Performance of the Interagency Integrated Triage Tool in a resource-constrained emergency department during the COVID-19 pandemic.

BACKGROUND: The Interagency Integrated Triage Tool (IITT) is a three-tier triage instrument recommended by the World Health Organization, but only the pilot version of the tool has been comprehensively assessed for its validity and reliability. This study sought to evaluate the performance of the IITT in a resource-constrained emergency department (ED) during the COVID-19 pandemic. METHODS: This prospective observational study was conducted at ANGAU Memorial Provincial Hospital in Lae, Papua New Guinea. The study period commenced approximately six weeks after introduction of the IITT, coinciding with a major COVID-19 wave. The primary outcome was sensitivity for the detection of time-critical illness, defined by eight pre-specified conditions. Secondary outcomes included the relationship between triage category and disposition. Inter-rater reliability was assessed using Cohen's Kappa. RESULTS: There were 759 eligible presentations during the study period. Thirty patients (4.0%) were diagnosed with one of the eight pre-specified time-critical conditions and 21 were categorised as red or yellow, equating to a sensitivity of 70.0% (95%CI 50.6-85.3). There was a clear association between triage category and disposition, with 22 of 53 red patients (41.5%), 72 of 260 yellow patients (27.7%) and 22 of 452 green patients (4.9%) admitted (p = <0.01). Negative predictive values for admission and death were 95.1% (95%CI 92.7-96.9) and 99.3% (95%CI 98.1-99.9) respectively. Among a sample of 106 patients, inter-rater reliability was excellent (κ = 0.83) and the median triage assessment time was 94 seconds [IQR 57-160]. CONCLUSION: In this single-centre study, the IITT's sensitivity for the detection of time-critical illness was comparable to previous evaluations of the tool and within the performance range reported for other triage instruments. There was a clear relationship between triage category and disposition, suggesting the tool can predict ED outcomes. Health service pressures related to COVID-19 may have influenced the findings.