Machine learning and comorbidity network analysis for hospitalized patients with COVID-19 in a city in Southern Brazil.

The large amount of data generated during the COVID-19 pandemic requires advanced tools for the long-term prediction of risk factors associated with COVID-19 mortality with higher accuracy. Machine learning (ML) methods directly address this topic and are essential tools to guide public health interventions. Here, we used ML to investigate the importance of demographic and clinical variables on COVID-19 mortality. We also analyzed how comorbidity networks are structured according to age groups. We conducted a retrospective study of COVID-19 mortality with hospitalized patients from Londrina, Parana, Brazil, registered in the database for severe acute respiratory infections (SIVEP-Gripe), from January 2021 to February 2022. We tested four ML models to predict the COVID-19 outcome: Logistic Regression, Support Vector Machine, Random Forest, and XGBoost. We also constructed a comorbidity network to investigate the impact of co-occurring comorbidities on COVID-19 mortality. Our study comprised 8358 hospitalized patients, of whom 2792 (33.40%) died. The XGBoost model achieved excellent performance (ROC-AUC = 0.90). Both permutation method and SHAP values highlighted the importance of age, ventilatory support status, and intensive care unit admission as key features in predicting COVID-19 outcomes. The comorbidity networks for old deceased patients are denser than those for young patients. In addition, the co-occurrence of heart disease and diabetes may be the most important combination to predict COVID-19 mortality, regardless of age and sex. This work presents a valuable combination of machine learning and comorbidity network analysis to predict COVID-19 outcomes. Reliable evidence on this topic is crucial for guiding the post-pandemic response and assisting in COVID-19 care planning and provision.

Severe Acute Respiratory Syndrome by SARS-CoV-2 Infection or Other Etiologic Agents Among Brazilian Indigenous Population: An Observational Study from the First Year of Coronavirus Disease (COVID)-19 Pandemic.

BACKGROUND: Indigenous peoples are vulnerable to pandemics, including to the coronavirus disease (COVID)-19, since it causes high mortality and specially, the loss of elderly Indigenous individuals. METHODS: The epidemiological data of severe acute respiratory syndrome (SARS) by SARS-CoV-2 infection or other etiologic agents (OEA) among Brazilian Indigenous peoples during the first year of COVID-19 pandemic was obtained from a Brazilian Ministry of Health open-access database to perform an observational study. Considering only Indigenous individuals diagnosed with SARS by COVID-19, the epidemiology data were also evaluated as risk of death. The type of sample collection for virus screening, demographic profile, clinical symptoms, comorbidities, and clinical evolution were evaluated. The primary outcome was considered the death in the Brazilian Indigenous individuals and the secondary outcome, the characteristics of Brazilian Indigenous infected by SARS-CoV-2 or OEA, as the need for intensive care unit admission or the need for mechanical ventilation support. The statistical analysis was done using Logistic Regression Model. Alpha of 0.05. FINDINGS: A total of 3,122 cases of Indigenous individuals with SARS in Brazil were reported during the first year of the COVID-19 pandemic. Of these, 1,994 were diagnosed with COVID-19 and 730/1,816 (40.2%) of them died. The death rate among individuals with SARS-CoV-2 was three-fold increased when compared to the group of individuals with OEA. Several symptoms (myalgia, loss of smell, and sore throat) and comorbidities (cardiopathy, systemic arterial hypertension, and diabetes mellitus) were more prevalent in the COVID-19 group when compared to Indigenous individuals with OEA. Similar profile was observed considering the risk of death among the Indigenous individuals with COVID-19 who presented several symptoms (oxygen saturation <95%, dyspnea, and respiratory distress) and comorbidities (renal disorders, cardiopathy, and diabetes mellitus). The multivariate analysis was significant in differentiating between the COVID-19-positive and non-COVID-19 patients [X2 (7)=65.187; P-value<0.001]. Among the patients' features, the following contributed in relation to the diagnosis of COVID-19: age [≥43 years-old [y.o.]; OR=1.984 (95%CI=1.480-2.658)]; loss of smell [OR=2.373 (95%CI=1.461-3.854)]; presence of previous respiratory disorders [OR=0.487; 95%CI=0.287-0.824)]; and fever [OR=1.445 (95%CI=1.082-1.929)]. Also, the multivariate analysis was able to predict the risk of death [X2 (9)=293.694; P-value<0.001]. Among the patients' features, the following contributed in relation to the risk of death: male gender [OR=1.507 (95%CI=1.010-2.250)]; age [≥60 y.o.; OR=3.377 (95%CI=2.292-4.974)]; the need for ventilatory support [invasive mechanical ventilation; OR=24.050 (95%CI=12.584-45.962) and non-invasive mechanical ventilation; OR=2.249 (95%CI=1.378-3.671)]; dyspnea [OR=2.053 (95%CI=1.196-3.522)]; oxygen saturation <95% [OR=1.691 (95%CI=1.050-2.723)]; myalgia [OR=0.423 (95%CI=0.191-0.937)]; and the presence of kidney disorders [OR=3.135 (95%CI=1.144-8.539)]. INTERPRETATION: The Brazilian Indigenous peoples are in a vulnerable situation during the COVID-19 pandemic and presented an increased risk of death due to COVID-19. Several factors were associated with enhanced risk of death, as male sex, older age (≥60 y.o.), and need for ventilatory support; also, other factors might help to differentiate SARS by COVID-19 or by OEA, as older age (≥43 y.o.), loss of smell, and fever. FUNDING: Fundação de Amparo à Pesquisa do Estado de São Paulo (Foundation for Research Support of the State of São Paulo; #2021/05810-7).

Characteristics and outcomes of pregnant women with SARS-CoV-2 infection and other severe acute respiratory infections (SARI) in Brazil from January to November 2020.

BACKGROUND: Knowledge about COVID-19 in pregnancy is limited, and evidence on the impact of the infection during pregnancy and postpartum is still emerging. AIM: To analyze maternal morbidity and mortality due to severe acute respiratory infections (SARI), including COVID-19, in Brazil. METHODS: National surveillance data from the SIVEP-Gripe (Sistema de Informação de Vigilância Epidemiológica da Gripe) was used to describe currently and recently pregnant women aged 10-49 years hospitalized for SARI from January through November, 2020. SARI cases were grouped into: COVID-19; influenza or other detected agent SARI; and SARI of unknown etiology. Characteristics, symptoms and outcomes were presented by SARI type and region. Binomial proportion and 95% confidence intervals (95% CI) for outcomes were obtained using the Clopper-Pearson method. RESULTS: Of 945,460 SARI cases in the SIVEP-Gripe, we selected 11,074 women aged 10-49 who were pregnant (7964) or recently pregnant (3110). COVID-19 was confirmed in 49.4% cases; 1.7% had influenza or another etiological agent; and 48.9% had SARI of unknown etiology. The pardo race/ethnic group accounted for 50% of SARI cases. Hypertension/Other cardiovascular diseases, chronic respiratory diseases, diabetes, and obesity were the most common comorbidities. A total of 362 women with COVID-19 (6.6%; 95%CI 6.0-7.3) died. Mortality was 4.7% (2.2-8.8) among influenza patients, and 3.3% (2.9-3.8) among those with SARI of unknown etiology. The South-East, Northeast and North regions recorded the highest frequencies of mortality among COVID-19 patients. CONCLUSION: Mortality among pregnant and recently pregnant women with SARIs was elevated among those with COVID-19, particularly in regions where maternal mortality is already high.

Characteristics and outcomes of pregnant women with SARS-CoV-2 infection and other severe acute respiratory infections (SARI) in Brazil from January to November 2020

ABSTRACT Background: Knowledge about COVID-19 in pregnancy is limited, and evidence on the impact of the infection during pregnancy and postpartum is still emerging. Aim: To analyze maternal morbidity and mortality due to severe acute respiratory infections (SARI), including COVID-19, in Brazil. Methods: National surveillance data from the SIVEP-Gripe (Sistema de Informação de Vigilância Epidemiológica da Gripe) was used to describe currently and recently pregnant women aged 10-49 years hospitalized for SARI from January through November, 2020. SARI cases were grouped into: COVID-19; influenza or other detected agent SARI; and SARI of unknown etiology. Characteristics, symptoms and outcomes were presented by SARI type and region. Binomial proportion and 95% confidence intervals (95% CI) for outcomes were obtained using the Clopper-Pearson method. Results: Of 945,460 SARI cases in the SIVEP-Gripe, we selected 11,074 women aged 10-49 who were pregnant (7964) or recently pregnant (3110). COVID-19 was confirmed in 49.4% cases; 1.7% had influenza or another etiological agent; and 48.9% had SARI of unknown etiology. The pardo race/ethnic group accounted for 50% of SARI cases. Hypertension/Other cardiovascular diseases, chronic respiratory diseases, diabetes, and obesity were the most common comorbidities. A total of 362 women with COVID-19 (6.6%; 95%CI 6.0-7.3) died. Mortality was 4.7% (2.2-8.8) among influenza patients, and 3.3% (2.9-3.8) among those with SARI of unknown etiology. The South-East, Northeast and North regions recorded the highest frequencies of mortality among COVID-19 patients. Conclusion: Mortality among pregnant and recently pregnant women with SARIs was elevated among those with COVID-19, particularly in regions where maternal mortality is already high.