How VAERS Tracks Vaccine Adverse Events in the US

A parent in Ohio reports fever and a rash after their child's vaccine. A pharmacist in Texas logs a fainting episode. A nurse in Florida files a note about an allergic reaction. All three accounts land in the same federal database within hours, sorted by vaccine type, symptom, and patient age. The Vaccine Adverse Event Reporting System, known as VAERS, has collected more than a million reports since its launch in 1990. Yet many people who search for vaccine safety data at night still wonder whether the numbers they see represent confirmed harm or simply coincidence recorded on paper.

VAERS is a passive surveillance tool, meaning it accepts any report from any source without requiring proof that the vaccine caused the event. That openness is both its strength and its limit. Public health teams scan the stream for patterns that stand out against expected background rates of illness. When a signal appears, regulators launch formal studies to confirm or rule out a causal link. Understanding how reports flow through the system, how analysts separate noise from signal, and what happens after a flag is raised can clarify what VAERS data do and do not prove about vaccine safety.

Who can file a report and what information goes in

Anyone may submit a VAERS report. Healthcare providers are required by law to report certain events listed on manufacturer package inserts and any serious outcome that follows immunization. Patients, parents, and vaccine manufacturers also file thousands of reports each year. The online form asks for vaccine name and lot number, the date of vaccination, the date symptoms began, a description of the event, medical history, and whether the person recovered.

No medical record is required at the time of filing. A parent who notices a high fever the evening after a dose can log that observation even if no doctor visit occurred. A clinician who sees an unexpected seizure will file a detailed note. Manufacturers must forward any adverse event reported to them, including those mentioned on social media or in customer service calls. This wide net captures rare signals that might be missed in controlled trials, but it also means the database includes events that occurred after vaccination without being caused by it.

Each report receives a unique identification number and enters a shared database managed by the Centers for Disease Control and Prevention and the Food and Drug Administration. Staff review incoming submissions for completeness and follow up by phone or letter when key details are missing. Duplicate reports are flagged. The system does not verify diagnoses or investigate causality at the point of entry. That work happens in the next stage.

How safety signals are detected in the data stream

Analysts run automated algorithms every week to compare observed report rates against expected background rates of medical events in the general population. If the number of reports for a specific symptom and vaccine combination crosses a statistical threshold, the system generates a signal. For example, if ten cases of a rare heart condition appear within two weeks of a newly authorized vaccine when only one or two would be expected by chance, the algorithm flags the cluster.

Manual review follows every automated alert. Epidemiologists read individual case narratives, check patient ages and medical histories, and look for alternative explanations such as concurrent infections or pre-existing conditions. They compare the timing of symptom onset to the known biological window in which a vaccine effect would be plausible. A rash that starts six weeks after a dose is less likely to be vaccine-related than one that appears within three days.

Not every signal becomes a confirmed safety issue. Many clusters dissolve under closer scrutiny. A spike in reports may reflect media coverage that prompts more people to file rather than a true increase in events. Seasonal illness patterns can create false associations. The review process filters out these artifacts before any public alert is issued.

What happens when a signal triggers a formal investigation

When a signal survives initial review, regulators launch studies that link VAERS data to other surveillance systems with verified medical records. The Vaccine Safety Datalink, a network of integrated healthcare organizations covering more than twelve million people, provides detailed electronic health data that can confirm diagnoses and establish timelines. The Clinical Immunization Safety Assessment Project recruits specialists to evaluate complex cases in hospital settings.

Investigators calculate incidence rates among vaccinated and unvaccinated groups, adjusting for age, sex, and underlying health conditions. They use case-control designs to compare vaccination history between patients who experienced the event and matched controls who did not. If the adjusted risk ratio remains elevated and the biological mechanism is plausible, the finding may lead to updated product labeling, revised clinical guidance, or in rare cases a recommendation to pause use while additional data are gathered.

One example unfolded in 2021 when VAERS reports flagged an unusual number of myocarditis cases in young men after mRNA COVID-19 vaccines. Within weeks, the CDC convened expert panels, analyzed medical charts from the Vaccine Safety Datalink, and confirmed a small but real increase in risk, particularly after the second dose. The finding led to updated fact sheets, closer monitoring protocols, and public communication about symptoms to watch. The vaccines remained recommended because the benefit of preventing severe COVID-19 far outweighed the low risk of typically mild and self-limiting heart inflammation.

Why raw VAERS numbers cannot prove causation

A report in VAERS documents a medical event that occurred after vaccination. It does not establish that the vaccine caused the event. People experience illness, injury, and death at baseline rates every day. When millions receive vaccines in a short window, some will coincidentally suffer unrelated events soon afterward. A person who has a stroke the day after a flu shot may have been on the verge of a stroke regardless of vaccination.

The database includes reports of events that are biologically implausible or have been thoroughly investigated and ruled out. A search for a specific symptom will return every report mentioning that term, whether the reporter suspected a link or simply noted it for completeness. Public misinterpretation of raw counts has fueled misinformation campaigns that present VAERS totals as proof of widespread harm without acknowledging the lack of verification or comparison to background rates.

Researchers use VAERS as a hypothesis-generating tool, not an endpoint. A high number of reports prompts further study. The conclusion comes from controlled analyses that account for confounders and establish temporal and biological plausibility. VAERS data alone cannot answer the question of whether a vaccine caused a specific outcome.

How the system balances transparency and public understanding

VAERS data are publicly accessible through an online search tool. Anyone can download reports, filter by vaccine type or symptom, and view anonymized case summaries. This transparency supports independent research and public accountability. It also creates opportunity for misuse when raw numbers are shared without context or statistical training.

The CDC and FDA publish regular safety updates that interpret VAERS findings in light of background rates and follow-up investigations. These summaries explain which signals led to confirmed risks and which were ruled out. Educational materials emphasize that a report does not equal a proven side effect and that the system is designed to catch even very rare events that would be invisible in pre-licensure trials.

Balancing open access with accurate interpretation remains an ongoing challenge. Some advocacy groups cite VAERS totals to argue that vaccines are dangerous, while public health agencies stress that the same data, when analyzed properly, consistently show that serious adverse events are rare and benefits overwhelmingly outweigh risks. Clear communication about how passive surveillance works is essential to maintaining trust in both the system and the immunizations it monitors.

What other countries do and how systems compare

Many countries operate similar passive reporting systems. The United Kingdom's Yellow Card scheme, the European Medicines Agency's EudraVigilance, and Australia's Therapeutic Goods Administration database all collect spontaneous reports from healthcare workers and the public. Each has its own reporting thresholds, follow-up protocols, and data-sharing agreements with manufacturers and academic researchers.

Some nations complement passive surveillance with active monitoring that recruits cohorts of vaccinated individuals and contacts them at regular intervals to ask about health events. This approach reduces underreporting and provides denominator data for calculating true incidence rates. The World Health Organization coordinates international signal sharing so that a cluster detected in one region can prompt alerts elsewhere before widespread exposure occurs.

No system is perfect. Passive reporting underestimates common, mild events because people do not bother to file. It may overestimate rare, dramatic events that attract media attention and prompt multiple reports of the same case. Active surveillance is more accurate but expensive and limited in scale. Most countries use a combination of methods, with VAERS-style databases serving as the early-warning layer that triggers deeper investigation when needed.

Key facts

• VAERS has collected more than one million reports since 1990, covering all vaccines licensed in the United States.
• Healthcare providers are legally required to report certain events specified in vaccine package inserts and any serious adverse event.
• Automated algorithms scan incoming reports weekly to detect statistical signals that exceed expected background rates.
• The Vaccine Safety Datalink links VAERS signals to verified electronic health records from more than twelve million people.
• Most VAERS reports describe mild, temporary events such as fever or injection-site soreness; serious events are rare.

Can I file a VAERS report myself?

Yes. The online form is open to patients, parents, and any member of the public. You do not need a doctor's signature or medical records to submit a report, though providing as much detail as possible helps analysts evaluate the event.

Does a high number of VAERS reports mean a vaccine is unsafe?

Not on its own. A large number of reports may reflect high vaccination rates, increased media attention, or reporting campaigns rather than a true safety problem. Regulators compare report rates to expected background illness rates and conduct follow-up studies before concluding that a vaccine caused the events.

What happens if I report an event that turns out to be unrelated to the vaccine?

The report remains in the database. VAERS is designed to capture all possible signals, even if most turn out to be coincidental. Analysts use statistical methods and follow-up investigations to separate true vaccine effects from background noise.

Related reading: About us · What we do