Analysis of the validity of the five-level TRIPED-GM paediatric triage system

To determine the validity of the five-level TRIPED-GM pediatric triage system. Unicentric, observational, descriptive, cross-sectional study of 485 patients aged 0–16 years in the pediatric emergency department of the HGU Gregorio Marañon. Two measures of validity were used: a direct measure calculated by the sensitivity and specificity obtained based on the number of infratriages and overtriages of the priorities given by classification nurses compared with a panel of experts and another indirect measure by the length of stay, the resources consumed and the percentage of income for each priority level. 10 patients were incorrectly classified, 4 (0.8%) were considered infratriages and 6 (1.2%) overtriages. The results showed a sensitivity of 99.45% (95% CI 96.5–99.97%) and a specificity of 99.01% (95% CI 96.9−99.7%) for high priorities (P2 and P3) and 98.99% (95% CI 96.8–99.6%) and 98.4% (95% CI 96.84−99.74%) respectively for low priorities (P4 and P5). The quadratic weighted Kappa index was 0.96 (95% CI 0.94−0.98; p = 0.0000). Resource consumption showed moderate Spearman correlation coefficients as the priority level increased. The percentage of admissions and the need for observation increased as the priority level p = 0,000 increased, not requiring observation or admitting any patients with priority 5. The TRIPED-GM pediatric triage system is valid for use in emergency departments with similar patients. Determinar la validez del sistema de cinco niveles de triage pediátrico TRIPED-GM. Estudio unicéntrico, observacional, descriptivo, transversal con 485 pacientes de 0 a 15 años realizado en un servicio de urgencias pediátricas de tercer nivel. Se utilizaron dos medidas de validez: una directa calculada mediante la sensibilidad y especificidad obtenidas en base al número de infratriages y sobretriages de las prioridades otorgadas por enfermeras de clasificación comparadas con un panel de expertos y otra indirecta medida por el tiempo de estancia, los recursos consumidos y el porcentaje de ingresos de cada nivel de prioridad. Diez pacientes fueron incorrectamente clasificados, 4 (0,8%) se consideraron infratriages y 6 (1,2%) sobretriages. Los resultados mostraron una sensibilidad del 99,45% (IC 95%: 96,5–99,97%) y una especificidad del 99,01% (IC 95%:96,9−99,7%) para las prioridades altas (P2 y P3) y del 98,99% (IC 95%: 96,8–99,6%) y 98,4% (IC 95%:96,84−99,74%) para las prioridades bajas (P4 y P5) respectivamente. El índice Kappa ponderado cuadrático fue