Vaccine Refusal, Mandatory Immunization, and the Risks of Vaccine-Preventable Diseases
List of authors.
Saad B. Omer, M.B., B.S., Ph.D., M.P.H., Daniel A. Salmon, Ph.D., M.P.H., Walter A.
Orenstein, M.D., M. Patricia deHart, Sc.D., et al.
May 7, 2009
N Engl J Med 2009; 360:1981-1988
DOI: 10.1056/NEJMsa0806477
Individual Risk and Vaccine Refusal
Children with nonmedical exemptions are at increased risk for acquiring and transmitting
vaccine-preventable diseases.34,35 In a retrospective cohort study based on nationwide
surveillance data from 1985 through 1992, children with exemptions were 35 times as likely
to contract measles as nonexempt children (relative risk, 35; 95% confidence interval [CI], 34
to 37).34 In a retrospective cohort study in Colorado based on data for the years 1987
through 1998, children with exemptions, as compared with unvaccinated children, were 22
times as likely to have had measles (relative risk, 22.2; 95% CI, 15.9 to 31.1) and almost six
times as likely to have had pertussis (relative risk, 5.9; 95% CI, 4.2 to 8.2).35 Earlier data
showed that lower incidences of measles and mumps were associated with the existence
and enforcement of immunization requirements for school entry.12,36-38
The consequences of delayed vaccination, as compared with vaccine refusal, have not been
studied in detail. However, it is known that the risk of vaccine-preventable diseases and the
risk of sequelae from vaccine-preventable diseases are not constant throughout childhood.
Young children are often at increased risk for illness and death related to infectious diseases,
and vaccine delays may leave them vulnerable at ages with a high risk of contracting several
vaccine-preventable diseases. Moreover, novel vaccine schedules that recommend
administering vaccines over a longer period may exacerbate health inequities, since parents
with high socioeconomic status are more likely to make the extra visits required under the
alternative schedules than parents with low socioeconomic status.39
Multiple studies have shown an increase in the local risk of vaccine-preventable diseases
when there is geographic aggregation of persons refusing vaccination. In Michigan,
significant overlap between geographic clusters of nonmedical exemptions and pertussis
clusters was documented.26 The odds ratio for the likelihood that a census tract included in
a pertussis cluster would also be included in an exemptions cluster was 2.7 (95% CI, 2.5 to
3.6) after adjustment for demographic factors.
In Colorado, the county-level incidence of measles and pertussis in vaccinated children from
1987 through 1998 was associated with the frequency of exemptions in that county.35 At
least 11% of the nonexempt children who acquired measles were infected through contact
with an exempt child.35 Moreover, school-based outbreaks in Colorado have been
associated with increased exemption rates; the mean exemption rate among schools with
outbreaks was 4.3%, as compared with 1.5% for the schools that did not have an outbreak
(P=0.001).35
High vaccine coverage, particularly at the community level, is extremely important for
children who cannot be vaccinated, including children who have medical contraindications to
vaccination and those who are too young to be vaccinated. These groups are often more
susceptible to the complications of infectious diseases than the general population of
children and depend on the protection provided by the vaccination of children in their
environs
Measles vaccination has been extremely successful in controlling a disease that previously
contributed to considerable morbidity and mortality. In the United States, the reported
number of cases dropped from an average of 500,000 annually in the era before vaccination
(with reported cases considered to be a fraction of the estimated total, which was more than
2 million) to a mean of 62 cases per year from 2000 through 2007.43-45 Between January 1,
2008, and April 25, 2008, there were five measles outbreaks and a total of 64 cases
reported.45 All but one of the persons with measles were either unvaccinated or did not
have evidence of immunization. Of the 21 cases among children and adolescents in the
vaccine-eligible age group (16 months to 19 years) with a known reason for nonvaccination,
14, or 67%, had obtained a nonmedical exemption and all of the 10 school-age children had
obtained a nonmedical exemption.45 Thirteen cases occurred in children too young to be
vaccinated, and in more than a third of the cases (18 of 44) occurring in a known
transmission setting the disease was acquired in a health care facility.45
Outbreaks of vaccine-preventable disease often start among persons who refused
vaccination, spread rapidly within unvaccinated populations, and also spread to other
subpopulations. For example, of the four outbreaks with discrete index cases (one outbreak
occurred by means of multiple importations) reported January through April 2008, three out
of four index cases occurred in people who had refused vaccination due to personal beliefs;
vaccination status could not be verified for the remaining cases.45,46 In Washington State, a
recent outbreak of measles occurred between April 12, 2008, and May 30, 2008, involving
19 cases. All of the persons with measles were unimmunized with the exception of the last
case, a person who had been vaccinated. Of the other 18 cases, 1 was an infant who was
too young to be vaccinated, 2 were younger than 4 years of age, and the remaining 15 were
of school age (unpublished data).
Some clinicians have discontinued or have considered discontinuing their provider
relationship with families that refuse vaccines.60,61 In a national survey of members of the
American Academy of Pediatrics, almost 40% of respondents said they would not provide
care to a family that refused all vaccines, and 28% said they would not provide care to a
family that refused some vaccines.61
The academy's Committee on Bioethics advises against discontinuing care for families that
decline vaccines and has recommended that pediatricians “share honestly what is and is not
known about the risks and benefits of the vaccine in question.”62 The committee also
recommends that clinicians address vaccine refusal by respectfully listening to parental
concerns, explaining the risk of nonimmunization, and discussing the specific vaccines that
are of most concern to parents.62 The committee advises against more serious action in a
majority of cases: “Continued refusal after adequate discussion should be respected unless
the child is put at significant risk of serious harm (e.g., as might be the case during an
epidemic). Only then should state agencies be involved to override parental discretion on the
basis of medical neglect.”62
Vaccine refusal not only increases the individual risk of disease but also increases the risk
for the whole community. As a result of substantial gains in reducing vaccine-preventable
diseases, the memory of several infectious diseases has faded from the public
consciousness and the risk–benefit calculus seems to have shifted in favor of the perceived
risks of vaccination in some parents' minds. Major reasons for vaccine refusal in the United
States are parental perceptions and concerns about vaccine safety and a low level of
concern about the risk of many vaccine-preventable diseases. If the enormous benefits to
society from vaccination are to be maintained, increased efforts will be needed to educate
the public about those benefits and to increase public confidence in the systems we use to
monitor and ensure vaccine safety. Since clinicians have an influence on parental decision
making, it is important that they understand the benefits and risks of vaccines and anticipate
questions that parents may have about safety. There are a number of sources of information
on vaccines that should be useful to both clinicians and parents (e.g., Appendix 1 in the fifth
edition of Vaccines, edited by Plotkin et al.; the list of Websites on vaccine safety posted on
the World Health Organization's Web site; and the Web site of the National Center for
Immunization and Respiratory Diseases).67-6
[Link]