Impact of Pain Assessment on Canadian Triage and Acuity Scale Prediction of Patient Outcomes

Study objective: How does the removal of patient-reported pain from the Canadian Triage Acuity Scale (CTAS) affect the scale's ability to predict admission, ICU consultation, and mortality? Methods: Retrospective observational cohort study of all adult visits to a tertiary emergency department. The standard CTAS algorithm combined patient-reported pain levels with other data to generate a triage score for each visit. We calculated a "pain-free" CTAS for each visit in the cohort, assuming that the patient had not reported any pain. We fit logistic regression models for each outcome (admission, ICU consultation, and mortality) using either the standard or the pain-free CTAS as the predictor. We compared the area under the receiver operator characteristic curves of the standard versus pain-free CTAS models for each outcome. Results: We analyzed a sample of 229,744 patients. The average reported pain level was 5.6/10 (SD, 3.0) among the 60.1% of the cohort who reported pain. Higher pain was slightly negatively correlated with hospital admission, ICU consultation, and 72-hour mortality (r = -0.008, -0.009, and -0.006, respectively). The area under the curve of the pain-free CTAS was higher than that of the standard scores for hospital admission (0.691 versus 0.641), ICU consultation (0.829 versus 0.773), and mortality (0.863 versus 0.810). Differences were statistically but not clinically significant. Conclusion: The removal of the pain scale from CTAS did not reduce its ability to predict hospital admission, ICU consultation, or the 72-hour mortality.