VAERS, the Vaccine Adverse Event Reporting System, is a national database that collects reports of health problems that occur after vaccination in the United States. It is co-managed by the CDC and the FDA and has been operating since 1990. The system serves as an early warning tool, helping federal agencies spot potential safety concerns with vaccines, but it does not prove that any vaccine caused a reported health problem.
How VAERS Works
VAERS is what’s known as a passive surveillance system. That means no one is actively tracking vaccinated people to see what happens to them. Instead, the system relies entirely on individuals choosing to submit a report. If you experience a health issue after getting a vaccine, you or someone aware of your situation can file a report through the VAERS website. No proof that the vaccine caused the problem is required.
Anyone can submit a report: patients, parents, caregivers, and healthcare providers. Healthcare providers are actually required by federal regulation to report certain serious adverse events, but for everyone else, reporting is voluntary. This open design casts a wide net, which is the point. By accepting all reports without filtering, VAERS increases the chances of catching a rare problem that might otherwise go unnoticed.
What Happens After a Report Is Filed
Each report is classified as either serious or non-serious. A report counts as serious if the health problem led to hospitalization, a life-threatening condition, lasting disability, a visit to the emergency room, or death. Everything else falls into the non-serious category, which includes things like injection-site soreness, mild fever, or fatigue.
CDC and FDA staff review incoming reports looking for what they call “signals.” A signal is a pattern in the data suggesting a possible new link between a vaccine and a health event, or a new aspect of a known link. Finding a signal doesn’t mean there’s a confirmed problem. It means something in the numbers warrants a closer look. The agencies then use more rigorous tools, like the Vaccine Safety Datalink, which actively tracks millions of patients in real time, to investigate whether the signal holds up under stricter analysis.
A Real Example: The RotaShield Vaccine
One of the clearest examples of VAERS working as intended involves RotaShield, a rotavirus vaccine licensed in August 1998. Soon after it hit the market, VAERS reports began showing cases of intussusception, a painful condition where part of the intestine folds into itself, in infants who had recently received the vaccine. The CDC moved quickly, suspending use of the vaccine and launching two emergency investigations. By October 1999, the advisory committee withdrew its recommendation, and the manufacturer pulled RotaShield from the market voluntarily. The entire process, from the first reports to withdrawal, took roughly 14 months.
Why VAERS Cannot Prove Cause and Effect
This is the single most important thing to understand about VAERS, and the point most often misunderstood. The system’s own data guide states it plainly: “No cause-and-effect relationship has been established” for any reported event. A VAERS report confirms only that a health problem happened sometime after a vaccination. It does not confirm the vaccine caused it.
Think of it this way. Millions of people get vaccinated every year. Some of those people will, by pure coincidence, develop an unrelated health issue in the days or weeks that follow. They might have a heart attack, get diagnosed with a new condition, or experience a seizure that was going to happen regardless. Because VAERS accepts all reports without judging whether the vaccine played any role, these coincidental events end up in the database alongside events that may genuinely be vaccine-related.
The biggest structural limitation of passive surveillance is the missing denominator. VAERS knows how many reports came in, but it doesn’t know the total number of people who were vaccinated without incident. Without that denominator, you can’t calculate a true rate of any given problem. Active surveillance systems do have that denominator, which is why they’re used to follow up on VAERS signals.
Reports can also be incomplete, contain errors, or reflect events that were never medically verified. Because anyone can file a report with no documentation required, the quality of individual entries varies widely.
Accessing VAERS Data
VAERS data is publicly available and searchable through the CDC’s WONDER database at wonder.cdc.gov. The database contains reports going back to 1990. In May 2025, the CDC and FDA expanded public access to provide more detailed information for all reported events. You can search by vaccine type, reported symptom, age group, and other filters. You can also look up a specific report by its VAERS ID number. The CDC provides both written instructions and video tutorials on its YouTube channel for navigating the system.
If you do explore the data, keep the limitations firmly in mind. Raw VAERS numbers, pulled without context, can be deeply misleading. A count of reports associated with a particular vaccine tells you how many people filed paperwork. It does not tell you how many people were actually harmed by that vaccine. Drawing conclusions from raw report counts is a common mistake, and it has fueled significant misinformation, particularly during periods of high vaccination volume when the total number of reports naturally rises alongside the number of doses given.
Where VAERS Fits in the Larger System
VAERS is not the only vaccine safety monitoring tool in the United States. It’s the front door, designed to detect early signals, but it works alongside several more precise systems. The Vaccine Safety Datalink uses electronic health records from multiple large healthcare organizations to run rapid, controlled analyses. The Clinical Immunization Safety Assessment Project connects the CDC with vaccine safety experts at medical research centers for in-depth clinical evaluations of individual cases.
Together, these systems form a layered approach. VAERS catches the initial hint that something might be off. The other systems determine whether that hint reflects a real safety issue or a statistical coincidence. No single system does the whole job on its own, and VAERS was never designed to be the final word on whether a vaccine is safe or dangerous. It’s the starting line, not the finish line.