Burden of COVID-19 disease and vaccine coverages in Apulian splenectomized patients: A retrospective observational study.

Splenectomy/asplenia is a condition associated with immune-compromission and specific vaccines are recommended for these patients, including the anti-COVID-19 vaccine. Among the high-risk group for which vaccination was prioritized in Italy, the immunocompromised patients after therapies or treatments were included. The Apulian regional archive of hospital discharge forms was used to define the list of splenectomized Apulian inhabitants, considering data from 2015 through 2020. The overall vaccination status of asplenic patients was assessed via data collected from the Regional Immunization Database. The history of SARS-CoV-2 infection and the infectious disease outcomes were extracted from the Italian Institute of Health platform "Integrated surveillance of COVID-19 cases in Italy". 1219 Apulian splenectomized inhabitants were included; the incidence rate of SARS-CoV-2 infection was 15.0 per 100 persons-year with a proportion of re-infection equal to 6.4%; the proportion of hospitalization was 2.9%, with a case-fatality rate of 2.6%. The vaccine coverage (VC) for the anti-COVID-19 vaccine basal routine was 64.2%, for the first booster dose was 15.4%, and for the second booster dose was 0.6%. A multifactorial approach is needed to increase the vaccination uptake in this sub-group population and to increase the awareness of the asplenia-related risks to patients and health personnel.

COVID-19 diagnosis prediction in emergency care patients: a machine learning approach (preprint)

The coronavirus disease (COVID-19) pandemic has increased the necessity of immediate clinical decisions and effective usage of healthcare resources. Currently, the most validated diagnosis test for COVID-19 (RT-PCR) is in shortage in most developing countries, which may increase infection rates and delay important preventive measures. The objective of this study was to predict the risk of positive COVID-19 diagnosis with machine learning, using as predictors only results from emergency care admission exams. We collected data from a sample of 235 adult patients from the Hospital Israelita Albert Einstein in Sao Paulo, Brazil, from 17 to 30 of March, 2020, of which 102 (43%) received a positive diagnosis of COVID-19 from RT-PCR tests. Five machine learning algorithms (neural networks, random forests, gradient boosting trees, logistic regression and support vector machines) were trained on a random sample of 70% of the patients, and performance was tested on new unseen data (30%). The best predictive performance was obtained by the support vector machines algorithm (AUC: 0.85; Sensitivity: 0.68; Specificity: 0.85; Brier Score: 0.16). In conclusion, we found that targeted decisions for receiving COVID-19 tests using only routinely-collected data is a promising new area with the use of machine learning algorithms.