Twenty-five studies were included in the review. Sample sizes ranged from 50 to 6,790 and included data from more than 35,000 patients. Only one study was an individually randomised controlled trial; two studies randomly allocated shifts to intervention or control. Patients were blinded to the intervention in all except one of the studies. The included studies scored an average of 3.6 (standard deviation 1.5) and ranged from 1 to 7 on the PEDro scale. Half of the included studies were considered poor quality.
Basic triage versus informal or no triage: Seven studies were included and found conflicting results on patient flow. Four were conducted in emergency departments, one in a delivery and labour unit, one in a dental service and one in a community mental health service.
Comparison of basic triage systems: Four studies were included and all found significant reductions in waiting time and/or length of stay. Three studies adapted triage criteria for specific patient groups (chest pain, paediatric and mental health). The other study examined the impact of prioritising treatable cases in a child and adolescent mental health service.
Triage with management options versus prioritisation only: Eight studies were conducted in a hospital emergency department and included a doctor at the triage desk in conjunction with the standard nurse triage role. All studies found some evidence of benefit for reduced length of stay, reduced number of patients who left without being seen and reduced waiting time in the emergency department.
Three studies evaluated use of a multidisciplinary triage clinic. All three found that more than 20% of patients were discharged at triage, reductions in waiting times or reduced number of patients on the waiting list. Two further studies in sexual health clinics also found evidence of improvements in patient flow.