Keep Factually independent
Whether you agree or disagree with our analysis, these conversations matter for democracy. We don't take money from political groups - even a $5 donation helps us keep it that way.
What percentage of VAERS adverse events are serious or fatal?
Executive Summary
VAERS is a passive U.S. safety‑signal system; multiple analyses of VAERS documentation converge on the finding that roughly 10–15% of VAERS reports are classified as serious (including hospitalization, life‑threatening events, disability, or death), while about 85–90% report minor events. COVID‑era subanalyses and short windows can show higher shares of serious or fatal reports in specific populations or timeframes, but these do not overturn the broader CDC/VAERS characterization of the overall dataset [1] [2] [3].
1. Why the headline numbers (10–15%) keep appearing — and what they mean for readers
Public VAERS guidance and multiple secondary analyses state that about 85% of reports describe minor reactions and roughly 10–15% are classified as serious, a framing that appears in CDC technical notes cited across analyses (one source explicitly summarized this in materials linked to VAERS) [1] [2] [3]. This percentage is a classification of report contents, not an estimate of true incidence after vaccination; VAERS accepts any report and tallies descriptors such as hospitalization or death. The 10–15% figure is a snapshot of report severity labels, not a validated rate of vaccine‑caused serious outcomes. The 2015 VAERS overview underscores the system’s role in early detection and limitations—signal generation rather than causality determination [4].
2. COVID‑period analyses show higher shares in narrow slices — why that can mislead
Several studies and rapid‑response pieces during the COVID vaccine rollout flagged spikes in serious or fatal VAERS reports in short timeframes or specific subgroups; for example, a pediatric COVID‑vaccine analysis found about 17.9% of reported events in 5–17‑year‑olds were serious (no publication date in the analysis summary), and a BMJ rapid‑response noted that 27% of all fatal VAERS reports since 1990 were for COVID‑19 vaccines within a four‑month window (BMJ rapid‑response dated 2025‑09‑07 in the summary) [5] [2]. These snapshots reflect exceptional reporting volumes, heightened public scrutiny, and focused searches, and therefore inflate the share of serious reports within those narrow datasets versus the broader, multi‑decade VAERS corpus [4] [6].
3. Limitations of VAERS: passive reporting, duplicates, and unverified causality
VAERS is explicitly a passive system; anyone can file a report, and entries vary widely in completeness and verification. The 2015 VAERS monitoring overview explains that VAERS is designed for early signal detection and not for incidence estimation or causal attribution [4]. Multiple analysis summaries reiterate this limitation, noting underreporting in some cases and overreporting or clustering in others, which means percentages of serious or fatal entries in VAERS can reflect reporting behavior as much as clinical reality [6] [7]. Public and academic commentators have pointed to this structural limitation when interpreting elevated shares during the COVID rollout [2].
4. Different data slices, different stories: age, time window, and vaccine type matter
When analysts restrict VAERS to specific vaccines, age groups, or short intervals, the proportion of serious or fatal reports can rise. The pediatric COVID analysis that found 17.9% serious events in 5–17‑year‑olds is an example of how subgroup analysis yields higher percentages than the systemwide average [5]. Research that isolates death reports or myocarditis cases similarly produces different shares and must be read as context‑dependent, not representative of VAERS overall [8] [9]. The systemwide 10–15% serious‑report figure persists because it aggregates across years, vaccines, ages, and reporting behaviors [1] [3].
5. What reporters and policymakers should do with these percentages
The proper use of the VAERS serious‑report percentage is as an early warning and prioritization tool, not as a standalone claim of vaccine harm. Public health authorities use VAERS signals to trigger controlled epidemiologic studies or active surveillance that can assess incidence and causality; the 2015 system description highlights this workflow [4]. Analysts who publicize elevated shares in focused windows often aim to draw attention to potential signals, which can reflect legitimate safety surveillance priorities or advocacy agendas; readers should weigh both the intent and the methodological scope behind any given percentage [2] [9].
6. Bottom line: a consistent systemwide picture plus important caveats
Across the provided analyses and official VAERS descriptions, the consistent systemwide picture is that most VAERS reports describe minor reactions, and about 10–15% are labeled serious, while focused COVID‑era or subgroup analyses can show higher proportions in short windows [1] [2] [3] [5]. This duality—stable overall proportions paired with volatile subgroup signals—explains public confusion: VAERS percentages are informative for surveillance but insufficient alone to establish causality or population risk. Follow‑up using active surveillance and controlled epidemiologic methods is necessary to determine whether VAERS signals reflect true vaccine‑related increases in serious outcomes [4] [7].