How science determines whether a vaccine caused an adverse event — and why that question is harder to answer than it sounds
This page provides a summary of how science determines whether a vaccine caused a specific adverse event. For the full framework — including detailed tables, historical case studies, published critiques, and primary sources — see the comprehensive VAERS Explained page.
Last updated: April 2026.
Causality — also called causal inference — means establishing that one event directly caused another. In vaccine safety, this is the difference between two statements:
"Event X happened after Vaccine Y" — a temporal association (an observation)
"Vaccine Y caused Event X" — a causal relationship (a conclusion requiring evidence beyond timing)
The assumption that because one event follows another, the first must have caused the second is a reasoning error known as post hoc ergo propter hoc — "after this, therefore because of this." The Institute of Medicine identified this fallacy directly in its 1994 report on childhood vaccine adverse events (IOM, 1994).
Adverse health events occur every day in every population regardless of vaccination status. Epidemiologists call these "background rates." Among 10 million vaccinated people, approximately 237 would die from any cause within one day purely by coincidence — not because of the vaccine (Abara et al., Journal of Infectious Diseases, 2022). This does not mean vaccines never cause harm. It means that observing an adverse event after vaccination is a starting point for investigation, not proof of causation.
Three questions help distinguish strong causal evidence from temporal coincidence:
No single question settles the matter, but together they provide a useful starting framework.
Several structured frameworks are used to evaluate whether vaccines cause specific adverse events:
Bradford Hill Viewpoints (1965). Nine considerations — including strength of association, consistency, temporality, and biological plausibility — used to evaluate whether an observed association is likely causal. Hill called these "viewpoints," not criteria, and emphasized that none are individually required. For the full 9-viewpoint table with vaccine-specific examples, see VAERS Explained.
IOM/NASEM Four-Category System. The National Academies evaluate vaccine-adverse event pairs using four categories: "establishes a causal relationship," "favors acceptance," "inadequate to accept or reject," and "favors rejection." The 2012 IOM report examined 158 pairs; 135 were classified as "inadequate" — reflecting limited evidence, not a verdict. The 2024 NASEM report established a causal relationship between mRNA vaccines and myocarditis. For the full category table and report findings, see VAERS Explained.
WHO AEFI Causality Assessment Algorithm. The WHO system classifies individual adverse event cases as Consistent, Inconsistent, Indeterminate, or Unclassifiable. It is designed for individual case assessment rather than population-level evidence review. Peer-reviewed critiques have argued the algorithm can make it structurally difficult to classify novel or multifactorial adverse events (Puliyel & Naik, 2018; Bellavite et al., 2024). For the full framework and published critiques, see VAERS Explained.
Evidence hierarchy. Different study designs carry different weight for causal claims — from randomized controlled trials (strongest for isolating a vaccine's effect) to case reports (useful for generating hypotheses but unable to establish causation). For the full study design comparison table, see VAERS Explained.
U.S. surveillance infrastructure. No single system determines causation. VAERS detects signals; the Vaccine Safety Datalink (VSD) and PRISM test hypotheses with controlled data; CISA provides clinical case review; V-safe actively solicits post-vaccination health reports. Together, these systems form a signal-detection-to-confirmation pipeline. For full system descriptions, see VAERS Explained.
Legal vs. scientific causation. The Vaccine Injury Compensation Program (VICP) uses a lower evidentiary standard than scientific causation — by design. Legal causation requires "more likely than not" (preponderance of evidence); scientific causation requires statistical significance replicated across studies. The two systems answer different questions for different purposes and are complementary, not contradictory. For the full comparison table and Althen test explanation, see VAERS Explained.
The causality assessment system has produced different outcomes in different directions. It has confirmed genuine vaccine-caused injuries (OPV paralysis, RotaShield intussusception, mRNA myocarditis), identified and acted on rare risks (J&J TTS, 1976 swine flu GBS), and found no causal association where evidence was extensively studied (MMR and autism, Hepatitis B and MS).
Academic debate over how causality frameworks should be applied is ongoing. Published critiques address whether Bradford Hill's viewpoints are applied too rigidly, whether institutional algorithms create structural barriers for novel adverse events, and whether the "correlation does not imply causation" principle has sometimes been used to dismiss signals before adequate investigation. These debates appear in peer-reviewed literature and reflect an ongoing effort to improve — not eliminate — causality assessment.
For the full 7-case historical examples table and detailed discussion of these debates, see VAERS Explained.
Causality assessment in vaccine safety is a layered process, not a single test. The strongest causal conclusions draw on multiple methods, multiple studies, and multiple independent populations. The weakest rely on any single source alone — whether that source is a VAERS report, a single study, or an institutional verdict.
Different frameworks can lead researchers to weigh the same evidence differently, especially when evidence is limited, evolving, or indirect. No single framework, study, or report settles every question.
For the complete causality framework with tables, sources, critiques, and case studies, see the full VAERS Explained page.
Hill AB. The Environment and Disease: Association or Causation? Proc Royal Soc Med. 1965. PMC →
IOM. Adverse Effects of Vaccines: Evidence and Causality. National Academies Press. 2012. NAP →
NASEM. Evidence Review of the Adverse Effects of COVID-19 Vaccination. 2024. NAP →
WHO. Causality Assessment of AEFI: User Manual (2nd ed.). 2019. WHO →
Abara WE, et al. Expected Rates of Select Adverse Events After Immunization. J Infect Dis. 2022. OUP →
For the full 20-source annotated bibliography, see the Primary Sources section on VAERS Explained.
Hill AB. The Environment and Disease: Association or Causation? Proc Royal Soc Med. 1965;58:295–300. PMC →
Original source of the nine viewpoints for evaluating causal associations.
Institute of Medicine. Adverse Events Associated with Childhood Vaccines. National Academies Press. 1994. NCBI Books →
Defined post hoc ergo propter hoc fallacy in vaccine context; foundational IOM vaccine safety report.
Institute of Medicine. Adverse Effects of Vaccines: Evidence and Causality. National Academies Press. 2012. NAP →
Most comprehensive vaccine safety review: 158 pairs evaluated across 4 causality categories.
NASEM. Evidence Review of the Adverse Effects of COVID-19 Vaccination. National Academies Press. 2024. NAP →
85 COVID-19 vaccine conclusions; established mRNA-myocarditis causal relationship.
WHO. Causality Assessment of AEFI: User Manual (2nd ed.). 2019. WHO →
Global standard for individual-case AEFI causality assessment.
Abara WE, et al. Expected Rates of Select Adverse Events After Immunization. J Infect Dis. 2022;225(9):1569–1578. OUP →
Calculated expected background rates to contextualize post-vaccination adverse events.
Phillips CV, Goodman KJ. The Missed Lessons of Sir Austin Bradford Hill. Epidemiologic Perspectives & Innovations. 2004;1:3. Springer →
Argued Hill's viewpoints are misapplied as rigid criteria rather than flexible considerations.
Shimonovich M, et al. Assessing Causality in Epidemiology: Revisiting Bradford Hill. Eur J Epidemiol. 2021;36:873–887. PMC →
Mapped Hill's viewpoints against modern causal inference methods.
Puliyel J, Naik P. Revised WHO's Causality Assessment of AEFI — A Critique. F1000Research. 2018;7:243. PMC →
Argued WHO AEFI classification creates structural paradox excluding novel and multifactorial adverse events.
Bellavite P, et al. WHO Algorithm for Causality Assessment of AEFI: Pitfalls and Suggestions. PMC. 2024. PMC →
Extended WHO critique to COVID-19 vaccines; flagged circular reasoning for novel platforms.
Bellavite P. Causality Assessment of AEFI: The Problem of Multifactorial Pathology. F1000Research. 2020. PubMed →
Argued WHO algorithm underestimates vaccine contribution when other factors interact.
Kohl KS, et al. The Brighton Collaboration. Advances in Patient Safety. 2005. NCBI Books →
International network standardizing AEFI definitions across 57 countries.
McNeil MM, et al. The Vaccine Safety Datalink. Vaccine. 2014;32(42):5390–5398. PMC →
Comprehensive overview of VSD methodology and achievements.
Grey B. The Plague of Causation in the NCVIA. Harvard J on Legislation. 2011;48:2. Harvard JOL →
Analysis of VICP legal causation standards; ~90% of claims now off-Table.
Althen v. Secretary of HHS, 418 F.3d 1274 (Fed. Cir. 2005). FindLaw →
Established the three-prong Althen test for off-Table VICP causation.
HRSA. VICP Data & Statistics. 2025. HRSA PDF →
Compensation statistics: 12,500+ petitions, ~$5.5B total since 1988.
Madsen KM, et al. MMR Vaccination and Autism. NEJM. 2002;347(19):1477–1482. PubMed →
537,303 Danish children; adjusted RR 0.92 — no increased autism risk.
Hviid A, et al. MMR Vaccination and Autism: Nationwide Cohort. Ann Intern Med. 2019;170(8):513–520. PubMed →
657,461 children; strongly supports no MMR-autism link.
CIDRAP. Vaccine Myths That Won't Die — Part 2. 2024. CIDRAP →
Teaching heuristic for temporal coincidence vs. causation.
Preprint (2025). Beyond Correlation and Causation. Preprints.org. Preprints.org →
Kuhnian paradigm critique; not peer-reviewed.