Comparative Data Analysis of COVID-19 Vaccine Post-Vaccination Symptoms: Challenges, Opportunities, and Recommendations

Individuals aged five years and older must receive vaccination promptly to safeguard against COVID-19 and its potential severe complications. As per the Centers for Disease Control and Prevention (CDC), the primary national public health agency in the United States, and other health agencies worldwide, the CDC recommends the use of mRNA vaccines (Pfizer-BioNTech or Moderna COVID-19 vaccines) over Johnson & Johnson's Janssen (Non-Replicating Viral Vector) COVID-19 vaccine. A comprehensive analysis of adverse effects following vaccination was conducted using data collected from the Vaccine Adverse Event Reporting System (VAERS) between October 2021 and December 2021. The study employed the random forest method to assess the significance of predictor variables. The analysis identified eight statistically significant predictors, including allergy, medicine intake, illness history, vaccine dose series, vaccine brands, age, sex, and current recipient illness, in adverse event reporting after vaccination. The total number of symptoms was considered as the dependent variable in this analysis. Furthermore, the study utilized various Poisson models such as Regular Poisson, Generalized Poisson (G.P.), and Quasi Poisson (Q.P.) to analyze the under dispersed count data from the Generalized Linear Model (GLM). The dataset analyzing suitable method was a generalized linear model. Additionally, the outcomes of Poisson methods in R and STATA software were compared, and similar results were obtained from both software applications.