Since its creation in 1990, the U.S. Centers for Disease Control and Prevention’s (CDC) Vaccine Adverse Event Reporting System (VAERS) has provided publicly available data on any adverse reactions that have been reported following a vaccination.
Since the arrival of COVID-19 vaccines, information contained in the VAERS system has been frequently referenced by those questioning vaccine safety.
The VAERS website includes disclaimers that the reports may contain “incomplete, inaccurate, coincidental, or unverifiable” information and requires users acknowledge these limitations. However, many of the social media posts encountered by Reuters Fact Check have failed to mention that reports in VAERS can be entered by anybody and are not deemed to show a causal connection to a vaccine until verified by the CDC.
From 1990 to 2001, VAERS data was only accessible through a Freedom of Information Act (FOIA, www.foia.gov/ ). In 2001, the data was made available on the VAERS website (vaers.hhs.gov/data.html), and later moved to CDC Wonder (wonder.cdc.gov/vaers.html) in 2006, where it is still hosted. Submissions for VAERS reports remain via vaers.hss.gov.
Martha Sharan, a spokesperson for the CDC, told Reuters via email, “Transparency was always a concern with VAERS, and in that spirit, the decision was made to make VAERS data publicly accessible at the earliest stages”.
When an adverse event report is filed into VAERS, it is processed and reviewed by certified coders, according to Sharan. The report receives a code based on the description given by the individual that filed it and enters a database made available to the CDC and FDA, which is updated daily.
Each report is labeled as a serious or non-serious adverse event. Serious events are defined as “death, life threatening illness, hospitalization or prolongation of hospitalization, permanent disability, congenital anomalies, or birth defects” by the code of Federal Regulation (here).
The CDC requests and reviews medical records, including hospital records, clinic records, death certificates, and autopsy reports, for each report labeled as “serious.”
VAERS staff do not decide if the vaccine caused the adverse event. To determine if there is a potential safety issue signal for a vaccine, statistical methods are used.
“If a safety concern is detected by VAERS, more detailed, quantitative analyses are conducted using more robust analytic systems, such as the Vaccine Safety Datalink,” Sharan explained. “Such analyses can determine if a statistically meaningful association exists between the vaccine and the adverse event in question.”
When a safety issue is recognized for the vaccine, the FDA and vaccine manufacturer work together to find a solution depending on the safety concern, whether it be an issue with a specific lot of vaccines, manufacturing, or the vaccine itself.
An FDA spokesperson told Reuters via email about CDC’s Vaccine Safety Datalink (VSD) or the Clinical Immunization Safety Assessment (CISA) project, saying: “These systems do not have the same limitations as VAERS and are used to help assess health risks and possible connections between adverse events and a vaccine.”
More information about these systems can be found here .
For example, “In the case of anaphylaxis, thrombosis with thrombocytopenia syndrome and GBS following Janssen COVID-19 Vaccine and myocarditis/pericarditis following Pfizer-BioNTech and Moderna COVID-19 Vaccines, FDA has worked with the companies to update information in the Fact Sheet for Health Care Providers Administering Vaccine (Vaccination Providers) and the Fact Sheet for Recipients and Caregivers for each of the vaccines with information about these potential adverse events,” FDA said.
The ability for anybody to submit a report into VAERS is both an important part of its transparency, and a limitation. Reuters has reviewed numerous examples of unverified information that have been shared widely on social media – such as story about a two-year-old who had allegedly died during Pfizer-BioNTech’s COVID-19 vaccine trial in children (here).
Participants in Pfizer’s clinical trial, which included children as young as six months, did not start until a month after the two-year-old child in Virginia allegedly received Pfizer’s COVID-19 vaccine on February 25, making it impossible for this child to have received a dose. A CDC spokesperson told Reuters via phone that the report was consequently removed from the database.
Dr. Dan Salmon, director of the Institute for Vaccine Safety at the Johns Hopkins Bloomberg School of Public Health, told Reuters via email about the risks with anyone being able to report to VAERS.
“If I got a flu vaccine and dog got hit by a car, I could report that to VAERS and it would end up on a publicly available database,” Salmon said.
The data being misinterpreted and has led to medical misinformation being shared widely online.
For example, Fox News host Tucker Carlson said in a broadcast (here) on May 5, 2021 that over 3,362 Americans had died from the COVID-19 vaccine between late December of 2020 and April 2021. But he was referring to VAERS reports and not deaths verified by the CDC.
Dr. Robert Leo Murphy, executive director of the Institute for Global Health at Northwestern University, told Reuters via phone call that users must be careful with interpreting raw data, as it takes investigation to determine causality.
Dr. Stephen Kissler, a research fellow at the department of immunology and infectious disease at Harvard T.H. Chan School of Public Health, said in a media call (here) that data accessibility for the public is very important.
“A knee jerk reaction might be to say that well, maybe we shouldn’t be making these data as publicly available as they are, or maybe they shouldn’t be as detailed, but I really hesitate to go that direction because I really do think that empowering people to make data-based decisions about their health and about community health, making data accessible is really important,” Kissler said.
Kissler said he believed VAERS to be extremely beneficial for the medical community to prompt research into possible links between the vaccine and adverse events that can be hinted at by the database.
According to Murphy, VAERS can be helpful for reporting adverse events that may have not surfaced or have been noticed during clinical studies. (Inevitably, a clinical trial will touch a smaller sample of individuals than a nationwide vaccination campaign, for example. A system like VAERS can help pick up on any other potential side effects.)
Giving members of the public the opportunity to view all user reports is both a strength and challenge of the system, according to Dr. Christin Gilmer, a global health scientist. Possible improvements to the database could include watermarking information, debunking false entries before they become viral, and requiring users to complete a training before accessing the data.
“For someone who is just looking at the data, who may not be aware of these caveats about using the data that they may not represent a causal link between vaccination and health effects, it can be very alarming, and it can seem like the vaccine is causing all of these different kinds of health effects,” Kissler said.
“Currently, it’s possible to download that data from VAERS and look at adverse reactions that followed vaccination but that doesn’t put it into the context of how frequent those adverse reactions are in the general population,” Kissler said. “That sort of thing is what I would like to see.”
“We must find a balance between being transparent and being proactive in preventing VAERS reports from being the cause of people avoiding vaccines because of misinformation,” Gilmer said.
This article was produced by the Reuters Fact Check team. Read more about our fact-checking work here .
Our Standards: The Thomson Reuters Trust Principles.