1. The Value of Prehospital Early Warning Scores to Predict in - Hospital Clinical Deterioration: A Multicenter, Observational Base-Ambulance Study.
- Author
-
Martín-Rodríguez, Francisco, Sanz-García, Ancor, Medina-Lozano, Elena, Castro Villamor, Miguel Ángel, Carbajosa Rodríguez, Virginia, del Pozo Vegas, Carlos, Fadrique Millán, Laura Natividad, Rabbione, Guillermo Ortega, Martín-Conty, José Luis, and López-Izquierdo, Raúl
- Subjects
MORTALITY risk factors ,LENGTH of stay in hospitals ,RESEARCH ,HOSPITALS ,EVALUATION of medical care ,PREDICTIVE tests ,SCIENTIFIC observation ,HOSPITAL emergency services ,CONFIDENCE intervals ,AMBULANCES ,ADVANCED cardiac life support ,RESEARCH methodology ,TIME ,MEDICAL cooperation ,DISCRIMINANT analysis ,MANN Whitney U Test ,FISHER exact test ,SEVERITY of illness index ,RISK assessment ,HOSPITAL care ,RESEARCH funding ,DESCRIPTIVE statistics ,GLASGOW Coma Scale ,CHI-squared test ,RECEIVER operating characteristic curves ,DATA analysis software ,ODDS ratio ,EMERGENCY medicine ,LONGITUDINAL method - Abstract
Early warning scores are clinical tools capable of identifying prehospital patients with high risk of deterioration. We sought here to contrast the validity of seven early warning scores in the prehospital setting and specifically, to evaluate the predictive value of each score to determine early deterioration-risk during the hospital stay, including mortality at one, two, three and seven- days since the index event. Methods: A prospective multicenter observational based-ambulance study of patients treated by six advanced life support emergency services and transferred to five Spanish hospitals between October 1, 2018 and December 31, 2019. We collected demographic, clinical, and laboratory variables. Seven risk score were constructed based on the analysis of prehospital variables associated with death within one, two, three and seven days since the index event. The area under the receiver operating characteristics was used to determine the discriminant validity of each early warning score. Results: A total of 3,273 participants with acute diseases were accurately linked. The median age was 69 years (IQR, 54–81 years), 1,348 (41.1%) were females. The overall mortality rate for patients in the study cohort ranged from 3.5% for first-day mortality (114 cases), to 7% for seven-day mortality (228 cases). The scores with the best performances for one-day mortality were Vitalpac Early Warning Score with an area under the receiver operating characteristic (AUROC) of 0.873 (95% CI: 0.81-0.9), for two-day mortality, Triage Early Warning Score with an AUROC of 0.868 (95% CI: 0.83-0.9), for three and seven-days mortality the Modified Rapid Emergency Medicine Score with an AUROC of 0.857 (0.82-0.89) and 0.833 (95% CI: 0.8-0.86). In general, there were no significant differences between the scores analyzed. Conclusions: All the analyzed scores have a good predictive capacity for early mortality, and no statistically significant differences between them were found. The National Early Warning Score 2, at the clinical level, has certain advantages. Early warning scores are clinical tools that can help in the complex decision-making processes during critical moments, so their use should be generalized in all emergency medical services. [ABSTRACT FROM AUTHOR]
- Published
- 2021
- Full Text
- View/download PDF