Machine learning models predict long COVID outcomes based on baseline clinical and immunologic factors

Area of Science:

• Infectious Diseases
• Computational Biology
• Immunology

Background:

• Post-acute sequelae of SARS-CoV-2 (PASC), or long COVID, is a complex condition lacking clear predictive biomarkers.
• Accurate prediction of long COVID development is crucial for timely intervention and patient management.

Purpose of the Study:

• To develop and evaluate machine learning models for predicting long COVID risk.
• To identify key clinical predictors of long COVID development from early infection data.

Main Methods:

• Utilized machine learning models trained on clinical data from acutely hospitalized COVID-19 patients.
• Included antibody titers, viral load measurements, comorbidities, and demographic information as input features.
• Assessed model performance using Area Under the Receiver Operating Characteristic Curve (AUROC) and Area Under the Precision-Recall Curve (AUPRC).

Main Results:

• Machine learning models achieved moderate predictive performance (median AUROC 0.64-0.66, AUPRC 0.51-0.54).
• Low antibody titers and high viral loads at hospital admission were identified as the strongest predictors of long COVID.
• Comorbidities (respiratory, cardiac, neurologic) and female sex were significant risk factors.

Conclusions:

• Machine learning models show promise in identifying individuals at high risk for long COVID using baseline clinical data.
• These predictive capabilities can inform early intervention strategies to improve patient outcomes.
• The findings contribute to understanding long COVID risk factors and mitigating its public health impact.