Sarcouncil Journal of Medicine and Surgery

Abstract: Background: The World Health Organization (WHO) Interagency Integrated Triage Tool (IITT) is designed to prioritize patients by acuity in resource-limited emergency units. We audited criterion concordance, care timeliness, selected resuscitation-equipment readiness, and staff training in a general hospital in Baghdad. Methods: This prospective single-centre clinical audit combined direct observation of 297 non-traumatic patients aged ≥12 years, 37 resuscitation-area checklist occasions, and 75 staff questionnaires from January to April 2026. Because the locally adapted observation form combined WHO emergency signs with high-risk criteria, analyses distinguished selected emergency signs requiring RED assignment from high-risk criteria requiring at least YELLOW assignment. Times were summarized as median [interquartile range]; categorical results include Wilson 95% confidence intervals (CI). Results: Patients had a median age of 53 [31–65] years; 101 (34.0%) were assigned RED, 129 (43.4%) YELLOW, and 67 (22.6%) GREEN. Arrival-to-triage time was available for 262 encounters and was 0 [0–5] minutes; 224/262 (85.5%, 95% CI 80.7–89.2) were triaged within 5 minutes. Fifty-one patients had ≥1 recorded selected emergency sign; 12/51 (23.5%, 95% CI 14.0–36.8) were assigned YELLOW rather than RED. Thirty-one had only high-risk criteria; 1/31 (3.2%, 95% CI 0.6–16.2) was assigned GREEN. First-intervention times were 7 [5–11.2], 12 [10–18], and 15 [10–19] minutes for RED, YELLOW, and GREEN patients, respectively (p<0.001). Among assigned RED patients, 72/100 (72.0%, 95% CI 62.5–79.9) received an intervention within 10 minutes. In an exploratory analysis, patients with emergency signs assigned below RED had slower intervention than concordantly assigned RED patients (10 [7–12.8] vs 5 [5–9.5] minutes; p=0.011). The crash cart was fully stocked on 25/37 (67.6%) checklist occasions. None of five nurse respondents reported formal triage training, compared with 61/69 (88.4%) physicians (Fisher exact p<0.001). Conclusions: Triage occurred rapidly, but selected emergency signs were sometimes assigned below RED and these discordant assignments were associated with slower intervention. Priority actions are criterion-based competency training, complete documentation, routine time-to-intervention monitoring, and standardized crash-cart checks. Independent expert re-triage and patient outcomes are required for formal IITT accuracy validation.