Development and validation of prediction models for mechanical ventilation, renal replacement therapy, and readmission in COVID-19 patients

Abstract

Objective Coronavirus disease 2019 (COVID-19) patients are at risk for resource-intensive outcomes including mechanical ventilation (MV), renal replacement therapy (RRT), and readmission. Accurate outcome prognostication could facilitate hospital resource allocation. We develop and validate predictive models for each outcome using retrospective electronic health record data for COVID-19 patients treated between March 2 and May 6, 2020. Materials and Methods For each outcome, we trained 3 classes of prediction models using clinical data for a cohort of SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2)–positive patients (n = 2256). Cross-validation was used to select the best-performing models per the areas under the receiver-operating characteristic and precision-recall curves. Models were validated using a held-out cohort (n = 855). We measured each model’s calibration and evaluated feature importances to interpret model output. Results The predictive performance for our selected models on the held-out cohort was as follows: area under the receiver-operating characteristic curve—MV 0.743 (95% CI, 0.682-0.812), RRT 0.847 (95% CI, 0.772-0.936), readmission 0.871 (95% CI, 0.830-0.917); area under the precision-recall curve—MV 0.137 (95% CI, 0.047-0.175), RRT 0.325 (95% CI, 0.117-0.497), readmission 0.504 (95% CI, 0.388-0.604). Predictions were well calibrated, and the most important features within each model were consistent with clinical intuition. Discussion Our models produce performant, well-calibrated, and interpretable predictions for COVID-19 patients at risk for the target outcomes. They demonstrate the potential to accurately estimate outcome prognosis in resource-constrained care sites managing COVID-19 patients. Conclusions We develop and validate prognostic models targeting MV, RRT, and readmission for hospitalized COVID-19 patients which produce accurate, interpretable predictions. Additional external validation studies are needed to further verify the generalizability of our results.

Development and validation of the MMCD score to predict kidney replacement therapy in COVID-19 patients

“…Logistic L1 had the best accuracy in the validation cohort. However, the discrimination results were inferior than the one observed in the present analysis (0.847 [95% CI, 0.772-0.936]) and the study has several limitations: many risk predictor variables, hindering the applicability of the score and high incidence of missing variables [29].…”

“…Using predictors available at baseline and within the first hours of the admission, we could objectively predict the probability of KRT of a COVID-19 patient with AKI. With an accurate prediction, it may help to organize resource allocation to patients who are at the highest risk of KRT requirement [29], in addition to selecting patients who may benefit from renal protection strategies, close assessment and follow-up by a nephrologist [31].…”

Predicting hospital readmission risk in patients with COVID-19: A machine learning approach

“…In Rodriguez's study (2021), a predictive model for readmission in COVID-19 patients was presented based on an ML classifier. They concluded that ML and data mining-based approaches have seemed fruitful for readmission prediction [ 20 ]. Koteswari (2020) proposed an intelligent model to predict the readmission probability of various COVID-19 cases using ML techniques.…”

“…Since the COVID-19 pandemic began, several studies selected clinically important predictors for post-discharge COVID-19. For example, Rodriguez's study (2021) indicated underline chronic disease, hypoxia (oxygen saturation ≤94%), increased LDH, CRP, and ESR as the most effective factors on hospital readmission [ 20 ]. In another study performed by Mendito (2021), several clinical features such as age, neutrophilia count, sequential organ failure assessment (SOFA), LDH, CRP, and D-dimer are recognized as highly contributing factors to the readmission of COVID-19 patients [ 31 ].…”

“…CDSSs attracted increasing interest because of the growing availability of a large amount of patient-level data [ 18 , 19 ]. CDSSs using available patient data at the time of admission may provide caregivers with valuable information regarding the likelihood risk of COVID-19 readmission [ 20 , 21 ]. Machine learning (ML) algorithms are complex and flexible classification modeling that leverage big datasets to reveal new and practical patterns [ 18 , 22 ].…”