Introduction

This article explores vaccine risk perceptions amongst parents who reported cases of autism in their children to the Vaccine Adverse Events Reporting System (VAERS). The VAERS is a passive surveillance system administered by the United States-based Food and Drug Administration (FDA) and Centers for Disease Control and Prevention (CDC) that allows health care professionals, manufacturers, and vaccinees and their parents to report adverse reactions. The authors interviewed 124 parents of children with autism who contacted VAERS to determine why they believe that vaccines may have contributed to their children's condition, and to determine how their perceptions of vaccine safety, disease severity, reasons for immunising, and credibility of key sources of vaccine differ from the general population. This study is taken in the context of continuing concern about the mumps-measles-rubella (MMR)-autism link, and considers 1) the impact of the highly publicized "Wakefield" case series of February 1998 that suggested a causal relationship between MMR vaccination, Crohn's disease, and the onset of autism, and 2) the subsequent (July 1999) US Public Health Service/American Academy of Pediatrics (PHS/AAP) recommendation that manufacturers remove thimerosal/mercury from their vaccines.

Key Findings

All respondents had reported some type of autism or similar developmental delay among their children to the VAERS. The majority of respondents were the biological, adoptive, or stepmother of the child (91%), over age 30 (93%), with at least some college education (91%), and White (91%). While the study examined reports registered over an 11-year period, the majority (66.9%) of reports had been logged between August 1999 and July 2001. 85.5% of the children were male with an average age of symptom onset at 19.9 months, of which 75% were diagnosed with autism or autism spectrum disorder. 65.3% of the vaccines involved were listed as MMR or component vaccines, though 47.6% of the reports listed multiple vaccines, the most common of which were Haemophilus influenzae type B, oral polio virus (OPV), and diphtheria-pertussis-tetanus (DPT).

The authors used open-ended questions to evaluate the reasons behind the reported association between vaccination and autism. The largest proportion, at 31.5%, volunteered responses that had to do with the temporal proximity of vaccination and the onset of symptoms. However, other reasons for making the association included having received information from magazine/newspaper articles (24.2%), nurses/doctors (21%), the web/Internet (19.4%), family/friends (14.9%), and consumer advocacy groups (11.3%). Responses were not mutually exclusive. The later the reports were logged with VAERS, the higher the proportion of respondents who reported temporal reasons. The percentages increased significantly after the media coverage associated with the Wakefield (February 1998) report and the PHS/AAP advisory (July 1999).

When asked about factors that contributed to their children's condition, 96% of respondents stated that vaccine ingredients had played a very strong or moderate role. Other factors cited included children receiving vaccines at too early an age (95.2%), too many vaccines at once (94.4%), thimerosal/mercury in vaccines (86.3%) and the MMR vaccine (78.2%). Less than half (41.1%) cited genes, family, or birth defects. Reports received after the Wakefield article was published that cited MMR increased from 66.7% to 81.4%.

The authors also compared respondents' perceptions surrounding immunisation with those of the general public, drawing data from an earlier general study on immunisation (Gellin et al., 2000). Only 15.3% of VAERS reporters felt that immunisation was extremely important, while the majority ranked its importance below nutrition, hand washing, and exercise (a finding in direct contrast to the general public, which ranked immunisation above these three). Respondents were also less likely to state that the reason to immunise was to prevent infections (56% vs. 83%) and more likely to cite government or school requirements (35% vs. 8%).

The respondents were also queried as to the refusal of further immunisations. 46% said that the symptomatic child had not had any further vaccinations, while 26.6% of siblings had not had any vaccinations subsequently. Respondents were also asked to rate the severity of vaccine-preventable diseases. Overall, the perceptions of severity were significantly lower than those found among the general population. 24.2% of respondents also felt that vaccines were "not at all safe" (a 0 score on a 10-point scale), while 78.2% rated them a 5 or lower, with a mean score of 3.4. This finding is compared with a mean score of 8.2 amongst the general public. One of the most substantial differences was the belief in the veracity of the safety testing carried out on vaccines: While 71% of the general populace agree or strongly agree with the statement "Vaccines are always proven to be very safe before they are approved for use", only 7% of VAERS reporters concurred.

Finally, the authors attempt to assess differences in the VAERS reporters' perceptions of the credibility of key sources of immunisation information. They were first evaluated on the basis of their familiarity with organisations such as the CDC, the National Vaccine Information Center (NVIC) (a consumer advocacy group), and groups such as the American Academy of Paediatrics (AAP). They were then asked to report how trustworthy they considered each group that they were familiar with, using a 10 point scale (where 0 = "very untrustworthy" and 10 = "very trustworthy"). The NVIC scored higher (mean: 7.2) than the CDC (mean: 5.0) and the AAP (mean: 4.9), while the general populace ranked both the CDC and AAP as very trustworthy (mean: 8.5 each).

The authors provide several conclusions. First, they find that, while most reporting parents based their associations on their own observations, media influences had a substantial effect. Second, the respondents had very little trust in the major policy-shaping institutions involved in immunisation. Third, respondents perceived vaccine-preventable diseases as being substantially less serious than other parents did and based immunisation decisions on these beliefs. Finally, the authors note the strong rise in association of autism with the MMR vaccine after the publishing of the Wakefield article and the advisory. Three-quarters of the VAERS reports were received after the Wakefield case series was published. Indeed it appears that several parents submitted reports after the case series even though the onset of their child's symptoms had occurred many years prior. Media exposure clearly contributed to the construction of an association between the two in the public eye.

In conclusion, the authors see the need for the development of fair and effective communication strategies to convey information about the health benefits and risks of vaccination. They suggest that there is a need to pay more attention to concerns about vaccine-related adverse events, and that medical practitioners of all stripes must be active participants in this two-way process.