RESUMO
OBJECTIVE: To design and validate a triage system for emergency services in primary care. METHODS: A classification of 5 priority levels was used (red, orange, yellow, green and blue). The first phase of the project was documentation, a classification based on the International Classification of Primary Care (ICPC2), and grouping the reasons for consultation with their respective decision trees into 97 different codes. The second phase was a pilot with 115 patients in which two independent evaluators performed triage. In the third stage of validation, triage was carried out on patients arriving at the emergency services (23,168 patients) over an eight month period, with continuous monitoring of the results. RESULTS: The level of concordance between two independent observers in the patients of the pilot obtained a Kappa of 0.7. The time of triage was 108.82sec (SD; 94.14). The main reasons for consultation were ear, nose and throat problems (27.6%), respiratory (15.7%), and musculoskeletal (14%). Distribution by priority: red (0.1%), orange (8.3%), yellow (17.9%), green (62.7%), blue (11%). Triage discharges/hospital admittance levels: 3.79/3.36 (P<.05). Professionals agreed with the triage level in more than 99% of cases. CONCLUSIONS: The triage system shows good interobserver agreement results, has a good correlation with the referral hospital, and there was no disagreement with the opinion of professionals.