Adverse Effects of Vaccines: Evidence and Causality
Immunizations are a cornerstone of the nation’s efforts to protect people from a host of infectious diseases. As required by the Food and Drug Administration, vaccines are tested for safety before they enter the market, and their performance is continually evaluated to identify any risks that might appear over time.
Vaccines are not free from side effects, or “adverse effects,” but most are very rare or very mild. Importantly, some adverse health problems following a vaccine may be due to coincidence and are not caused by the vaccine. As part of the evaluation of vaccines over time, researchers assess evidence to determine if adverse events following vaccination are causally linked to a specific vaccine, and if so, they are referred to as adverse effects. Under the National Childhood Vaccine Injury Act of 1986, Congress established the National Vaccine Injury Compensation Program (VICP) to provide compensation to people injured by vaccines. Anyone who thinks they or a family member—often a child—has been injured can file a claim.
The Health Resources and Services Administration (HRSA), the agency within the Department of Health and Human Services that administers VICP, can use evidence that demonstrates a causal link between an adverse event and a vaccine to streamline the claim process. As such, HRSA asked the Institute of Medicine (IOM) to review a list of adverse events associated with vaccines covered by VICP and to evaluate the scientific evidence about the event—vaccine relationship. The vaccines covered by VICP include all vaccines recommended by the Centers for Disease Control and Prevention (CDC) for routine administration in children. Adults who experience an adverse event following one of these childhood vaccines also are covered by the program. HRSA asked the IOM to review 8 of the 12 covered vaccines. These eight are the varicella zoster vaccine (used against chickenpox); the influenza vaccines(except for the H1N1 influenza vaccine distributed in 2009); the hepatitis B vaccine; the human papillomavirus(HPV) vaccine; the measles, mumps, and rubella (MMR) vaccine; the hepatitis A vaccine; the meningococcal vaccines, and tetanuscontaining vaccines that do not carry the wholecell pertussis component.
Examining the Evidence
The adverse events selected by HRSA for IOM review are ones for which people have submitted claims—successful or not—to VICP. The committee appointed to this study was not asked to assess the benefits or effectiveness of vaccines but only the risk of specific adverse events. Its conclusions reflect the best evidence available at the time. Some of the adverse events the committee examined already are accepted in the medical community, but they are minor or manageable—for example, a sudden allergic reaction called anaphylaxis that can follow the administration of some vaccines.
In its report, the committee explains its process for evaluating the list of adverse events and provides a set of 158 causality conclusions. The committee examined two types of evidence: epidemiologic evidence, which derives from studies of populations, and mechanistic evidence, which draws from biological and clinical studies. The committee evaluated each scientific article for its strengths and weaknesses and then assigned a “weight of evidence” ranking to both the epidemiologic and mechanistic bodies of studies.
The committee considered the weights of evidence and then reached a conclusion about the causal relationship between each vaccine and adverse health problem pairing. The committee began from a position of neutrality, presuming neither causation nor lack of causation, and moved from that position only when the combination of evidence suggested a more definitive assessment regarding causation. The figure provides an explanation of how the evidence influenced the causality conclusions.
Based on the totality of the evidence, the committee assigned each relationship to one of four categories of causation in which the evidence:
- convincingly supports a causal relationship;
- favors acceptance of a causal relationship;
- favors rejection of a causal relationship; or
- is inadequate to accept or reject a causal relationship.
The committee did not use a category to designate evidence that convincingly supports no causal relationship, because it is virtually impossible to prove the absence of a very rare relationship with the same certainty that is possible to establish the presence of one.