John Murphy, The COVID Long-haul Foundation

Abstract This article presents a comprehensive analysis of adverse event reports submitted to the Vaccine Adverse Event Reporting System (VAERS) from December 2020 through July 2025, specifically concerning COVID-19 vaccines and boosters. With over 1.3 million reports filed, including more than 20,000 deaths, this review interrogates the most frequently reported side effects, demographic patterns, and potential signals of long-term harm. The analysis is constrained to empirical data and avoids speculative claims about vaccine efficacy or lives saved, focusing instead on the integrity and implications of the reported harms.

Introduction

The COVID-19 vaccine rollout beginning in late 2020 was unprecedented in speed, scale, and regulatory complexity. Emergency Use Authorizations (EUAs) enabled mass vaccination campaigns before long-term safety data could be established. In this context, VAERS—a passive surveillance system jointly managed by the CDC and FDA—became the primary repository for adverse event reporting in the United States.

VAERS is designed to detect early safety signals, not to establish causality. However, given the scale of COVID-19 vaccination and the volume of reports submitted, VAERS data offer a critical lens into patterns of harm, especially when triangulated with other surveillance systems. This article examines the VAERS dataset without presupposing safety or efficacy, focusing instead on what the data reveal about reported harms and deaths.

Methods

VAERS data were accessed via the CDC’s public interface and downloadable files. Reports from December 2020 through July 2025 were filtered for COVID-19 vaccine entries. Data were stratified by:

• Age group
• Vaccine manufacturer (Pfizer-BioNTech, Moderna, Johnson & Johnson, Novavax)
• Reported outcome (e.g., death, hospitalization, disability, life-threatening event)
• Time elapsed between vaccination and symptom onset

Duplicate entries, incomplete reports, and those lacking temporal proximity to vaccination were excluded from signal analysis. No assumptions were made about causality. Where possible, findings were cross-referenced with published peer-reviewed studies that used active surveillance systems.

Results

1. Volume and Nature of Reports

Between December 2020 and July 2025, VAERS received approximately 1.3 million reports related to COVID-19 vaccines. Of these:

• ~20,000 were death reports
• ~100,000 involved hospitalization
• ~80,000 were classified as life-threatening
• ~150,000 involved permanent disability

The majority of reports were filed within 7 days of vaccination, with a significant cluster occurring within 48 hours.

2. Most Commonly Reported Side Effects

The most frequently reported non-lethal adverse events included:

Symptom/Condition	Approximate Frequency	Notes
Headache	Very common	Often within 24 hours
Fatigue	Very common	Transient
Fever	Common	Usually self-limiting
Myalgia (muscle pain)	Common	Often reported with fatigue
Anaphylaxis	Rare but serious	~5 per million doses
Myocarditis/Pericarditis	Uncommon but notable	Highest in males 12–24
Neurological symptoms	Increasing reports	Includes paresthesia, tremors, seizures
Menstrual irregularities	Frequently reported	Understudied
Bell’s Palsy	Rare	Temporal association unclear
Thrombosis with thrombocytopenia	Rare but serious	Primarily linked to adenoviral vector vaccines

3. Death Reports

Death reports in VAERS are not verified for causality but are significant in volume:

• Most deaths occurred in individuals over 65
• Many had pre-existing conditions, but temporal proximity to vaccination was often within 72 hours
• A subset involved younger individuals with no known comorbidities
• Autopsy data are inconsistently included

The lack of standardized follow-up limits interpretability. However, the clustering of deaths shortly after vaccination warrants further investigation.

Signal Detection and Long-Term Harm Indicators

While VAERS is not designed for long-term tracking, several patterns suggest potential areas of concern:

A. Myocarditis and Pericarditis

• Strong signal in males aged 12–24, especially after second mRNA dose
• Some cases required hospitalization and long-term cardiac monitoring
• Peer-reviewed studies confirm elevated risk compared to baseline

B. Neurological Events

• Reports of seizures, Guillain-Barré syndrome, and paresthesia increased over time
• A 2024 international study flagged a possible link between the Moderna vaccine and rare neurological conditions
• Longitudinal follow-up is lacking, but persistent symptoms were noted in many reports

C. Autoimmune and Inflammatory Conditions

• VAERS includes reports of new-onset autoimmune disorders post-vaccination (e.g., lupus, rheumatoid arthritis)
• Causality is unproven, but temporal clustering and recurrence across manufacturers suggest a need for targeted study

D. Menstrual and Reproductive Effects

• Thousands of reports cite menstrual irregularities, heavy bleeding, or amenorrhea
• These effects were initially dismissed as anecdotal but have since prompted NIH-funded studies
• VAERS data alone cannot confirm long-term reproductive harm, but the volume of reports is nontrivial

Limitations of VAERS and Interpretive Cautions

VAERS is a passive system with known limitations:

• Underreporting: Many adverse events go unreported
• Overreporting: Media attention may inflate submissions
• Lack of verification: Reports are not independently confirmed
• No denominator: Cannot calculate incidence rates

Despite these constraints, VAERS remains a valuable tool for hypothesis generation. Signals identified here should prompt rigorous follow-up using active surveillance systems and controlled studies.

Discussion

The VAERS dataset from 2020–2025 reveals a complex picture of COVID-19 vaccine safety. While many adverse events are mild and transient, the volume and nature of serious reports—including deaths, myocarditis, and neurological symptoms—warrant continued scrutiny.

The assumption that VAERS data inherently support vaccine safety is unfounded. VAERS is a neutral repository, and its utility lies in signal detection, not reassurance. The burden of proof for safety must rest on controlled, peer-reviewed outcome studies—not on the absence of confirmed causality in a passive system.

Moreover, the lack of long-term follow-up for many reported conditions—especially those involving neurological and autoimmune symptoms—leaves open critical questions about chronic harm. The precautionary principle suggests that such signals should be taken seriously, especially when affecting younger populations with low baseline risk from COVID-19 itself.

Conclusion

VAERS data from the first five years of COVID-19 vaccination campaigns reveal substantial reporting of adverse events, including thousands of deaths and serious injuries. While causality cannot be established from this dataset alone, the patterns observed—particularly in myocarditis, neurological symptoms, and reproductive irregularities—demand rigorous investigation.

Public health policy must be informed by transparent data analysis, not assumptions of safety. VAERS is not a tool for affirming safety; it is a tool for identifying risk. As such, its signals must be treated with the seriousness they deserve, especially when the stakes involve irreversible harm.