• 
• 

Three Part Question

In [non-trauma adult emergency department patients] does the initial [Modified Early Warning Score] predict [ICU admission in the first 48 hours]?

Search Strategy

EMBASE 1974 to 2016 June 20 using the OVID interface.
MEDLINE 1966 to 2016 June 20 using the PubMed interface.
( (early warning score) OR (NEWS) OR (EWS) OR (MEWS) )
AND
(emergency department)

Limited to patients aged over 17 and to English language.

Search Outcome

Altogether 144 papers were found in EMBASE and 150 in PubMed, of which 14 titles were relevant to the question. After discarding irrelevant abstracts and papers, 4 relevant articles were summarised in the table.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Christensen, et al 2011 Denmark	ED patients: Age 2-98, medical and surgical	2b: Retrospective cohort	ICU admission in first 48 hours.	RR 4.1 (1.5-10.9), PPV 6%, NPV 98% for BEWS >=5.	Sample taken from a streamed cohort ('red'), of which half excluded. BEWS cut-off (5) based on local protocol, with no subset analysis possible.
			Death in first 48 hours.	RR 20.3 (6.9-60.1), PPV 16%, NPV 99% for BEWS >=5
Corfield, et al 2014 UK	2003 ED patients aged >16 with suspected sepsis who were admitted for at least 2 days (or died within 2 days) to 20 Scottish hospitals	2b: Retrospective cohort	ICU admission in first 48 hours	Significantly higher admission NEWS in patients who went to ICU, predicted by NEWS >=7 (27% admitted to ICU)	Data collection stopped early (though large sample completely analysed). Use of poorly-evidence age-adjusted NEWS. Patients discharged <2 days not considered.
			30-day in-hospital mortality	Higher NEWS associated with higher mortality.
Heitz, et al 2010 USA	300 randomly selected adult ED patients, excluding trauma and cardiology	2b: Retrospective cohort	Mortality or admission to intermediate or intensive care within 24 hours.	All MEWS max >9 patients died or were admitted to higher care. Proposed MEWS max cut-off >=4 gave sensitivity 62%, PPV 52%	Excluded cardiological presentations (disease-specific risk tool). Carried-forward rather than excluded absent data. Variable use of MEWS initial/max/admit
Jafar, et al 2016 UK	200 resuscitation room adult patients excluding trauma	1b: Inception cohort study	New organ failure, death or escalation of care within 48hrs.	MEWS predicted 48hr death (OR 1.32) and organ failure (OR 1.19) but not care escalation. All who died had MEWS >= 4	Unclear cut-off MEWS in OR calculation. Does not study patients admitted to ICU directly from ED.

Comment(s)

High Modified Early Warning Scores predict hospital mortality reasonably well but there is no evidence as yet for utility in the Emergency Department in terms of patient prioritisation.

References

1. Christensen D. Jensen N.M. Maaloe R. Rudolph S.S. Belhage B. Perrild H. Nurse-administered early warning score system can be used for emergency department triage. Danish Medical Bulletin 2011; 58 (6): (no pagination)
2. Corfield A.R. Lees F. Zealley I. Houston G. Dickie S. Ward K. McGuffie C. Utility of a single early warning score in patients with sepsis in the emergency department. Emergency Medicine Journal 2014; 31 (6): 482-487
3. Heitz CR, Gaillard JP, Blumstein H, Case D, Messick C, Miller CD. Performance of the maximum modified early warning score to predict the need for higher care utilization among admitted emergency department patients. J Hosp Med. 2010; 5(1): E46-52
4. Jafar A.J.N. Junghans C. Kwok C.S. Hymers C. Monk K.J. Gold E. Harris T.R. Do physiological scoring and a novel point of care metabolic screen predict 48-h outcome in admissions from the emergency department resuscitation area?. European Journal of Emergency Medicine 2016; 23 (2): 130-136