a unit within an ED with the same or similar staffing, monitoring, and capability for therapies as an ICU [8].” An emergency medicine specialist, a senior resident, and 2 nurses were responsible for the patients at the EDICU with 8 beds. Critical care was provided for nonadmitted patients at the EDICU (such as invasive and noninva- sive mechanical ventilation (non-IMV), invasive monitoring, use of va- soactive drugs, weaning, hemodialysis, paracentesis, and thoracentesis).

0735-6757/(C) 2014

Patients with unknown results and patients who died within 4 hours after ED presentation were both excluded from the study.

Patients were divided into 4 priority groups by Society of Critical Care Medicine (SCCM) criteria [9]. “Priority” 1 group included “unsta- ble” patients requiring ICU monitoring and treatment (or those who would most benefit from the ICU), “priority” 2 group included patients requiring intensive monitoring with possible need for intervention, “priority” 3 group included “unstable” patients with slim chances of re- covery due to chronic illness (such as metastatic neoplasm or bedridden patients), and “priority” 4 group included patients inappropriate for ICU placement (“too sick” or “too well to benefit” from ICU care).

Acute Physiology and Chronic Health Evaluation II (APACHE II), Rapid Emergency Medicine Score (REMS), modified early warning score , and Sequential Organ Failure Assessment Score (SOFA) were calculated for all patients’ vitals, laboratory values, age, and comorbidities. Scores were derived from the first values obtained at the ED. Demograph- ic characteristics, time of presentation to ED, LOS, use of IMVs, diagnosis, cost, admission, discharge, and death status were all recorded.

A triage system developed by the Elderly in European Intensive Care Units (ELDICUS) project was used for admission to the ICU. “Initial refusal score (ELDICUS I),” consisting of 15 variables and “final triage score,” consisting of 11 variables were in addition applied to patients. Patients with initial refusal scores more than 173.5 or with final triage scores = 0 were refused ICU admission due to being “too sick or too well for ICU” (99.5%-99.3% specificity). Admission to ICU was appropriate if scores were between or included values of 1 and 173.5 [10].

The grouping of patients and the organization of follow-up data were conducted by the senior resident under the guidance of the emergency med- icine specialist. Data were continuously reviewed to minimize redundancies.

Statistical analysis

Statistical analyses were performed using IBM SPSS for Windows version 21.0 statistical package. Continuous variables were presented

as either mean values +- SD or median values. Categorical variables were summarized as frequencies and percentages. Normality of the continuous variables was evaluated using the Shapiro-Wilk test. The differences between the 2 groups according to continuous variables were determined either by the independent samples t test or by the Mann-Whitney U test. Comparisons among more than 2 groups were conducted by 1-way analysis of variance or Kruskal-Wallis tests. Categorical variables were compared using Pearson ?2 or Fisher exact test. Correlation between mortality and scores was determined by the ?2 coefficient (with significance defined as P b .05).


Between April 15 and September 14, 2013, a total of 13 963 pa- tients presented to the ED; 400 medical patients (excluding trauma patients) met the inclusion criteria and were enrolled in the study. Priority patients’ baseline characteristics, comorbidities, and diagno- ses are shown in Table 1. The SCCM priority models were grouped as “priority” 1 patients (n = 102 or 25.5% of all patients), “priority” 2 patients (n = 148 or 37% of all patients), “priority” 3 patients (n = 118 or 29.5% of all patients), and “priority” 4 patients (n = 32 or 8% of all patients).

In the group of “priority” 1 and 2 patients, ELDICUS I, APACHE II, SOFA, and REMS were significant scores (P b .05). Scores predicting death were found for “priority” 1 patients using ?2 as ELDICUS I N APACHE II N SOFA N REMS. In the “priority” 3 group, SOFA was the only significant score apart from the morbidity score. The MEWS score was not significant in any group (Table 1).

After consultation, 269 patients (67.2% of all patients) were admit- ted to medical-surgical (med-surg) ICUs. Seventy patients (17.6% of all patients) were admitted to non-ICU wards after initial treatment pro- vided, and 61 patients (15.2% of all patients) were treated in the EDICU because they were not admitted to hospital.

Table 1

Characteristics of priority groups consulted for admission to med-surg ICUs from the ED


Priority 1 (n = 102)

Priority 2 (n = 148)

Priority 3 (n = 118)

Priority 4 (n = 32)


Age, mean +- SD

60.0 +- 16.0

62.5 +- 15.6

67.1 +- 14.2

52.4 +- 20.0


Sex, male 62 (60.8%)

92 (62.2%)

73 (61.9%)

17 (54.8%)




49 (48.0%)

69 (46.6%)

40 (33.9%)

12 (38.7%)


History of malignancy

23 (22.5%)

23 (15.5%)

87 (73.7%)

4 (12.9%)


Coronary artery disease

28 (27.5%)

60 (40.5%)

25 (21.2%)

3 (9.7%)


Diabetes mellitus

18 (17.6%)

38 (25.7%)

25 (21.2%)

3 (9.7%)


Heart failure/arrhythmia

18 (17.6%)

41 (27.7%)

22 (18.6%)

2 (6.5%)



9 (8.8%)

22 (14.8%)

13 (11.0%)

4 (12.5%)


Unit/ward admission

Med-surg ICUs

83 (81.4%)

122 (82.4%)

32 (27.1%)

32 (100%)



9 (8.8%)

2 (1.4%)

50 (42.4%)

0 (0.0%)


Acute care

4 (3.9%)

14 (9.5%)

4 (3.4%)

0 (0.0%)


Palliative care/wards

6 (5.9%)

10 (6.8%)

32 (27.1%)

0 (0.0%)


Diagnosis Pulmonary

31 (30.4%)

20 (13.5%)

63 (53.4%)

3 (9.2%)



31 (30.4%)

58 (39.2%)

8 (6.8%)

2 (6.0%)



12 (11.7%)

31 (20.9%)

29 (24.5%)

11 (35.2%)



16 (15.7%)

11 (7.4%)

66 (55.9%)

0 (0.0%)



27 (25.5%)

33 (32.3%)

12 (10.2%)

8 (25.9%)


Endocrine metabolic

1 (1.0%)

5 (3.4%)

3 (2.5%)

2 (6.0%)


Drug overdose

6 (5.9%)

4 (2.7%)

1 (0.8%)

3 (9.2%)


Mortality med-surg ICUs

Scores predicting death (?2b/median value)

17 (20.5%)

15 (12.3%)

22 (68.8%)

1 (3.1%)



0.186 (39.0)

0.058 (33.0)




0.118 (12.0)

0.122 (10.0)




0.115 (3.0)

0.183 (2.0)




0.050 (6.0)

0.023 (5.0)








a COPD, chronic obstructive pulmonary disease.

b Bigger is better.

Emergency department intensive care unit patients

The treatments of 61 patients were not admitted to hospital and were continued in the EDICU (Table 2). These patients were older (mean age, 66.6 years) and had a higher percentage belonging to the “priority” 3 group (50 patients or 82.0%) in comparison with med- surg ICU patients (mean age, 60.4 years and 32 patients or 11.9%, re- spectively) (P b .05). Length of stay of “priority” 3 patients in the EDICU was median 120 hours (4-1176) that was also the same for those of all EDICU patients. Although IMVs were used for 79% of EDICU patients (48 patients), 81% of these patients (39 patients) were in the “priority” 3 group. Furthermore, length of IMV duration was me- dian 19 hours (0.1-336) in “priority” 3 patients.

Nearly 50% of all EDICU patients were diagnosed with Pulmonary system diseases and sepsis; the prevalence of these particular diagnoses was higher in the EDICU than in the other ICUs (P b .001). Patients ad- mitted to the EDICU had a 75.4% mortality rate, and 46 patients had died during follow-up. Patients admitted to med-surg ICUs, however, had a lower mortality rate of 20.4% in the all ICU admissions (P b .001). From April 15 to September 14, 2013, the occupancy rate for the EDICU was 94.0%; rates for medical ICUs were between 86.0% and 94.7% and those for surgical ICUs were between 51.0% and 78.3%.

Priorities of patients

Rates of ED visits in “priority” 1 and 2 groups were 25.5% and 37%, re- spectively, and rates of admission to med-surg ICUs were more than 80% for both. admission diagnosis of these patients was mostly cardiovascu- lar, neurologic, and less pulmonary system diseases.

In total, 118 patients (29.5% of total patients) presenting to the ED were triaged as “priority” 3 patients (unstable with reduced likelihood of recovery due to chronic illness). Also the history of malignancy and the ages of these patients were greater than these values in other prior- ity models (P b .001).

In the group of “priority” 3 patients, EDICU admissions were higher than admissions to other units (42.4% of total admissions, P b .001); per- centage of med-surg ICU admissions and palliative care services admis- sions were equal in this group (27.1% of total admissions).

Two groups of unstable patients were evaluated: among “priority” 1 patients, those who were not admitted to med-surg ICUs had higher mortality rates than those of admitted patients (52.6% vs 20.5%, P = .01); among “priority” 3 patients, there was no significant differ- ence between mortality rates of nonadmitted (74.4%) and admitted (68.8%) patients (Fig. 1). When “priority” 3 patients were admitted to med-surg ICUs, hospital LOS increased by 83%, and costs increased by 135% compared with those values of their nonadmitted counterparts (P b .05) (Fig. 2).

Table 2

Patient characteristics in the EDICU and in med-surg ICUs

Variables EDICU (n = 61) ICUs (med-surg) (n = 269) P

Age, mean +- SD

66.6 +- 14.9

60.4 +- 16.7


Priority models

Priority 1

9 (14.8%)

83 (30.9%)


Priority 2

2 (3.3%)

122 (45.4%)


Priority 3

50 (82.0%)

32 (11.9%)


Priority 4

0 (0.0%)

32 (11.9%)




14 (23.0%)

79 (29.4%)



31 (50.8%)

39 (14.5%)



30 (49.2%)

16 (5.9%)



4 (6.6%)

58 (21.6%)



8 (13.2%)

10 (3.7%)


Endocrine metabolic

1 (1.6%)

4 (1.5%)


Drug overdose

1 (1.6%)

9 (3.3%)



46 (75.4%)

55 (20.4%)


Fig. 1. “Priorities” 1 and 3 patient mortality rates: med-surg ICU admission vs nonadmission.


Although urgencies of critically ill patients are the same, their prior- ities can be different. Similarly, although some patients receive further benefit from critical care, the others receive limited [11]. Even many critically ill patients presenting emergently have poor survival. There- fore, the areas where the patients receiving critical care are different from each other such as intensive care, acute care, or palliative care units.

Similar to findings in previous studies, this study showed that criti- cally ill patients with older ages had higher Mortality scores, especially among “priority” 3 groups [1,12,13]. Although mortality scores did tend to increase with older age and with comorbidities, not all scores were found to be significant; only SOFA scores were significant among “priority” 3 patients. ELDICUS project, a new score to ICU triage, can be useful for predicting death in “priority” 1 and 2 patients.

Emergency department intensive care unit patients

Critical care is not only provided in the ICU; care starts amidst prehospital conditions and continues through the ED [14,15]. Tradition- ally, resuscitation and stabilization of critically ill patients are provided in the ED. However, many EDs are in change to have capable of units, which provide critical care, called as EDICU because of increasing need for critical care [8]. From 2001 to 2008, ED treatments occurring in ob- servation units increased from 642 000 to 2318 000 due to the de- creased number of inpatient Hospital beds and critical care services [16]. When there are insufficient critical care beds (particularly in the ICU), critical care areas in the ED become release valves for hospitals. The aim of EDICU is to continue critical care after resuscitation and sta- bilization of the critically ill patients until admission to the hospital. If these patients are not properly managed or admitted to the appropriate critical care area, resuscitation and the continuing critical care retire from the aims of the EDICU.

Between 2001 and 2009, median LOS of critically ill patients in- creased by 32% in US EDs [1]. In our study, LOS of “priority” 3 patients in the EDICU was median 120 hours, and the longest LOS in the EDICU was 49 days. Several other studies in the literature indicate similar find- ings of LOS lasting for hours or even days [8,17,18]. Emergency depart- ment crowding rapidly exceeds the critical level because of the prolonged LOS and increased workload. In such instances, Optimal care for critically ill patients is not available; consequences include re- duced quality of care, increased adverse effects, and increased mortality in the ED [1,3,5,6,19].

In this study, diagnoses of pulmonary system diseases and sepsis were found in more than half of the “priority” 3 patients. The use of IMV was found to be prevalent in 81% of “priority” 3 patients, and length of IMV duration was also median 19 hours for these patients in the EDICU. Green and MacIntyre [4] showed that the rates of IMV were

P = .004

P < .001

Fig. 2. “Priority” 3 patients’ hospital LOS and costs: med-surg ICU admission vs nonadmission.

between 60% and 70% for all patients in the ED. Rose et al [17] found that ED LOS of patients requiring mechanical ventilation exceeded 24 hours. A recent study by Pitts et al [20] (2012) showed that increased ED crowding is strongly associated with both increased intensity of ED care and prolongED boarding times of admitted patients.

From April 15 to September 14, 2013, the occupancy rate of EDICU was 94.0%. Mortality rate was also 75.4% for these patients who tended to be older adults with malignancy and/or organ dysfunctions and belonged to “priority” 3 group by 82%. Over the last 2 decades, older pa- tients with serious and complex comorbidities have continued to increase in the ED [21]. When these patients with end-stage diseases have present- ed to the ED, they have usually received acutely life-sustaining treat- ments. Hospitalization becomes crucial for these “reduced benefit” patients, and appropriate placement can be difficult, leading to prolongED wait times and intensive treatments in the ED. Consequently, prolonged critical and/or end-of-life care leads to ED crowding, resulting in poor quality of critical care and high mortality rates [21].

Priorities of patients

The percentage of “priority” 3 patients in this study was higher than the percentage in the other 2 priority studies conducted by ICU physi- cians [12,13]. Our hospital is a reference oncology center, where unsta- ble patients with malignancy first present to the ED. When there are no critical care beds available, treatment of patients is continued in the EDICU. Of our “priority” 3 patients, 27.1% were admitted into the med- surg ICU; this percentage was similar to that of “priority” 3 patients ad- mitted to the palliative care services. And 42.4% of “priority” 3 patients were followed up in the EDICU. Many previous studies have reported that ICU admission refusal is a common occurrence, with approximately 70% of such cases occurring due to resource constraints [12,22-24].

Mortality rates were up to 70% in “priority” 3 patients admitted to the ICU (as in cases of malignancy and/or organ dysfunctions). Caldeira et al [13] showed that mortality rates were more than 80% in “reduced benefit” patients; other studies have reported mortality rates between 60% and 90% in such patients [23-26]. In distinction from the study of Shum et al [12], our results showed that whether “priority” 3 patients were admitted to the ICU, there were no significant differences in mor- tality rates among these patients. In addition, hospital LOS and costs in- creased in these patients based on their admission or nonadmission to the ICU. Initial studies had reported that patients with malignancy only poorly benefit from ICU care. However, more recent studies have suggested that patients with certain subgroup malignancies or with newly diagnosed malignancies may benefit from critical care. Thus, ICU admission policies should be continuously developed for patients with malignancy [27,28]. Admission policies should include early ICU admission; ICU admission for non-ICU care; and admission to prophy- lactic, palliative, and terminal care units [27].

Presently, palliative care teams are frequently inadequate to coordi- nate ED-based consultations; crowding and substandard critical and/or end-of-life care in EDs can be improved with the development of poli- cies allowing for ED-based palliative care consultations, homecare, and hospice services [21,29,30]. Palliative care providers should work in conjunction with the EDICUs and the EDs to provide higher quality crit- ical care services.


There is an urgent need to develop protocols to best manage this challenging patient population. Possibilities include the following:

      • Rapidly reducing population needing palliative care who are waiting for hospitalization;
      • Increasing development of palliative care units and non-ICU floors to manage unstable and/or Mechanically ventilated patients with reduced benefit;
      • Providing coordination of critical care between the ED, intensive care, and palliative care;
      • Management of the patients with serious and complex comorbidi- ties, end-stage diseases, and malignancies by policies and coordi- nation of critical care units; and
      • Providing homecare and hospice services for these patients to re- duce ED visits.


Emergency department intensive care unit occupancy appears driven by SCCM “priority” 3 patients. When these patients cannot be admitted to appropriate critical care areas, EDs tend to become alterna- tive dumping grounds for palliative care services. Such circumstances can cause crowding and poor health outcomes for all critically ill patients. Hospitals and health care administrators should take special care to develop policies for improving the management of these “reduced benefit” patients.


The authors thank Miss Pelin Aytekin; Sevilay Karahan, PhD; and Serkan E. Eroglu, MD for the generous support.


  1. Herring AA, Ginde AA, Fahimi J, Alter HJ, Maselli JH, Espinola JA, et al. Increasing critical care admissions from U.S. emergency departments, 2001-2009. Crit Care Med 2013;41(5):1197-204.
  2. Hospital and emergency department overcrowding. Ann Emerg Med 1990;19(3): 336.
  3. Cowan RM, Trzeciak S. Clinical review: emergency department overcrowding and the potential impact on the critically ill. Crit Care 2005;9(3):291-5.
  4. Green RS, MacIntyre JK. Critical care in the emergency department: an assessment of the length of stay and invasive procedures performed on critically ill ED patients. Scand J Trauma Resusc Emerg Med 2009;17:47.
  5. Hoot NR, Aronsky D. Systematic review of emergency department crowding: causes, effects, and solutions. Ann Emerg Med 2008;52(2):126-36.
  6. Chalfin DB, Trzeciak S, Likourezos A, Baumann BM, Dellinger RP, group D-Es. Impact of Delayed transfer of critically ill patients from the emergency department to the intensive care unit. Crit Care Med 2007;35(6):1477-83.
  7. Fromm Jr RE, Gibbs LR, McCallum WG, Niziol C, Babcock JC, Gueler AC, et al. Critical care in the emergency department: a time-based study. Crit Care Med 1993;21(7): 970-6.
  8. Weingart SD, Sherwin RL, Emlet LL, Tawil I, Mayglothling J, Rittenberger JC. ED intensivists and ED intensive care units. Am J Emerg Med 2013;31(3):617-20.
  9. Guidelines for intensive care unit admission, discharge, and triage. Task Force of the American College of Critical Care Medicine, Society of Critical Care Medicine. Crit Care Med 1999;27(3):633-8.
  10. Sprung CL, Baras M, Iapichino G, Kesecioglu J, Lippert A, Hargreaves C, et al. The Eldicus prospective, observational study of triage decision making in European intensive care units: part I–European intensive care admission Triage Scores. Crit Care Med 2012;40(1):125-31.
  11. Sprung CL, Danis M, Iapichino G, Artigas A, Kesecioglu J, Moreno R, et al. Triage of intensive care patients: identifying agreement and controversy. Intensive Care Med 2013;39(11):1916-24.
  12. Shum HP, Chan KC, Lau CW, Leung AK, Chan KW, Yan WW. Triage decisions and outcomes for patients with triage priority 3 on the Society of Critical Care Medicine scale. Crit Care Resusc 2010;12(1):42-9.
  13. Caldeira VM, Silva Junior JM, Oliveira AM, Rezende S, Araujo LA, Santana MR, et al. Criteria for patient admission to an intensive care unit and related mortality rates. Rev Assoc Med Bras 2010;56(5):528-34.
  14. Nguyen HB, Rivers EP, Havstad S, Knoblich B, Ressler JA, Muzzin AM, et al. Critical care in the emergency department: a physiologic assessment and outcome evaluation. Acad Emerg Med 2000;7(12):1354-61.
  15. Rivers E, Nguyen B, Havstad S, Ressler J, Muzzin A, Knoblich B, et al. Early goal- directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med 2001;345(19):1368-77.
  16. Venkatesh AK, Geisler BP, Gibson Chambers JJ, Baugh CW, Bohan JS, Schuur JD. Use of observation care in US emergency departments, 2001 to 2008. PLoS One 2011;6(9): e24326.
  17. Rose L, Gray S, Burns K, Atzema C, Kiss A, Worster A, et al. Emergency department length of stay for patients requiring mechanical ventilation: a prospective observa- tional study. Scand J Trauma Resusc Emerg Med 2012;20:30.
  18. Mullins PM, Goyal M, Pines JM. National growth in intensive care unit admissions from emergency departments in the United States from 2002 to 2009. Acad Emerg Med 2013;20(5):479-86.
  19. Hospital-Based Emergency Care. At the breaking point. The National Academies Press; 2007.
  20. Pitts SR, Pines JM, Handrigan MT, Kellermann AL. National trends in emergency department occupancy, 2001 to 2008: effect of inpatient admissions versus emergency department practice intensity. Ann Emerg Med 2012;60(6):679-686.e3.
  21. Grudzen CR, Richardson LD, Morrison M, Cho E, Morrison RS. Palliative care needs of seriously ill, older adults presenting to the emergency department. Acad Emerg Med 2010;17(11):1253-7.
  22. Azoulay E, Pochard F, Chevret S, Vinsonneau C, Garrouste M, Cohen Y, et al. Compliance with triage to intensive care recommendations. Crit Care Med 2001; 29(11):2132-6.
  23. Garrouste-Orgeas M, Montuclard L, Timsit JF, Misset B, Christias M, Carlet J. Triaging patients to the ICU: a pilot study of factors influencing admission decisions and patient outcomes. Intensive Care Med 2003;29(5):774-81.
  24. Garrouste-Orgeas M, Montuclard L, Timsit JF, Reignier J, Desmettre T, Karoubi P, et al. Predictors of intensive care unit refusal in French intensive care units: a multiple- center study. Crit Care Med 2005;33(4):750-5.
  25. Joynt GM, Gomersall CD, Tan P, Lee A, Cheng CA, Wong EL. Prospective evaluation of patients refused admission to an intensive care unit: triage, futility and outcome. Intensive Care Med 2001;27(9):1459-65.
  26. Sprung CL, Geber D, Eidelman LA, Baras M, Pizov R, Nimrod A, et al. Evaluation of triage decisions for intensive care admission. Crit Care Med 1999;27(6):1073-9.
  27. Azoulay E, Soares M, Darmon M, Benoit D, Pastores S, Afessa B. Intensive care of the cancer patient: recent achievements and remaining challenges. Ann Intensive Care 2011;1(1):5.
  28. Lecuyer L, Chevret S, Thiery G, Darmon M, Schlemmer B, Azoulay E. The ICU trial: a new admission policy for cancer patients requiring mechanical ventilation. Crit Care Med 2007;35(3):808-14.
  29. Mahony SO, Blank A, Simpson J, Persaud J, Huvane B, McAllen S, et al. Preliminary report of a palliative care and Case management project in an emergency department for chronically ill elderly patients. J Urban Health 2008;85(3):443-51.
  30. Beynon T, Gomes B, Murtagh FE, Glucksman E, Parfitt A, Burman R, et al. How common are palliative care needs among older people who die in the emergency department? Emerg Med J 2011;28(6):491-5.