Navegando por Palavras-chave "Accidents, Traffic"

Agora exibindo 1 - 1 de 1
  • Nenhuma Miniatura disponível
    Item
    Somente Metadadados
    Método para análise das complicações decorrentes de trauma por acidentes de trânsito
    (Universidade Federal de São Paulo (UNIFESP), 2021) Lopes, Maria Carolina Barbosa Teixeira [UNIFESP]; Whitaker, Iveth Yamaguchi [UNIFESP]; Universidade Federal de São Paulo
    Background: Patients with injuries resulting from traffic accidents, who survive the severe traumatic condition, become vulnerable to the occurrence of complications during hospitalization. To improve patient outcomes and safety, complications need to be monitored and controlled. Objective: To develop a prediction model for in-hospital complication (POC) for patients with traffic accidents injuries; to verify the association between complications and independent variables; to identify risk factors associated with complications and to verify the predictive capacity of the proposed in-hospital POC model. Methods: These analysis from secondary data that sample was prospectively collected, from patients aged 14 years or older, with traumatic injuries from traffic accidents, from January 2015 to July 2016, in a municipal hospital in the city of São José dos Campos, São Paulo. To derive a POC model, the association and correlation of independent variables with complications was verified and the types of complications were analyzed according to length of stay and mortality using parametric and non-parametric statistical tests. Multivariate logistic regression was used to identify the predictive variables of the model and the adjustment was verified by the Hosmer-Lemeshow (HL) test. POC performance was assessed using the ROC and PR Curve curves. Results: The sample consisted of 327 patients and 82 (25.1%) had complications during hospitalization and infectious ones were the most frequent. The occurrence of complications was statistically associated with a higher average age, pedestrian accidents and the greater severity of trauma. The length of stay in the emergency room, the length of hospital stay (LOS), ICU stay (ICU-LOS), percentage of deaths and hospital readmission were higher in patients with complications. Were identified as independent risk factors for the occurrence of complications age, systolic blood pressure, Glasgow Coma Scale score, Revised Trauma Score, Injury Severity Score (ISS), New Injury Severity Score (NISS), ICU-LOS, LOS and readmission. The number of complications correlated with the severity of the trauma, ICU-LOS and mortality. The types of complications were associated with higher mortality. Cardiovascular and neurological complications were not associated with length of stay. The variables age and severity of trauma, measured by ISS and NISS, were identified as risk factors for the occurrence of complications in the final model of multivariate logistic regression, with HL of 0.84 and 1.0, respectively. The area under the ROC curve of the models with ISS and the NISS, reached 0.81 and 0.82, respectively, indicating excellent discrimination capacity and the area of the PR Curve for both models was compatible with those presented in the ROC curve. Conclusion: The in-hospital POC model of trauma patients resulting from traffic accidents included the independent variables age and ISS or NISS. The proposed model showed excellent predictive capacity. However, as it is an initial proposal, it requires application in a larger sample of multiple institutions to confirm the predictive capacity and enable its improvement.