Análise da confiabilidade do Sistema de Triagem de Manchester para determinar o grau de prioridade pacientes em serviços de urgência.

Abstract

Manchester Triage System (MTS) has been adopted in most Brazilianemergency services to triage patients. However, few studies aimed to evaluate the reliability of this instrument. Objective: To analyze the MTS reliability to determine the priority of patients in emergency departments. Method: It is a reliability study with a sample of 361 nurses to evaluate the external reliability, and 153 to evaluate internal reliability of the MTS. The study was developed in three stages: Stage 1 - instrumental: getting clinical cases from the Brazilian Group Risk Classification andvalidate these with experts; Step 2 - Test - assessment of interobserver agreement through self-applied electronic questionnaires; Step 3 - Re-test - assessment of intraobservers agreement to the nurses who participated in the phase 2 study. Data were analyzed using Statistical Package for Social Science - version 19, Minitab version 15 and GraphPad Prism version 6.0. The tests used was: Kappa Index, Kruskal- Wallis test, Mann-Whitney test with Bonferroni correction, descriptive statistics with tables of frequency distribution, measures of central tendency and dispersion, and linear regression analysis. Results: The majority (79.23%) of nurses had between one and ten years of graduation, 66.48% say no had content on risk classification, and 78.12% no had content about MTS at graduation. The correct choice of the flowchart explained 16% of the variation in the correct indication of the risk level (R²: 0.16; p <0.0001), and the correct choice of discriminating explained 77% of the correct choice of the risk level (R² = 0.77; p <0.0001). The overtriage was more frequent in the fifth level of severity, with ranging between 17% and 18%. The "undertriage" was more frequent in level II of severity, and was 27%. The MTS external and internal reliability varied from moderate to substantial, with Kappavalues respectivelybetween 0.55 and 0.72, (p <0.001) and between 0.57 and 0.78 (p <0.05). The variables "length of professional experience as a nurse," "time experience as a nurse in emergency care" and "time experience as a nurse in CR" were associated with external and internal reliability of the MTS. The better the selfassessment of nurses about the ability to use the MTS, the greater the percentage of agreement on the choice of discriminating. Conclusion: The MTS reliability varied from moderate to substantial. The insertion of nurses in clinical practice, previous experience in emergency services and with triage are important for external and 9 internal reliability of the MTS. It is recommended that content about CR and MTS are included as mandatory conten in nurse graduation. It has been proven that the correct choice of the flowchart has little influence in the correct choice of level of risk. It is recommended conducting future studies to understand the reasons of errors intriage in order to devise strategies aimed at increasing the nurse's evaluation of reliability using the MTS.