Predicting SARS-CoV-2 infection duration at hospital admission:a deep learning solution.

COVID-19 cases are increasing around the globe with almost 5 million of deaths. We propose here a deep learning model capable of predicting the duration of the infection by means of information available at hospital admission. A total of 222 patients were enrolled in our observational study. Anagraphical and anamnestic data, COVID-19 signs and symptoms, COVID-19 therapy, hematochemical test results, and prior therapies administered to patients are used as predictors. A set of 55 features, all of which can be taken in the first hours of the patient's hospitalization, was considered. Different solutions were compared achieving the best performance with a sequential convolutional neural network-based model merged in an ensemble with two different meta-learners linked in cascade. We obtained a median absolute error of 2.7 days (IQR = 3.0) in predicting the duration of the infection; the error was equally distributed in the infection duration range. This tool could preemptively give an outlook of the COVID-19 patients' expected path and the associated hospitalization effort. The proposed solution could be viable in tackling the huge burden and the logistics complexity of hospitals or rehabilitation centers during the pandemic waves. With data taken ad admission, entering a PCA-based feature selection, a k-fold cross-validated CNN-based model was implemented. After external texting, a median absolute error of 2.7 days [IQR = 3 days].

The methodology of a "living" COVID-19 registry development in a clinical context.

OBJECTIVE: The aim of this study was to describe an innovative methodology of a registry development, constantly updated for the scientific assessment and analysis of the health status of the population with COVID-19. STUDY DESIGN AND SETTING: A methodological study design to develop a multi-site, Living COVID-19 Registry of COVID-19 patients admitted in Fondazione Don Gnocchi centres started in March 2020. RESULTS: The integration of the living systematic reviews and focus group methodologies led to a development of a registry which includes 520 fields filled in for 748 COVID-19 patients recruited from 17 Fondazione Don Gnocchi centres. The result is an evidence and experience-based registry, according to the evolution of a new pathology which was not known before outbreak of March 2020 and with the aim of building knowledge to provide a better quality of care for COVID-19 patients. CONCLUSION: A Living COVID-19 Registry is an open, living and up to date access to large-scale patient-level data sets that could help identifying important factors and modulating variable for recognising risk profiles and predicting treatment success in COVID-19 patients hospitalized. This innovative methodology might be used for other registries, to be sure which the data collected is an appropriate means of accomplishing the scientific objectives planned. CLINICAL TRIAL REGISTRATION NUMBER: not applicable.

Mortality and characteristics of older people dying with COVID-19 in Lombardy nursing homes, Italy: An observational cohort study.

BACKGROUND: The aim of the study was to describe the epidemiological characteristics of Nursing Homes (NHs) residents infected by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection and to compute the related case-fatality rate. MATERIALS AND METHODS: The outcomes were mortality and case-fatality rate with related epidemiological characteristics (age, sex, comorbidity, and frailty). RESULTS: During the COVID-19 outbreak lasted from March 1 to May 7, 2020, 330 residents died in Fondazione Don Gnocchi NHs bringing the mortality rate to 27% with a dramatic increase compared to the same period of 2019, when it was 7.5%. Naso/oropharyngeal swabs resulted positive for COVID-19 in 315 (71%) of the 441of the symptomatic/exposed residents tested. The COVID-19 population was 75% female, with a 17% overall fatality rate and sex-specific fatality rates of 19% and 13% for females and males, respectively. Fifty-six percent of deaths presented SARS-CoV-2-associated pneumonia, 15% cardiovascular, and 29% miscellaneous pathologies. CONCLUSION: Patients' complexity and frailty might influence SARS-CoV-2 infection case-fatality rate estimates. A COVID-19 register is needed to study COVID-19 frail patients' epidemiology and characteristics.