RESUMEN
BackgroundPatients with severe COVID-19 have overwhelmed healthcare systems worldwide. We hypothesized that Machine Learning (ML) models could be used to predict risks at different stages of management (at diagnosis, hospital admission and ICU admission) and thereby provide insights into drivers and prognostic markers of disease progression and death. MethodsFrom a cohort of approx. 2.6 million citizens in the two regions of Denmark, SARS-CoV-2 PCR tests were performed on subjects suspected for COVID-19 disease; 3944 cases had at least one positive test and were subjected to further analysis. A cohort of SARS- CoV-2 positive cases from the United Kingdom Biobank was used for external validation. FindingsThe ML models predicted the risk of death (Receiver Operation Characteristics - Area Under the Curve, ROC-AUC) of 0.904 at diagnosis, 0.818, at hospital admission and 0.723 at Intensive Care Unit (ICU) admission. Similar metrics were achieved for predicted risks of hospital and ICU admission and use of mechanical ventilation. We identified some common risk factors, including age, body mass index (BMI) and hypertension as driving factors, although the top risk features shifted towards markers of shock and organ dysfunction in ICU patients. The external validation indicated fair predictive performance for mortality prediction, but suboptimal performance for predicting ICU admission. InterpretationML may be used to identify drivers of progression to more severe disease and for prognostication patients in patients with COVID-19. Prognostic features included age, BMI and hypertension, although markers of shock and organ dysfunction became more important in more severe cases. We provide access to an online risk calculator based on these findings. FundingThe study was funded by grants from the Novo Nordisk Foundation to MS (#NNF20SA0062879 and #NNF19OC0055183) and MN (#NNF20SA0062879). The foundation took no part in project design, data handling and manuscript preparation.
RESUMEN
Background: Knowledge about the COVID-19 outbreak is still sparse especially in a cross-national setting. COVID-19 is caused by a SARS-CoV-2 infection. The aim of the study is to contribute to the surveillance of the pandemic by bringing new knowledge about SARS-CoV-2 seropositivity among healthcare workers and evaluating whether certain job functions is associated with a higher risk of being infected, and to clarify if such association is mediated by the number of individuals that the employees meet during a workday. Finally, we will investigate regional and national differences in seroprevalence. Methods: A bi-national prospective observational cohort study including 3,272 adults employed at Falck in Sweden and Denmark. Participants were tested for SARS-CoV-2 antibodies every second week for a period of 8 weeks from June 22, 2020 until August 10, 2020. Descriptive statistics as well as multivariable logistic regression analyses were applied. Results: Of the 3,272 Falck employees participating in this study, 159 (4.9%) tested positive for SARS-CoV-2 antibodies. The seroprevalence was lower among Danish Falck employees than among those from Sweden (2.8% in Denmark and 8.3% in Sweden). We also found that number of customer or patient contacts during a workday was the most prominent predictor for seropositivity, and that ambulance staff was the most vulnerable staff group. Conclusions: Our study presents geographical variations in seroprevalence within the Falck organization and shows evidence that social interaction is one of the biggest risk factors for getting infected with SARS-CoV-2.