The review concluded patient flow in emergency departments could be improved with the introduction of fast track processes for patients with less severe symptoms. Team triage, with a physician in the team, would probably result in improvements. The authors conclusions reflect the evidence presented but potential limitations in the review process mean the reliability of the conclusions is uncertain.

To determine if triage-related interventions improve patient flow in emergency departments.

British Nursing Index, Business Source Premier, CINAHL, The Cochrane Library, EMBASE, ProQuest ABI, PubMed and Science Direct were searched from 1966 to March 2009. Reference lists and review articles were scanned for additional papers. Search terms were reported in an online appendix. Further details were reported in the full Health Technology Assessment, see publications of related interest.

Randomised controlled trials (RCTs) or observational studies with historical controls that evaluated triage-related interventions to improve patient flow in emergency departments were eligible for inclusion. Studies had to include patients (aged 15 years or over) who presented for somatic reasons. Studies also had to report data on waiting time, length of stay, patients leaving the emergency department without being seen or other flow patterns.

Triage related interventions evaluated in the included studies were: streaming which used routines to divide patients based on various criteria after brief evaluation; fast track using coherent processes; team triage where patients were handled by a team including a physician; point of care testing which moved laboratory analysis to the emergency department and nurse requested x-ray.

Two reviewers independently selected studies for inclusion.

Study quality was assessed using methods of Eldridge et al. and Akobeng. Methodological quality and the clinical relevance of each study was graded as high, medium or low. In addition the GRADE system was used to appraise the overall strength of evidence based on quality and relevance. Factors considered included study quality, concordance/consistency, transferability/relevance, precision of data, risk of publication bias, effect size and dose-response. Studies were graded as having strong, moderately strong, limited or insufficient scientific evidence.

The authors did not state how many reviewers assessed study quality.

Studies were grouped by intervention and data on relevant outcomes were extracted to calculate median (minimum and maximum) differences between intervention and control groups.

The authors did not state how many reviewers extracted data.

Data were combined in a narrative summary with additional data reported in tables. Studies of low quality and relevance were included when studies of medium quality and relevance were unavailable.

Thirty-three studies (over 800,000 participants) were included in the review although only details of 31 studies were reported; nine RCTs and twenty two controlled observational studies. No studies were graded as high quality. Twenty-two studies were graded as medium quality and eleven studies as low quality.

There was moderately strong evidence of the effect of fast track interventions on waiting time (median 24.5 minutes, nine studies including one RCT), length of stay (median 27 minutes, 10 studies including two RCTs) and number of patients leaving the emergency department without being seen (median 3.1%, five studies), but there was insufficient evidence for increased patient satisfaction (two studies, including one RCT).

There was limited evidence of the effect of streaming (excluding fast-track) on waiting time (median 31 minutes, three studies) and length of stay (median 9.5 minutes, two studies).

There was a moderately strong effect of team triage on patients leaving without being seen (median 1.3%, four studies, including one RCT), but limited evidence for waiting time (median 18 minutes, three studies), and length of stay (median 40.5 minutes, four studies including two RCTs).

There was moderately strong evidence of the effect of point of care testing for response time (median 51 minutes, three studies), but limited evidence for length of stay (median 21 minutes, five studies including two RCTs).

There was limited evidence of the effect of nurse-requested x-ray (three RCTs).

Introducing fast track for patients with less severe symptoms resulted in shorter waiting time, shorter length of stay and fewer patients leaving without being seen. Team triage, with a physician in the team, would probably result in shorter waiting time and shorter length of stay and most likely in fewer patients leaving without being seen. There was limited scientific evidence that streaming of patients into different tracks, performing laboratory analysis in the emergency department or having nurses to request certain x-rays resulted in shorter waiting time or length of stay.

The review question was clear with defined inclusion criteria. Several relevant sources were searched. No attempts were made to locate unpublished data and it was unclear whether language restrictions were applied so the potential for language and publication bias could not be ruled out. Study quality was assessed and some results were reported. Appropriate methods to reduce reviewer error and bias were used for selecting studies but it was unclear whether similar methods were used for quality assessment and data extraction.

None of the included studies were high quality and most were observational studies. The authors report including thirty three studies in the review but details and results were only reported for 31 studies. Studies were combined in a narrative synthesis which was appropriate given the variability of the studies in terms of study design and quality, interventions and outcomes. The authors conclusions reflected the evidence presented but some potential methodological limitations in the review process mean the reliability of the conclusions is uncertain.

Note: This abstract was based on the journal article and English summary from a full report which was only available in Swedish.

Practice: The authors stated that introducing fast track for patients with less severe symptoms results in shorter waiting time, length of stay and fewer patients leaving without being seen. Team triage with a physician in the team would probably result in shorter waiting time, length of stay and most likely in fewer patients leaving without being seen.

Research: The authors stated that further research was needed into the economics and patient satisfaction of fast track interventions in the emergency department. Further research was also required to fully evaluate the effects of organisational interventions.

Statens beredning for medicinsk utvardering (SBU) (Swedish Council for Health Technology Assessments).

Statens beredning for medicinsk utvardering (SBU): Triage och flodesprocesser pa akutmottagningen. En systematisk litteraturoversikt Stockholm: 2010, 197.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.