Using Automated Machine Learning to Predict the Mortality of Patients With COVID-19: Prediction Model Development Study.

BACKGROUND: During a pandemic, it is important for clinicians to stratify patients and decide who receives limited medical resources. Machine learning models have been proposed to accurately predict COVID-19 disease severity. Previous studies have typically tested only one machine learning algorithm and limited performance evaluation to area under the curve analysis. To obtain the best results possible, it may be important to test different machine learning algorithms to find the best prediction model. OBJECTIVE: In this study, we aimed to use automated machine learning (autoML) to train various machine learning algorithms. We selected the model that best predicted patients' chances of surviving a SARS-CoV-2 infection. In addition, we identified which variables (ie, vital signs, biomarkers, comorbidities, etc) were the most influential in generating an accurate model. METHODS: Data were retrospectively collected from all patients who tested positive for COVID-19 at our institution between March 1 and July 3, 2020. We collected 48 variables from each patient within 36 hours before or after the index time (ie, real-time polymerase chain reaction positivity). Patients were followed for 30 days or until death. Patients' data were used to build 20 machine learning models with various algorithms via autoML. The performance of machine learning models was measured by analyzing the area under the precision-recall curve (AUPCR). Subsequently, we established model interpretability via Shapley additive explanation and partial dependence plots to identify and rank variables that drove model predictions. Afterward, we conducted dimensionality reduction to extract the 10 most influential variables. AutoML models were retrained by only using these 10 variables, and the output models were evaluated against the model that used 48 variables. RESULTS: Data from 4313 patients were used to develop the models. The best model that was generated by using autoML and 48 variables was the stacked ensemble model (AUPRC=0.807). The two best independent models were the gradient boost machine and extreme gradient boost models, which had an AUPRC of 0.803 and 0.793, respectively. The deep learning model (AUPRC=0.73) was substantially inferior to the other models. The 10 most influential variables for generating high-performing models were systolic and diastolic blood pressure, age, pulse oximetry level, blood urea nitrogen level, lactate dehydrogenase level, D-dimer level, troponin level, respiratory rate, and Charlson comorbidity score. After the autoML models were retrained with these 10 variables, the stacked ensemble model still had the best performance (AUPRC=0.791). CONCLUSIONS: We used autoML to develop high-performing models that predicted the survival of patients with COVID-19. In addition, we identified important variables that correlated with mortality. This is proof of concept that autoML is an efficient, effective, and informative method for generating machine learning-based clinical decision support tools.

Correlation of Coagulation Parameters With Clinical Outcomes During the Coronavirus-19 Surge in New York: Observational Cohort.

IMPORTANCE: COVID-19 has caused a worldwide illness and New York became the epicenter of COVID-19 in the United States from Mid-March to May 2020. OBJECTIVE: To investigate the coagulopathic presentation of COVID and its natural course during the early stages of the COVID-19 surge in New York. To investigate whether hematologic and coagulation parameters can be used to assess illness severity and death. DESIGN: Retrospective case study of positive COVID inpatients between March 20, 2020-March 31, 2020. SETTING: Montefiore Health System main hospital, Moses, a large tertiary care center in the Bronx. PARTICIPANTS: Adult inpatients with positive COVID tests hospitalized at MHS. EXPOSURE FOR OBSERVATIONAL STUDIES: Datasets of participants were queried for demographic (age, sex, socioeconomic status, and self-reported race and/or ethnicity), clinical and laboratory data. MAIN OUTCOME AND MEASURES: Relationship and predictive value of measured parameters to mortality and illness severity. RESULTS: Of the 225 in this case review, 75 died during hospitalization while 150 were discharged home. Only the admission PT, absolute neutrophil count (ANC) and first D-Dimer could significantly differentiate those who were discharged alive and those who died. Logistic regression analysis shows increased odds ratio for mortality by first D-Dimer within 48 hrs. of admission. The optimal cut-point for the initial D-Dimer to predict mortality was found to be 2.1 µg/mL. 15% of discharged patients required readmission and more than a third of readmitted patients died (5% of all initially discharged). CONCLUSION: We describe here a comprehensive assessment of hematologic and coagulation parameters in COVID-19 and examine the relationship of these to mortality. We demonstrate that both initial and maximum D-Dimer values are biomarkers that can be used for survival assessments. Furthermore, D-Dimer may be useful to follow up discharged patients.

Anticoagulation in COVID-19: Effect of Enoxaparin, Heparin, and Apixaban on Mortality.

BACKGROUND: Mortality in coronavirus disease of 2019 (COVID-19) is associated with increases in prothrombotic parameters, particularly D-dimer levels. Anticoagulation has been proposed as therapy to decrease mortality, often adjusted for illness severity. OBJECTIVE: We wanted to investigate whether anticoagulation improves survival in COVID-19 and if this improvement in survival is associated with disease severity. METHODS: This is a cohort study simulating an intention-to-treat clinical trial, by analyzing the effect on mortality of anticoagulation therapy chosen in the first 48 hours of hospitalization. We analyzed 3,625 COVID-19+ inpatients, controlling for age, gender, glomerular filtration rate, oxygen saturation, ventilation requirement, intensive care unit admission, and time period, all determined during the first 48 hours. RESULTS: Adjusted logistic regression analyses demonstrated a significant decrease in mortality with prophylactic use of apixaban (odds ratio [OR] 0.46, p = 0.001) and enoxaparin (OR = 0.49, p = 0.001). Therapeutic apixaban was also associated with decreased mortality (OR 0.57, p = 0.006) but was not more beneficial than prophylactic use when analyzed over the entire cohort or within D-dimer stratified categories. Higher D-dimer levels were associated with increased mortality (p < 0.0001). When adjusted for these same comorbidities within D-dimer strata, patients with D-dimer levels < 1 µg/mL did not appear to benefit from anticoagulation while patients with D-dimer levels > 10 µg/mL derived the most benefit. There was no increase in transfusion requirement with any of the anticoagulants used. CONCLUSION: We conclude that COVID-19+ patients with moderate or severe illness benefit from anticoagulation and that apixaban has similar efficacy to enoxaparin in decreasing mortality in this disease.