Research article

BMC Emergency Medicine

, 13:17

First online:

Open Access This content is freely available online to anyone, anywhere at any time.

Implementing wait-time reductions under Ontario government benchmarks (Pay-for-Results): a Cluster Randomized Trial of the Effect of a Physician-Nurse Supplementary Triage Assistance team (MDRNSTAT) on emergency department patient wait times

  • Ivy ChengAffiliated withEmergency Services, Sunnybrook Health Sciences CenterKarolinska Institutet Email author 
  • , Jacques LeeAffiliated withClinical Epidemiology Unit, Sunnybrook Health Sciences Center
  • , Nicole MittmannAffiliated withHOPE Research Centre, Sunnybrook Health Sciences Centre
  • , Jeffrey TybergAffiliated withEmergency Services, Sunnybrook Health Sciences Center
  • , Sharon RamagnanoAffiliated withLawrence S. Bloomberg Faculty of Nursing, University of Toronto
  • , Alex KissAffiliated withInstitute of Clinical Evaluative Sciences, Sunnybrook Health Sciences Centre
  • , Michael SchullAffiliated withInstitute of Clinical Evaluative Sciences, Sunnybrook Health Sciences Centre
  • , Fergus KerrAffiliated withDepartment of Emergency Medicine, Austin Health
  • , Merrick ZwarensteinAffiliated withDepartment of Family Medicine - Schulich School of Medicine and Dentistry, Western University



Internationally, emergency departments are struggling with crowding and its associated morbidity, mortality, and decreased patient and health-care worker satisfaction. The objective was to evaluate the addition of a MDRNSTAT (Physician (MD)-Nurse (RN) Supplementary Team At Triage) on emergency department patient flow and quality of care.


Pragmatic cluster randomized trial. From 131 weekday shifts (8:00–14:30) during a 26-week period, we randomized 65 days (3173 visits) to the intervention cluster with a MDRNSTAT presence, and 66 days (3163 visits) to the nurse-only triage control cluster. The primary outcome was emergency department length-of-stay (EDLOS) for patients managed and discharged only by the emergency department. Secondary outcomes included EDLOS for patients initially seen by the emergency department, and subsequently consulted and admitted, patients reaching government-mandated thresholds, time to initial physician assessment, left-without being seen rate, time to investigation, and measurement of harm.


The intervention’s median EDLOS for discharged, non-consulted, high acuity patients was 4:05 [95th% CI: 3:58 to 4:15] versus 4:29 [95th% CI: 4:19–4:38] during comparator shifts. The intervention’s median EDLOS for discharged, non-consulted, low acuity patients was 1:55 [95th% CI: 1:48 to 2:05] versus 2:08 [95th% CI: 2:02–2:14]. The intervention’s median physician initial assessment time was 0:55 [95th% CI: 0:53 to 0:58] versus 1:21 [95th% CI: 1:18 to 1:25]. The intervention’s left-without-being-seen rate was 1.5% versus 2.2% for the control (p = 0.06). The MDRNSTAT subgroup analysis resulted in significant decreases in median EDLOS for discharged, non-consulted high (4:01 [95th% CI: 3:43–4:16]) and low acuity patients (1:10 95th% CI: 0:58–1:19]), as well as physician initial assessment time (0:25 [95th% CI: 0:23–0:26]). No patients returned to the emergency department after being discharged by the MDRNSTAT at triage.


The intervention reduced delays and left-without-being-seen rate without increased return visits or jeopardizing urgent care of severely ill patients.

Trial registration number