Predicting the Risk of COVID-19 Among Adult Patients With Diabetes: A Machine Learning Approach

In this study our aim was to develop a machine learning model that could accurately predict the risk of acquiring COVID-19 in community-dwelling adults with type 1 and/or type 2 diabetes in Alberta, Canada.

This predictive supervised machine learning study included adults (≥18 years old) living in Alberta, Canada, between April 1, 2019, and March 31, 2021, with pre-existing diabetes (n=372,055, excluding 2,541 due to migration; final sample size 369,514). The outcome of interest was a positive severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) polymerase chain reaction test result between March 1, 2020, and March 1, 2021. Model features were extracted from routinely collected Alberta administrative health data from March 1, 2015, to March 1, 2020. Fifteen algorithms were trained on 67% of the data and the top performer (Light Gradient Boost [LGBoost] model) was validated on the remaining 33%. The model was calibrated and model performance was assessed using area under the receiver-operating characteristic curve (AUROC), area under the precision recall curve (AUPRC), and threshold analyses.

Among the 369,514 individuals with diabetes, 140,511 were tested, of whom 13,082 had a positive SARS-CoV-2 test. The LGBoost model incorporated 367 features with AUROC and AUPRC of 0.69 and 0.08, respectively. The model was well-calibrated for common risk thresholds (<0.2 probability) with high specificity (≥0.98 at all thresholds); however, sensitivity and positive predictive values were low at all thresholds (≤0.08 and ≤0.18, respectively).

The LGBoost model lacked the sensitivity to be clinically useful in predicting SARS-CoV-2 infection in Albertans with diabetes. Alternative data sources may be required to improve future COVID-19 prediction models from the community.