Important: VAERS reports alone cannot determine if a vaccine caused an adverse event. Reports may contain incomplete, inaccurate, or unverified information. Correlation does not equal causation.
Which adverse events are reported disproportionately often for each vaccine? This dashboard uses Proportional Reporting Ratio (PRR) analysis — the same statistical method used by the FDA and EMA for pharmacovigilance — to identify potential safety signals across 1.98 million VAERS reports.
A PRR ≥ 2 with at least 3 cases and statistical significance (χ² ≥ 3.84, p < 0.05) is classified as a signal. This is a screening tool — signals require further investigation to confirm causality.
Signal detection is the process of identifying potential safety concerns from post-market surveillance data. Regulatory agencies like the FDA and EMA routinely perform this type of analysis on VAERS and other pharmacovigilance databases to catch safety issues that may not have been detected in clinical trials.
The PRR compares the proportion of a specific adverse event for one vaccine against the proportion for all other vaccines. If Vaccine A has 5% of its reports mentioning myocarditis, but all other vaccines combined only have 0.1% of reports mentioning myocarditis, the PRR would be 50 — indicating myocarditis is reported 50 times more frequently for Vaccine A.
The FDA uses a combination of PRR, Empirical Bayesian Geometric Mean (EBGM), and other methods to screen VAERS weekly for new signals. When a signal is detected, it triggers a more detailed investigation including clinical review, epidemiological studies, and sometimes label changes or safety communications.
Our PRR analysis correctly identifies several well-established vaccine safety signals that have been confirmed through extensive research: intussusception after rotavirus vaccine, myocarditis after mRNA COVID vaccines, thrombosis after adenoviral COVID vaccines, and post-herpetic neuralgia after Zostavax. This validation confirms the methodology is working as expected.