Question explored with the scientific record
What ratio of vaccine injury is captured
The short version: passive surveillance systems like VAERS capture only a tiny fraction of real vaccine injuries, and the evidence lets us estimate just how small that fraction is.
The best direct evidence comes from a 14-year prospective study that compared active surveillance (culturing every platelet unit before transfusion) against passive surveillance (waiting for clinicians to report a reaction) [10]. Active culture found 38 contaminated units out of 28,454 tested. Passive surveillance caught exactly 1 of those 38. That is a 2.6% capture rate. The same study found that Gram stain and pH testing, which are closer to how passive clinical reporting works, had sensitivities of 27% and 0% respectively [10]. In other words, for every injury passively reported, roughly 37 went undetected.
The measles outbreak study from the Netherlands tells a similar story for disease surveillance: only 9% of cases were reported by general practitioners to public health authorities [8]. The factors that predicted whether a case got reported were not the severity of the illness but whether the parent thought medical care was needed for respiratory infections [8]. That is selection bias, not medical triage.
For vaccine safety specifically, the VAERS system has known structural problems. A 2006 analysis found that by 2002, litigation-related reports made up the majority of VAERS reports for overdose, neuropathy, and thimerosal, and nearly half of mental retardation reports [7]. That means the signal in the database is shaped by who is suing, not by who is injured. The authors concluded that longitudinal VAERS analyses that ignore reporting source "will not portray accurately the trend that they imply in the data" [7].
| Surveillance method | True positives detected | Sensitivity | What it misses |
|---|---|---|---|
| Active culture (platelet study) | 38 of 38 contaminated units | 97.4% | Nearly nothing |
| Passive clinical reporting (same study) | 1 of 38 | 2.6% | 97% of events |
| Gram stain (same study) | 6 of 22 | 27% | Low-density contamination |
| pH testing (same study) | 0 of 5 | 0% | Everything |
| Passive measles surveillance (Netherlands) | 15 of 164 cases | 9% | Mild/moderate cases |
My call: the ratio of captured to true vaccine injury is somewhere between 1:10 and 1:40 based on the best available active-vs-passive comparison, and the evidence that exists is thin because the most informative study design (active surveillance of a vaccinated vs unvaccinated cohort) has almost never been run. Confidence: moderate for the capture ratio itself, low for generalizing it to every vaccine, because the platelet study is the only direct head-to-head comparison in the retrieved evidence.
Sources examined 20
-
The role of complement in long COVID pathogenesis
This JCI Insight review synthesizes evidence that dysregulated complement activation contributes to long COVID pathogenesis via endothelial injury and thromboinflammation, surveys testing/biomarker strategies, and discusses therapeutic targeting potential and limitations.
DOI: 10.1172/jci.insight.194314 -
Benefits of Inactivated Vaccine and Viral Vector Vaccine Immunization on COVID-19 Infection in Kidney Transplant Recipients
In a retrospective cohort of 45 kidney transplant recipients with COVID-19, vaccination with inactivated or viral vector vaccines was associated with significantly lower mortality (13% vs 36%; RR 0.56) and remained significant after adjustment (OR 0.54).
DOI: 10.3390/vaccines10040572 -
Cardiac injury in patients with pandemic 2009 influenza A (H1N1) infection
A retrospective, single-center study of 172 hospitalized patients with 2009 influenza A(H1N1) in Zhejiang, China, showing that severe/critically severe infection is associated with marked myocardial injury evidenced by elevated CK, CK-MB, hs-CRP, and cardiothoracic ratio, with w…
DOI: 10.1080/ac.66.4.2126589 -
Safety of anthrax vaccine: an expanded review and evaluation of adverse events reported to the Vaccine Adverse Event Reporting System (VAERS)
Review of 1841 VAERS reports covering ~2 million doses of anthrax vaccine (AVA) found no evidence of an unusual rate of serious or medically important adverse events attributable to AVA, supporting deltoid rather than triceps injection.
DOI: 10.1002/pds.936 -
Towards a Humanized Mouse Model of Liver Stage Malaria Using Ectopic Artificial Livers
Macroporous PEG cryogel-based human ectopic artificial livers (p-HEALs) enable liver-stage Plasmodium infection in vitro and in vivo, delivering a scalable, host-liver-injury-free humanized mouse model for malaria liver-stage research.
DOI: 10.1038/srep45424 -
Meningococcal conjugate vaccine safety surveillance in the Vaccine Safety Datalink using a tree‐temporal scan data mining method
Using the tree-temporal scan data mining method on 1,253,403 first doses of MenACWY-D in the Vaccine Safety Datalink, this pilot study detected expected adverse events (skin/subcutaneous tissue infections, fever, urticaria) and a likely false pleurisy signal, demonstrating the m…
DOI: 10.1002/pds.4397 -
Vaccine Adverse Event Reporting System Reporting Source: A Possible Source of Bias in Longitudinal Studies
Analysis of VAERS public-use data from 1990-2003 found a sharp rise in litigation-related reports, with most overdose, neuropathy, and thimerosal reports and many autism and mental retardation reports coming from legal sources, potentially biasing longitudinal VAERS analyses.
DOI: 10.1542/peds.2004-2687 -
Measles surveillance: the importance of finding the tip of the iceberg
A Dutch community measles outbreak study shows that only about 9% of measles cases are reported by general practitioners to municipal health services, illustrating that passive surveillance captures only the outbreak’s tip of the iceberg and underestimates true incidence.
DOI: 10.1016/s0140-6736(05)17715-0 -
Strengths and limitations of assessing influenza vaccine effectiveness using routinely collected, passive surveillance data in Ontario, Canada, 2007 to 2012: balancing efficiency versus quality
Routine Ontario passive influenza surveillance can generate vaccine effectiveness estimates via screening methods, but these estimates tend to underestimate true effectiveness compared with sentinel-network test-negative design due to incomplete immunisation data and related bia…
DOI: 10.2807/1560-7917.es2015.20.16.21100 -
Evolution of surveillance methods for detection of bacterial contamination of platelets in a university hospital, 1991 through 2004
A 14-year prospective surveillance study at University Hospitals of Cleveland evaluating active (culture/Gram stain/pH) versus passive methods for detection of bacterial contamination in platelet units, demonstrating that active at-issue culture is the most sensitive method, Gra…
DOI: 10.1111/j.1537-2995.2006.00790.x -
An Overview of COVID-19 Vaccine Safety and Post-marketing Surveillance Systems
This article provides a comprehensive review of COVID-19 vaccine safety, emphasizing pre- and post-marketing surveillance systems, methods for signal detection and causality assessment, risk management, and real-world safety data trends from international pharmacovigilance datab…
DOI: 10.24926/iip.v12i4.4294 -
A nested real-time PCR assay for the quantification of Plasmodium falciparum DNA extracted from dried blood spots
This study develops and validates a nested real-time PCR assay to quantify Plasmodium falciparum DNA from dried blood spots, achieving substantially higher sensitivity than qPCR alone and showing strong correlation with microscopy and external qPCR tests, enabling improved detec…
DOI: 10.1186/1475-2875-13-393 -
Time is of the essence: effectiveness of dairy farm control of H5N1 is limited by fast spread
This study investigates the rapid spread of H5N1 avian influenza in dairy farms, revealing that current surveillance and control measures are insufficient due to the swift transmission dynamics of the virus.
DOI: 10.1101/2024.11.15.623774 -
Evaluating sampling strategies for the detection of avian influenza viruses in the environment
A year-long environmental surveillance study in two New Zealand urban ponds comparing four eRNA-based sampling methods with RT-qPCR and metagenomics to detect avian influenza viruses, finding RT-qPCR on actively filtered water most sensitive with inconsistent detections across s…
DOI: 10.1101/2025.09.18.677014 -
Feasibility study for identifying adverse events attributable to vaccination by record linkage
Feasibility of identifying vaccine-attributable adverse events by linking hospital admissions with vaccination records, showing high match quality and modest sensitivity sufficient for surveillance.
DOI: 10.1017/s0950268800052183 -
Safety surveillance of respiratory syncytial virus (RSV) vaccine among pregnant individuals: a real-world pharmacovigilance study using the Vaccine Adverse Event Reporting System
In a real-world pharmacovigilance analysis of VAERS reports from September 2023 to February 2024, 77 adverse event reports after RSVpreF vaccination in pregnant individuals were identified; preterm birth was the most frequently reported pregnancy-specific adverse event and showe…
DOI: 10.1136/bmjopen-2024-087850 -
Surveillance of Wildlife Viruses: Insights from South Australia’s Monitoring of Rabbit Haemorrhagic Disease Virus (RHDV GI.1 and GI.2)
A long-term, multi-method surveillance study in South Australia comparing passive and active wildlife virus monitoring (including capture–mark–recapture with serology, carcass and spotlight sampling, and vector fly surveillance) to track Rabbit Haemorrhagic Disease Virus GI.1 an…
DOI: 10.3390/v16101553 -
Molecular Epidemiology of Novirhabdoviruses Emerging in Iranian Trout Farms
From 2015 to 2019, surveillance and diagnostic investigation of 13 mortality outbreaks in Iranian rainbow trout farms identified 9 VHSV isolates (genogroup Ia, clade Ia-2) and 4 IHNV isolates (genogroup E, clade E-1), with G-gene sequence identity and haplotype networks supporti…
DOI: 10.3390/v13030448 -
An enhanced technique combining pre-enrichment and passive filtration increases the isolation efficiency of Campylobacter jejuni and Campylobacter coli from water and animal fecal samples
This study evaluates an enhanced technique combining selective enrichment in Bolton Broth and passive filtration for the efficient isolation of Campylobacter jejuni and Campylobacter coli from environmental water and animal fecal samples, demonstrating improved specificity and c…
DOI: 10.1016/j.mimet.2012.09.005 -
A quantitative risk assessment to evaluate the efficacy of mitigation strategies to reduce highly pathogenic avian influenza virus, subtype H5N1 (HPAI H5N1) in the Menoufia governorate, Egypt
Monte Carlo–based quantitative risk assessment shows vaccination and surveillance reduce H5N1 prevalence in Menoufia, Egypt, but incomplete vaccination coverage and imperfect detection leave substantial residual prevalence in both farms and backyard poultry.
DOI: 10.1186/s12917-021-02917-7