• Text Resize A A A
  • Print Print
  • Share Share on facebook Share on twitter Share

Vaccine Safety: Evidence and Belief

Harvey Fineberg
President, Institute of Medicine

Over the past 35 years, the Institute of Medicine has conducted more than 60 studies related to vaccine safety. Two years ago, a committee of experts reviewed possible adverse effects of eight vaccines, assessing the possible relation to immunization of 158 specific health problems.  Some vaccines were found to cause anaphylaxis in rare instances . The committee found evidence sufficient to favor rejection of a causal association in five instances, including measles, mumps, and rubella vaccine and autism, and inactivated influenza vaccine and asthma. While few health problems are clearly associated with vaccines and some putative associations can be rejected based on evidence, in the majority of cases, evidence was inadequate to accept or to reject a causal relationship.

Physicians and public health champions, mindful of the devastating consequences of vaccine-preventable illness and the profound role of vaccines in saving lives and increasing life expectancy, stress the few demonstrated associations of adverse events and vaccines and their enormous value for health. And yet, some individuals and groups in the United States reject immunization for themselves and their children. Some are convinced of an association because the timing of an adverse event's presentation followed soon after immunization, such as the first symptoms of autism appearing a few weeks or months after immunization and a febrile reaction—and no amount of epidemiological evidence will dislodge this conviction. Others may be suspicious of science, resist all medical interventions on the basis of religious belief, or simply believe that avoiding vaccines is the safer course for their child, especially when the evidence is inconclusive on most possible side effects.

Identifying potential adverse events in connection with vaccines depends in the first instance on surveillance and reporting. Surveillance tools that have been developed and deployed over the years include the Vaccine Adverse Event Reporting System jointly administered by the Centers for Disease Control and Prevention (CDC) and the Food and Drug Administration (FDA), the Vaccine Safety Datalink that connects electronic data systems at selected health maintenance organizations with CDC, and the Post-Licensure Rapid Immunization Safety Monitoring system at FDA. Although these systems each have their limitations, they can help trigger attention to possible side effects and aid in their assessment.

Confidence in vaccine safety requires more than surveillance and reporting in real time. In light of the paucity of strong conclusions about possible vaccine side effects, continued and selective investment in epidemiologic and other investigations into the risks of immunization will be necessary. A scientific research design is generally intended to test whether an effect is present, or more precisely, whether the evidence for an association is sufficient to reject the assumption of no causal association. However, insufficient evidence for a causal connection is not the same as evidence for the absence of any association. About the best one can do is to estimate, based on the evidence, the probability that the frequency of an adverse event is less than a specified, low level. This may be enough for the physician who weighs the public health and personal health benefit against a very low risk, but not enough to satisfy a wary parent.

Continued, candid, and open communication is also an essential ingredient to a successful vaccine safety regime. This means more than the experts explaining the benefits and risks to parents and families. It means listening carefully to the anxieties and doubts, staying true to the strength of evidence without exaggeration or misrepresentation, and reporting fully and fairly on scientifically sound investigations into possible adverse events.

Content created by National Vaccine Program Office
Content last reviewed on March 27, 2016
Back to Top