Objectives & Background Despite the increasing burden of trauma and acute illness in low and middle-income countries, emergency care in these countries remains immature and there is a lack of robust data regarding mortality. Emergency care at the main governmental referral hospital in Freetown, Sierra Leone has been redeveloped in the wake of the Ebola epidemic. Initiatives include twenty-four hour triage, a resuscitation bay and an acute medical unit. Mortality within the unit is noted down for emergency admissions, but to date, the information has not been collated or reviewed. This study aimed to determine the baseline mortality rate within the acute medical unit and to assess correlation with triage category on arrival.

Methods The aims of this study were achieved by carrying out a retrospective review of emergency admissions at a tertiary referral hospital with 300 beds. The number of new admissions, deaths within the acute medical unit and the admission triage category of deceased patients were obtained from ledgers for a six-month period between January and June 2016. The triage categories are based on the South African Triage Scale (SATS) and include, in an order of increasing severity, green, yellow, orange and red. Patients that arrived dead to the department were excluded from the study. Pearson's correlation was used to test for an association between triage category and mortality.

Results A total of 1657 acute medical unit admissions were recorded during the study period. The overall mortality was 10.5% (n=174). There was a strong positive correlation between triage category on arrival and mortality, which was statistically significant (r=.986, n=145, p=0.014). Only one deceased patient was assigned to the green category, which on review was found to belong to the orange category. Triage category was not recorded for 29 deceased patients (16.7%).

Conclusion The study represents the first effort in the acute medical unit to review emergency patient mortality and serves as a baseline for future data collection and comparison. This basic method of mortality data collection could be adopted in other units and local healthcare facilities in order to produce comparable data, which does not currently exist. The strong correlation between the assigned triage category and mortality highlights the value of triage in identifying the patients most at risk of dying.