U.S. flag An official website of the United States government

On Oct. 1, 2024, the FDA began implementing a reorganization impacting many parts of the agency. We are in the process of updating FDA.gov content to reflect these changes.

  1. Home
  2. Vaccines, Blood & Biologics
  3. Safety & Availability (Biologics)
  4. COVID-19 Vaccine Safety Surveillance
  1. Safety & Availability (Biologics)

COVID-19 Vaccine Safety Surveillance

January 10, 2025

The Center for Biologics Evaluation and Research (CBER) at the FDA is monitoring the safety of approved and authorized COVID-19 vaccines through both passive and active safety surveillance systems. CBER is doing so in collaboration with the Centers for Disease Control and Prevention (CDC), the Centers for Medicare and Medicaid Services (CMS), the Department of Veterans Affairs (VA), and other academic and large non-government healthcare data systems. In addition, CBER participates actively in ongoing international pharmacovigilance efforts, including those organized by the International Coalition of Medicines Regulatory Authorities (ICMRA) and the World Health Organization (WHO). These efforts are in addition to the pharmacovigilance efforts being undertaken by the individual manufacturers of approved and authorized vaccines. A coordinated and overlapping approach using state-of the art technologies has been implemented to maximize the ability to detect potential safety concerns.


Passive Surveillance
Active Surveillance
BEST
CMS
Summaries of Monitoring Efforts
 

Passive Surveillance

Passive surveillance is defined as unsolicited reports of adverse events (possible side effects) that are sent to a central database or health authority. In the United States, these are received and entered into the Vaccine Adverse Event Reporting System (VAERS), a national vaccine safety surveillance system that is co-managed by FDA and CDC. VAERS collects and analyzes information from reports of adverse events that occur after the administration of US licensed and authorized vaccines. Reports are received from patients, caregivers, health care providers, pharmacists, and vaccine manufacturers. Reporting adverse events by patients, caregivers, or providers is generally voluntary, but it can be required under certain circumstances regardless of whether the reporter or provider thinks the event is related to the vaccine; for example, the COVID-19 vaccine emergency use authorizations (EUAs) require providers to report all serious adverse events following vaccination, regardless of whether providers believed the event was related to the vaccine.i Initial reports made to the VAERS system are subsequently further investigated to determine whether they are likely to be related to vaccine administration. A report to VAERS does not mean the vaccine caused the event. Some VAERS reports might represent true vaccine reactions or side effects; others are coincidental events not related to vaccination.

FDA and CDC continually monitor incoming and accumulated VAERS data using various methods, including frequent assessment of report counts and trends, comparisons of current versus historical data, and reporting trends over time, as well as evaluating individual reports, as needed, to identify possible safety concerns associated with vaccines. These reports are used to monitor the occurrence of both known and unknown adverse events. Once adverse events of concern are identified in VAERS, they may be monitored in other immunization safety systems to evaluate whether the event occurs more frequently than expected after vaccination or to conduct more controlled scientific studies to confirm whether a particular adverse event is related to a vaccine. An example of the work done with passive safety surveillance during the current pandemic has been the evaluation of severe allergic reactions, such as anaphylaxis, and myocarditis or pericarditis as serious types of adverse events following vaccination with COVID-19 vaccines. Through this work, we have come to understand that these reactions are quite rare, happening in less than 1 in 200,000 vaccinated individuals, and we were able to provide additional information to patients and providers about individuals who might be at higher risk for these adverse events.

CBER also uses disproportionality analysis, also known as data mining, in its vaccine safety monitoring efforts. Data mining refers to the use of statistical or mathematical tools to discover patterns of associations or unexpected occurrences in large databases, such as VAERS for vaccines. Data mining involves the automated analysis of adverse events and vaccine administration errors report data in VAERS to inform FDA staff about a potential disproportionality of reports of a certain adverse event or vaccine administration error for one vaccine relative to other vaccines in the VAERS database.ii

Data mining results are evaluated by FDA staff and may inform FDA staff about specific adverse events or administration errors that may warrant additional review to determine whether there is a safety concern associated with vaccination. Importantly, data mining results may provide insights into the patterns of adverse events or errors reported for a given product relative to other vaccine products, but data mining results do not, by themselves, demonstrate causal associations between a vaccine and an adverse event. Note that data mining results are derived exclusively from ‘‘numerator’’ data (i.e., numbers of adverse events or errors passively reported to VAERS); as a passive, numerator-only system, VAERS lacks information on the total number of individuals vaccinated (i.e., the “denominator”) and, as such, cannot provide or compare rates of adverse events after vaccination.

The limitations of data mining include but are not limited to the following:

  • Results may be impacted by stimulated reporting (for example, the use of v-safe and media reporting around the COVID-19 vaccines);
  • There may be false alerts from statistical interactions (for example, if two vaccines are given at the same time, a potential statistical signal for one vaccine may be driven by the other vaccine); and
  • There may be underreporting, incomplete reporting, and duplicate reporting.

As part of the agencies’ surveillance efforts during the COVID-19 public health emergency, FDA and CDC exchanged raw data mining information about COVID-19 vaccines. Using data mining, FDA and CDC did not find any new signals that led to regulatory actions or changes in clinical recommendations. More information on these exchanges may be found in CBER’s Biologics Electronic Reading Room (eFOIA) | FDA. Because data mining results are preliminary and form only one part of CBER’s passive vaccine surveillance, CBER does not ordinarily make them public. However, given the extraordinary interest in the COVID-19 vaccines, CBER has decided to do so here.

Active Surveillance

Active surveillance involves proactively obtaining and rapidly analyzing information occurring in millions of individuals recorded in large healthcare data systems to verify safety signals identified through passive surveillance or to detect additional safety signals that may not have been reported as adverse events to passive surveillance systems. FDA is conducting active surveillance using the Sentinel BEST (Biologics Effectiveness and Safety) System and the CMS system. Our efforts complement those of the Vaccine Safety Datalink (VSD) and the v-safe text-based monitoring system for conducting surveillance of adverse events that CDC has implemented. FDA is also collaborating with other federal and non-federal partners.

BEST

To elaborate further, the BEST system, which is part of the Sentinel initiative, comprises large-scale claims data, electronic health records (EHR), and linked claims-EHR databases. The system makes use of multiple data sources and enables rapid queries to detect or evaluate adverse events as well as studies to answer specific safety questions for vaccines. The linked claims-EHR database makes it possible to study the safety of vaccines in sub-populations with pre-existing conditions or in pregnant women. The major partners for BEST currently are Acumen, IBM Federal HealthCare, IQVIA, and Columbia University and many affiliated partners such as MedStar Health, BlueCross BlueShield of America, the Observational Health Data Sciences and Informatics (OHDSI), OneFlorida, University of California, and several others. iii

Using BEST, CBER has monitored about 15 adverse events that were chosen based on experience with evaluating safety for the deployment of previous vaccines. CBER will use the BEST system to conduct more in-depth analyses should a safety concern be identified from sources such as VAERS.

CMS

CBER has worked over the past several years with CMS to develop capabilities for routine and time-sensitive assessments regarding the safety of vaccines for people 65 years of age and older using the Medicare Claims database. Because it was already in place, this system was immediately put into use for COVID-19 vaccine surveillance to monitor for adverse events.iv

In summary, in collaboration and coordination with several different partners, CBER has assembled passive and active surveillance systems that can detect and refine safety findings with the approved and authorized COVID-19 vaccines. These systems can also potentially be leveraged to assess safety in specific subpopulations and to assess vaccine effectiveness, including against emerging variants.

Summaries of Monitoring Efforts

FDA is conducting intensive monitoring of COVID-19 vaccine safety in the U.S. using a variety of approaches. Based on available information, FDA strongly believes that the known and potential benefits of COVID-19 vaccination greatly outweigh their known and potential risks. As part of our efforts to be transparent about our COVID-19 vaccine safety monitoring activities, FDA is posting summaries of the key safety monitoring findings on this page.


Notes:

i Serious adverse events are defined as: death; a life-threatening adverse event; inpatient hospitalization or prolongation of existing hospitalization; a persistent or significant incapacity or substantial disruption of the ability to conduct normal life functions; or a congenital anomaly/birth defect. For reporting requirements for vaccination providers for the COVID-19 vaccines authorized for emergency use, please see the Fact Sheets for Healthcare Providers Administering Vaccine, which may be accessed here and here.

ii Best Practices for FDA Staff in the Postmarketing Safety Surveillance of Human Drug and Biological Products

iii To confirm the utility of the BEST system for situations such as COVID-19 vaccine surveillance, a test case was conducted. This study aimed to replicate a previous study by the CDC’s Vaccine Safety Datalink (VSD) (Klein et al. Pediatrics 2010) that examined the databases and analytic capabilities of the new system. The objective of this study was to test the new system’s ability to reproduce the increased risk of febrile seizures in children receiving the first dose of measles-mumps-rubella-varicella (MMRV) vaccine, compared to that of MMR and varicella vaccines separately but on the same day. The results of the study met the objectives and demonstrated the ability of the BEST Initiative data network to run a complex study protocol at multiple sites using a distributed data network and the Observational Medical Outcomes Partnership Common Data Model (organizing disparate data sources into the same database design using a common format).

iv As one example of the capabilities of this system, FDA, CMS, and CDC evaluated the risk of Guillain-Barré syndrome (GBS) following influenza vaccination after CDC’s Vaccine Safety Datalink identified safety signals suggesting an increased risk of GBS following high-dose influenza vaccinations and Shingrix vaccinations during the 2018-2019 influenza season. CBER, CDC, and CMS formed working groups in February 2019 to refine these safety signals in the CMS data.


Back to Top