How well the Vaccine Adverse Event Reporting System captures adverse events, what the studies found, and what those findings mean
This page examines the reporting performance of the Vaccine Adverse Event Reporting System (VAERS) — how completely it captures adverse events, how sensitivity varies by event type, and why commonly cited reporting estimates require careful context.
This page is about reporting performance, not whether vaccines are safe overall and not whether any single VAERS report proves causation.
Last updated: April 2026.
VAERS is a national passive surveillance system operated jointly by the Centers for Disease Control and Prevention (CDC) and the U.S. Food and Drug Administration (FDA). It was established in 1990 under the National Childhood Vaccine Injury Act of 1986.
VAERS is designed to serve as an early-warning system. Its primary functions include detecting previously unrecognized adverse events following vaccination, identifying unusual increases in known adverse events, flagging possible lot-specific or product-specific safety concerns, and monitoring safety patterns after newly introduced vaccines enter the market (Zhou et al., 2003; Singleton et al., 1999).
Because VAERS accepts reports from healthcare providers, manufacturers, patients, and family members, it can capture events across a broader range of clinical settings and populations than any single hospital or health plan could monitor alone. However, as a passive system, VAERS depends on voluntary submission for most report types, which means its completeness varies (CDC Surveillance Manual, Ch. 21, 2024).
VAERS reporting is mandatory for vaccine manufacturers. It is also mandatory for healthcare providers in certain Vaccine Injury Table situations when a reportable event occurs within specified time windows. In other situations, reporting is voluntary.
In practice, this means that the completeness of VAERS data depends on several factors: whether a healthcare provider recognizes an event as potentially vaccine-related, whether the provider or patient takes the time to file a report, and whether the event is severe enough to prompt documentation. Serious or unusual events are more likely to be reported than mild or expected reactions (Rosenthal & Chen, 1995).
Periods of high public attention — such as the introduction of a new vaccine or widespread media coverage of a safety concern — can also increase reporting volume. During the COVID-19 vaccination campaign, VAERS received more reports in 2021–2022 than in the previous 30 years combined. CDC refers to this pattern as "stimulated reporting," which can temporarily inflate raw report counts without reflecting a change in the true rate of adverse events (CDC Surveillance Manual, Ch. 21, 2024).
Published estimates of VAERS reporting sensitivity vary widely depending on the event being studied, the study design, the time period, and the reporting environment. No single percentage describes all VAERS reporting.
In general, serious events — such as anaphylaxis, Guillain-Barré syndrome (GBS), and intussusception — are captured at higher rates than mild events such as injection-site soreness or low-grade fever. For serious events, published estimates range from roughly 12% to 76% depending on the condition and comparison data source. For mild events, estimates fall below 1% (Rosenthal & Chen, 1995; Shimabukuro et al., 2020).
This variation means that any single underreporting figure applied broadly to "all VAERS reporting" oversimplifies what the literature actually shows. Reporting completeness is better understood as a range that shifts with the type, severity, and visibility of the event in question.
The following table summarizes published estimates of how completely VAERS captures selected adverse events. These studies are not directly comparable to one another, because they examined different outcomes in different reporting environments. Some studies used capture-recapture methods, which estimate how many cases may have been missed by comparing overlapping reporting sources. Readers should note that these estimates cannot be combined into a single universal VAERS reporting rate, because each study measured a different event under different conditions.
| Study | Event / Context | Reported Sensitivity | What This Likely Reflects | Main Caution |
|---|---|---|---|---|
| Rosenthal & Chen, 1995 | Vaccine-associated paralytic polio (OPV) | ~68% | Serious, clinically distinct events are reported at higher rates | OPV is no longer used in the U.S.; modern vaccines may have different reporting patterns |
| Rosenthal & Chen, 1995 | MMR-associated rash | <1% | Mild, self-limiting events are rarely reported to passive systems | Does not mean these events are unimportant — only that passive systems are not designed to capture them completely |
| Verstraeten et al., 2001 | Intussusception after rotavirus vaccine (RotaShield) | ~47% | Capture-recapture estimate for a serious GI event during active surveillance period | Reporting was heightened by media coverage and regulatory attention around RotaShield withdrawal |
| Shimabukuro et al., 2020 | Anaphylaxis (multiple vaccines) | 13–76% | Range reflects variation across vaccines and comparison data sources | Higher estimates came from vaccines with more established reporting infrastructure |
| Shimabukuro et al., 2020 | Guillain-Barré syndrome (multiple vaccines) | 12–64% | GBS is serious and clinically recognizable, but reporting still varies | Sensitivity depends on how the comparison dataset defined and identified GBS cases |
These estimates reinforce a consistent pattern: VAERS captures a meaningful share of serious adverse events, but its completeness for any given event depends on severity, clinical recognizability, public awareness, and the reporting environment at the time.
The phrase "only 1% of adverse events are reported" appears frequently in public discussions of VAERS. However, this figure does not come from a single VAERS-specific study. It draws on at least three distinct sources, each measuring something different. Presenting them as interchangeable overstates what any individual source actually found.
| Source | Year | Product Context | What Was Actually Measured | Exact or Near-Exact Claim | Why It Is Often Overgeneralized |
|---|---|---|---|---|---|
| Kessler, JAMA | 1993 | Drugs and devices reported to FDA (MedWatch system) | Reporting rate for serious adverse drug events to the FDA's MedWatch system | "Only about 1% of serious events are reported to the FDA" | The article introduced MedWatch for drugs and devices. It explicitly directed vaccine reports to VAERS, not MedWatch. Later retellings sometimes apply this figure directly to VAERS vaccine reporting, which broadens the original source beyond its stated scope. |
| Lazarus et al. / Harvard Pilgrim (AHRQ grant) | 2010 | Vaccines administered at Harvard Pilgrim Health Care | Automated detection of possible adverse events following vaccination, compared with actual VAERS reports filed | "Fewer than 1% of vaccine adverse events are reported" | This figure included all detected events — mild reactions such as injection-site soreness alongside serious events. The project did not complete its planned system-performance comparison with VAERS because CDC declined to collaborate on the evaluation phase. The report is a grant final report, not a peer-reviewed journal article. |
| Hazell & Shakir, Drug Safety | 2006 | Adverse drug reactions (37 studies, systematic review) | Median underreporting rate for adverse drug reactions across passive surveillance systems | Median underreporting rate of 94% (i.e., approximately 6% reported) | This was a systematic review of drug adverse reaction reporting, not a vaccine-specific or VAERS-specific estimate. It is sometimes cited alongside the other two sources as though all three measured the same thing. |
Each of these sources addresses a real aspect of passive surveillance limitations. The issue is not that they are wrong, but that they are often cited more broadly than the original source supports. When combined into a single "1%" claim applied to all VAERS vaccine reporting, the resulting figure blurs important distinctions between drugs and vaccines, serious and mild events, and different surveillance systems. These distinctions matter because VAERS is often asked to do more than passive reporting systems are designed to do.
VAERS can identify possible safety signals — unusual clusters of reports, unexpected adverse events, or patterns that suggest a vaccine-related concern may warrant further study. It can detect rare events that clinical trials were not large enough to identify, flag increases in known adverse events that may indicate a manufacturing or distribution problem, and provide early data on newly licensed vaccines entering widespread use. These capabilities have led to several documented safety actions, including the withdrawal of the RotaShield rotavirus vaccine in 1999 and the identification of myocarditis following mRNA COVID-19 vaccines in 2021 (Zhou et al., 2003; Shimabukuro et al., 2015).
VAERS reports alone cannot determine the true incidence of an adverse event in the vaccinated population, because the system lacks complete denominator data — that is, the total number of vaccine doses administered. VAERS reports also cannot establish that a vaccine caused a reported event, because a temporal association (Event X happened after Vaccine Y) is not the same as a causal relationship. Additionally, raw VAERS counts cannot reliably be used to compare the safety profiles of different vaccines without additional analysis, because reporting rates, clinical attention, and public awareness differ across vaccine types and time periods (FDA VAERS Q&A, 2024; CRS IN12486, 2025).
Despite its limitations as a passive system, VAERS has contributed to several important safety actions. The following examples illustrate cases where VAERS reporting played a role in identifying signals that were subsequently confirmed through follow-up investigation.
| Year | Vaccine / Event | VAERS Role | Outcome |
|---|---|---|---|
| 1999 | RotaShield (rotavirus) / Intussusception | VAERS reports flagged an unusual cluster of intussusception cases within weeks of the vaccine's introduction | RotaShield was withdrawn from the market. Capture-recapture analysis later estimated VAERS captured approximately 47% of cases (Verstraeten et al., 2001). |
| 2010 | Influenza vaccination / Febrile seizures in children | VAERS detected an increase in febrile seizure reports in young children who received influenza and pneumococcal vaccines together during the same visit | Follow-up studies assessed the signal and informed later clinical guidance. |
| 2021 | mRNA COVID-19 vaccines (Pfizer-BioNTech, Moderna) / Myocarditis | VAERS reports identified myocarditis cases concentrated in younger males after second doses | VSD and other active surveillance systems confirmed the association. ACIP updated clinical guidance and informed consent materials. |
| 2021 | Janssen (J&J) COVID-19 vaccine / Thrombosis with thrombocytopenia syndrome (TTS) | VAERS reports identified a rare clotting syndrome within days of reports being filed | CDC and FDA paused administration, conducted follow-up review, and updated risk-benefit guidance. |
| Ongoing | Hepatitis B vaccine / Alopecia | VAERS reports contributed to identification of a possible association between hepatitis B vaccination and hair loss; research identifies this as a rare, weak association requiring further study | Listed as a potential signal warranting continued monitoring. Illustrates how VAERS can surface associations that are not yet confirmed or refuted. |
These examples show how VAERS has contributed to signal detection in practice. In each case, however, the initial VAERS signal required confirmation through stronger follow-up methods before regulatory or clinical action was taken.
While these signal detections demonstrate the system's intended function, a 2023 investigative report in The BMJ identified significant operational limitations during high-volume periods, including understaffing and delays in processing and public record updates (Block, The BMJ, 2023).
VAERS is designed to function as the first stage in a broader vaccine safety monitoring infrastructure, not as a standalone system. When VAERS identifies a potential safety signal, federal agencies and researchers use additional systems and methods to determine whether the signal reflects a genuine safety concern.
The Vaccine Safety Datalink (VSD) is one of the primary verification tools. VSD is an active surveillance system that links vaccination records with medical records across multiple large health plans, covering approximately 12 million people annually. Because VSD includes both vaccination and outcome data, it can calculate rates, compare vaccinated and unvaccinated populations, and assess whether an observed signal is occurring more often than expected (Shimabukuro et al., 2015).
The Clinical Immunization Safety Assessment (CISA) Project provides expert clinical evaluation of individual cases and complex adverse events. CISA can conduct detailed medical-record review and offer case-level assessment that VAERS reporting alone cannot provide.
Beginning in 2024–2025, the CDC's Immunization (IZ) Gateway expanded more timely data sharing with the FDA on how many vaccine doses were administered. This improvement helps address one of VAERS's longstanding limitations — the missing denominator — by providing more accurate dose counts for a given period, which allows more meaningful interpretation of VAERS report volumes.
In addition to these systems, follow-up epidemiologic studies, medical-record reviews, and formal risk-benefit analyses are used when a VAERS signal warrants deeper investigation. The combination of these methods means that a VAERS report is a starting point for investigation, not an endpoint.
VAERS is a passive early-warning system used in U.S. vaccine safety monitoring. It has contributed to the identification of several confirmed safety signals, including the withdrawal of RotaShield and the recognition of myocarditis following mRNA COVID-19 vaccines.
At the same time, VAERS has well-documented limitations. In published studies of selected events, reporting completeness has ranged from under 1% for some mild events to as high as 76% for some serious events. Raw VAERS data cannot establish causation, measure true incidence, or replace follow-up investigation.
The literature most consistently describes VAERS as useful for early signal detection, but limited as a stand-alone system for measuring true adverse-event rates or proving that a vaccine caused a specific outcome (Zhou et al., 2003; CDC Surveillance Manual, 2024; FDA VAERS Q&A, 2024; CRS IN12486, 2025).
Zhou W, Pool V, Iskander JK, et al. Surveillance for Safety After Immunization: Vaccine Adverse Event Reporting System (VAERS) — United States, 1991–2001. MMWR Surveill Summ. 2003;52(1):1–24. CDC MMWR →
Foundational federal summary of VAERS purpose, methodology, strengths, and limitations over the system's first decade.
Centers for Disease Control and Prevention. Manual for the Surveillance of Vaccine-Preventable Diseases, Chapter 21: Surveillance for Adverse Events Following Immunization Using VAERS. Updated 2024. CDC Surveillance Manual →
Current CDC operational guide documenting VAERS strengths (national scope, timeliness, rare-event detection) and limitations (underreporting, stimulated reporting, no causal proof).
U.S. Food and Drug Administration. Vaccine Adverse Event Reporting System (VAERS) Questions and Answers. Updated 2024. FDA VAERS Q&A →
FDA guidance stating VAERS reports generally cannot determine whether a vaccine caused or contributed to an adverse event.
Congressional Research Service. The Vaccine Adverse Events Reporting System: Understanding the Data. CRS Insight IN12486. 2025. CRS Report →
Neutral government summary explaining VAERS as an early-warning system not designed to determine causation on its own.
Kessler DA. Introducing MEDWatch: A New Approach to Reporting Medication and Device Adverse Effects and Product Problems. JAMA. 1993;269(21):2765–2768. FDA Download →
Source of the frequently cited "only about 1% of serious events are reported" figure. Article addressed the MedWatch system for drugs and devices; vaccine reporting was directed to VAERS.
Scott HD, Rosenbaum SE, Waters WJ, et al. Rhode Island Physicians' Recognition and Reporting of Adverse Drug Reactions. RI Med J. 1987;70:311–316. PMID: 3476980. PubMed →
Original source for the "1% reporting" figure later cited by Kessler. Study of drug (not vaccine) adverse reaction reporting by Rhode Island physicians.
Lazarus R, Klompas M, et al. Electronic Support for Public Health — Vaccine Adverse Event Reporting System (ESP:VAERS) Final Report. AHRQ Grant R18 HS017045. 2010. AHRQ PDF →
Source of the "fewer than 1% of vaccine adverse events are reported" figure. Grant final report, not peer-reviewed. Included all detected events (mild and serious).
Rosenthal S, Chen R. The Reporting Sensitivities of Two Passive Surveillance Systems for Vaccine Adverse Events. Am J Public Health. 1995;85(12):1706–1709. PMID: 7503352. PubMed →
First VAERS sensitivity study. Found 68% capture for vaccine-associated paralytic polio vs. less than 1% for MMR rash.
Verstraeten T, Baughman AL, Cadwell B, et al. Enhancing Vaccine Safety Surveillance: A Capture-Recapture Analysis of Intussusception After Rotavirus Vaccination. Am J Epidemiol. 2001;154(11):1006–1012. PMID: 11724716. PubMed →
Estimated VAERS captured approximately 47% of intussusception cases after RotaShield using capture-recapture methodology.
Shimabukuro TT, Nguyen M, Martin D, DeStefano F. Safety Monitoring in the Vaccine Adverse Event Reporting System (VAERS). Vaccine. 2015;33(36):4398–4405. PMID: 26209838. PubMed →
Comprehensive overview of VAERS methodology, strengths, limitations, and role in U.S. vaccine safety monitoring.
Shimabukuro TT, Miller ER, et al. The Reporting Sensitivity of the Vaccine Adverse Event Reporting System (VAERS) for Anaphylaxis and for Guillain-Barré Syndrome. Vaccine. 2020;38(47):7458–7463. PMID: 33039207. PubMed →
Most comprehensive VAERS sensitivity analysis to date. Found 13–76% capture for anaphylaxis, 12–64% for GBS across multiple vaccines.
Hazell L, Shakir SAW. Under-Reporting of Adverse Drug Reactions: A Systematic Review. Drug Safety. 2006;29(5):385–396. PMID: 16689555. PubMed →
Systematic review of 37 studies finding 94% median underreporting rate for adverse drug reactions. Drug-focused, not vaccine-specific.
Varricchio F, Iskander J, DeStefano F, et al. Understanding Vaccine Safety Information from the Vaccine Adverse Event Reporting System. Pediatr Infect Dis J. 2004;23(4):287–294. PMID: 15071280. PubMed →
Key interpretive guide describing VAERS as a signal-detection and hypothesis-generation system, not a causality tool.
Iskander JK, Miller ER, Chen RT. The Role of the Vaccine Adverse Event Reporting System (VAERS) in Monitoring Vaccine Safety. Pediatr Ann. 2004;33(9):599–606. PMID: 15462575. PubMed →
Presents VAERS as the front line of vaccine safety monitoring within a broader signal-detection framework.
Loughlin AM, Marchant CD, Adams W, et al. Causality Assessment of Adverse Events Reported to the Vaccine Adverse Event Reporting System (VAERS). Vaccine. 2012;30(50):7253–7259. PMID: 23063829. PubMed →
Found only 3% of reviewed VAERS reports classified as definitely causally related to vaccination. Most were classified as unrelated, unlikely, or possibly related.
Shimabukuro TT, Kim SY, Myers TR, et al. Safety of mRNA Vaccines Administered During the Initial 6 Months of the US COVID-19 Vaccination Programme. Lancet Infect Dis. 2022;22(6):802–812. Lancet →
Analysis of 298 million mRNA vaccine doses acknowledging VAERS limitations including underreporting and variable reporting as inherent to passive surveillance.
CDC/FDA. Guide to Interpreting VAERS Data. HHS.gov (VAERS). 2025. VAERS Data Guide →
Official guidance stating underreporting is one of the main limitations and VAERS receives only a small fraction of actual adverse events.
CDC. Chapter 21: Surveillance for Adverse Events Following Immunization Using VAERS. Manual for Surveillance of VPDs. 2026. CDC Surveillance Manual →
Documents underreporting causes, biased reporting, stimulated reporting, and system limitations.
Block J. Is the US Reporting System for Vaccine Safety Broken? The BMJ. 2023;380:p556. The BMJ →
Investigative report raising concerns about VAERS staffing, responsiveness, and potential missed signals during the COVID era. Operational critique, not an epidemiological study.
Johns Hopkins Bloomberg School of Public Health. What VAERS Is (And Isn't). 2022. JHU Public Health →
Plain-language explainer emphasizing that VAERS is public, searchable, and unverified, and that entries are often assumed to be vetted or causal.
Congressional Research Service. The ACA Preventive Services Coverage Requirement. CRS In Focus IF13010. 2025. CRS Report →
Explains ACA Section 2713 vaccine coverage requirements. Cross-reference for policy context, not VAERS-specific.