Why a VAERS report cannot establish that a vaccine caused an adverse event — and what happens after a signal is detected
The most common misunderstanding about VAERS data is treating a reported event as proof that a vaccine caused it. This page explains why VAERS reports cannot establish causation on their own, what happens after a potential signal is detected, and how causality is actually determined through stronger methods.
For a comprehensive explanation of causality frameworks, study designs, and institutional evaluation systems, see the full Causality in Vaccine Safety page.
Last updated: April 2026.
VAERS is a passive surveillance system that accepts reports of adverse events occurring after vaccination from anyone — healthcare providers, patients, family members, and manufacturers. A VAERS report documents that an event happened after vaccination. It does not establish that the vaccine caused the event.
There are four structural reasons why VAERS cannot determine causation on its own:
No denominator data. VAERS records events but does not know how many people were vaccinated. Without knowing the total number of doses administered, it is impossible to calculate how often an event actually occurs in the vaccinated population.
No verification. Reports are accepted as submitted. They are not investigated, confirmed, or validated before entering the database. A report may contain incomplete information, an unconfirmed diagnosis, or inaccurate details.
No comparison group. VAERS has no built-in way to compare adverse event rates in vaccinated people against rates in unvaccinated people. Without that comparison, there is no way to determine whether an event is occurring more often than expected.
Temporal association is not causation. The fact that an event happened after vaccination does not mean the vaccine caused it. Adverse health events occur every day in every population regardless of vaccination status. Epidemiologists call these ordinary expected events "background rates" (FDA VAERS Q&A, 2024; CDC Surveillance Manual, Ch. 21, 2024).
When millions of people receive a vaccine over a short period, some will experience heart attacks, strokes, new diagnoses, and deaths in the days following vaccination — events that would have occurred regardless of whether they were vaccinated.
Researchers at the CDC calculated what this looks like in practice. Among 10 million vaccinated people, purely by coincidence from background rates: approximately 237 would die from any cause within one day of vaccination, approximately 1,656 within seven days, and approximately 9,933 within 42 days — all from causes entirely unrelated to the vaccine (Abara et al., Journal of Infectious Diseases, 2022).
This does not mean that vaccines never cause harm. It means that observing adverse events after vaccination — even serious ones — is expected statistically and cannot by itself distinguish a vaccine-caused injury from a coincidental event. That distinction requires the controlled studies and formal frameworks described below.
When VAERS identifies an unusual pattern — a cluster of similar reports, an unexpected event type, or a reporting rate that exceeds historical norms — federal agencies treat it as a signal that warrants further investigation, not as a confirmed safety problem.
The investigation typically follows this sequence:
Step 1 — Signal detection (VAERS). The passive system flags an unusual pattern in reported events.
Step 2 — Hypothesis testing (VSD / PRISM). Active surveillance systems with linked medical records test whether the signal reflects a genuine increase in risk by comparing vaccinated and unvaccinated populations using controlled study designs. The Vaccine Safety Datalink (VSD) covers more than 12 million people annually. FDA's PRISM system accesses data on more than 190 million people.
Step 3 — Clinical case review (CISA). For complex individual cases, the Clinical Immunization Safety Assessment Project provides expert medical evaluation and detailed record review.
Step 4 — Formal studies and institutional review. If the signal is confirmed, epidemiological studies, risk-benefit analyses, and expert review by bodies such as ACIP and the National Academies inform regulatory and clinical decisions.
This sequence has worked in practice. VAERS detected the intussusception signal for RotaShield within months of licensure in 1998; controlled studies confirmed the association; ACIP withdrew its recommendation within 14 months. VAERS detected myocarditis reports following mRNA COVID-19 vaccines in 2021; VSD and international surveillance systems confirmed the association; ACIP updated clinical guidance.
In both cases, VAERS was the starting point — not the conclusion.
Determining whether a vaccine caused a specific adverse event requires structured evidence that goes well beyond VAERS reporting. The key elements include:
Temporality. The vaccine must have been administered before the adverse event occurred. This is necessary but never sufficient on its own.
Background rate comparison. Researchers compare the observed rate of the event in vaccinated people against the expected background rate in the general population. If the event occurs significantly more often after vaccination than would be expected by chance, that strengthens the case for a causal association.
Controlled comparison. Studies comparing vaccinated and unvaccinated groups — using cohort designs, self-controlled case series, or case-control methods — can estimate whether the vaccine increases the risk of a specific event while controlling for other factors.
Consistency across datasets. A signal that appears in VAERS, is confirmed in VSD, and is replicated in international surveillance systems carries more weight than a signal from any single source.
Biological plausibility. A known or proposed biological mechanism explaining how the vaccine could cause the event — such as an immune-mediated pathway — strengthens the causal case, though the absence of a known mechanism does not automatically rule out causation.
For the full framework — including the Bradford Hill viewpoints, the IOM/NASEM four-category system, the WHO AEFI algorithm, study design hierarchy, legal vs. scientific causation, and historical case studies — see the comprehensive Causality in Vaccine Safety page.
A VAERS report is a starting point for investigation, not an endpoint.
VAERS serves a genuine and essential function as an early-warning system. It has contributed to the detection of real vaccine safety signals, including RotaShield intussusception and mRNA vaccine myocarditis. But a VAERS report, by itself, cannot establish that a vaccine caused an adverse event — because the system lacks denominator data, does not verify reports, has no comparison group, and cannot distinguish causal relationships from temporal coincidence.
Causation is determined through controlled studies, active surveillance systems, formal evidence review, and institutional evaluation — not through passive report counts. Readers who want to understand how that process works in detail can find the full explanation on the Causality in Vaccine Safety page.
CDC/FDA. Guide to Interpreting VAERS Data. HHS.gov (VAERS). 2025. VAERS Data Guide →
Official guidance: no cause-and-effect relationship established for any reported event.
U.S. Food and Drug Administration. VAERS Questions and Answers. 2024. FDA →
FDA states VAERS reports generally cannot determine whether a vaccine caused an adverse event.
CDC. Chapter 21: Surveillance for Adverse Events Following Immunization Using VAERS. Manual for Surveillance of VPDs. 2024. CDC →
CDC operational guide documenting VAERS strengths and limitations including underreporting and no causal proof.
Abara WE, et al. Expected Rates of Select Adverse Events After Immunization. J Infect Dis. 2022;225(9):1569–1578. OUP →
Calculated expected background rates to contextualize post-vaccination adverse events.
Shimabukuro TT, et al. Safety Monitoring in VAERS. Vaccine. 2015;33(36):4398–4405. PubMed →
Comprehensive overview of VAERS methodology, strengths, and limitations.
McNeil MM, et al. The Vaccine Safety Datalink. Vaccine. 2014;32(42):5390–5398. PMC →
VSD methodology for active surveillance and hypothesis testing.
Hill AB. The Environment and Disease: Association or Causation? Proc Royal Soc Med. 1965;58:295–300. PMC →
Original source of the nine viewpoints for evaluating causal associations.
For the full 84-source causality research inventory, see the Primary Sources section of the Causality in Vaccine Safety page.