Validation of the Kuwait Progression Indicator Score for predicting progression of severity in COVID19

Authors:
Preprint
Publisher website
Google Scholar
Abstract
Introduction: Identifying patients with COVID19, at risk of having a severe clinical course during their hospitalization is important for appropriate allocation of clinical resources. We recently described the Kuwait Progression Indicator based on laboratory findings, in an initial training cohort derived from the first series of 1096 consecutive patients admitted to Jaber AlAhmad AlSabah Hospital in Kuwait. The aim of this study was to validate the KPI scoring system in an independent cohort of patients with COVID19. Methodology: Data was collected prospectively for consecutive patients admitted to Jaber AlAhmad AlSabah Hospital in Kuwait between 24th February till 28th April 2020. Patients were grouped according to the severity of their clinical course as their main outcome, based on clinical and radiological parameters, with ICU admission and death as secondary outcomes. Model discrimination was assessed through the area under the receiver operating characteristic curve (AUC) while model calibration was assessed through a calibration plot and measures of slope and calibration in the large (CITL). Results: Of 752 patients not used in model development previously, 414 met the criteria for inclusion in this validation study. The baseline characteristics for these 752 patients were similar to the patients that were included in our validation cohort. The area under the curve was equal to 0.904 (95% CI, between 0.867 to 0.942), indicating good model discrimination. The calibration plot and CITL confirmed reasonably good model calibration. Sensitivity and specificity were above 90% for the low and high risk levels respectively. Conclusions: We were able to validate our previously described laboratory based prognostic scoring system for COVID19 patients, to predict which patients progressed to a severe clinical course.
All Related Versions

This publication has 11 references indexed in Scilit: