Inter-rater agreement of CTAS between paramedics and emergency department nurses

Introduction In Canada, paramedics and hospital emergency department (ED) nurses utilize the Canadian Triage and Acuity Scale (CTAS) to assign acuity. This study examined concordance of assessment between paramedics and emergency department triage nurses. We also assessed operational factors to provide confidence that paramedic triage scores were not being influenced by non-patient-related factors such as workload, delays in handover or destination type. Methods We conducted a retrospective cohort study of all paramedic-transported patients to hospitals from 1 January 2018 to 31 December 2022. All records of patients transported by paramedics to in-region EDs or urgent care centres (UCCs) where CTAS was recorded at two different points by paramedics (at time of transport) and triage nurses (at time of ED triage) were reviewed. We analyzed concordance using crude percentages, inter-rater agreement using a weighted kappa ( kW ) analysis, and computed an adjusted multilevel binomial regression model to determine independent associations between agreement and operational and clinical factors using odds ratios (OR) with 95% confidence intervals (CIs). Results Inter-rater reliability analysis showed ‘substantial’ agreement for the most critical patients (CTAS 1: 78.8%, κW = 0.66) and ‘fair’ agreement for the least urgent patients (CTAS 4: 32.8%, κW = 0.30; CTAS 5: 25.9%, κW = 0.34). Discordance increased with patient complexity (i.e. higher numbers of medical conditions [OR 0.98, 95% CI 0.97–0.98], patients aged ≥18, primary problems of mental health or soft tissue injury/pain presentations). Concordance increased with patient age and paramedic certification level. No operational factors examined demonstrated an independent association with CTAS agreement. Conclusions Patient acuity agreement declined with lower acuity scores. Operational influences were not detected as significant predictors of CTAS disagreement. Several clinical factors were found to potentially affect agreement, particularly with increasing patient complexity. The implications for paramedic practice, paramedic education and EMS response priority planning warrant further investigation.