CVSARRP: A framework to predict the risk of adverse to severe adverse reactions for 10855 diseases after COVID-19 vaccination

Abstract

COVID-19 vaccines greatly reduce the risk of infection with SARS-CoV-2. However, some people have adverse reactions after vaccination, and these can sometimes be severe. Gender, age, vaccines, and especially certain diseases histories are related to severe adverse reactions following COVID-19 vaccination. However, there are thousands of diseases and only some are known to be related to these severe adverse reactions. The risk of severe adverse reactions with other diseases remains unknown. Therefore, there is a need for predictive studies to provide improved medical care and minimize risk. Herein, we analyzed the statistical results of existing COVID-19 vaccine adverse reaction data and proposed a COVID-19 vaccine severe adverse reaction risk prediction method, named CVSARRP. The performance of the CVSARRP method was tested using the leave-one-out cross-validation approach. The correlation coefficient between the predicted and real risk is greater than 0.86. The CVSARRP method predicts the risk from adverse reactions to severe adverse reactions after COVID-19 vaccination for 10855 diseases. People with certain diseases, such as central nervous system diseases, heart diseases, urinary system disease, anemia, cancer, and respiratory tract disease, among others, may potentially have increased of severe adverse reactions following vaccination against COVID-19 and experiencing adverse events.

Keywords

• COVID-19 vaccine
• Severe adverse reaction
• Risk factor
• Network module division
• Disease ontology
• Disease similarity