By preventing CRS, rubella vaccination has prevented hundreds, and perhaps thousands, of ASD cases from 2001 through 2010 in the US. These results demonstrate that the CRS-ASD association is not trivial, though the prevented cases represent only a small fraction of current ASD prevalence.
This study relies on estimates from Chess's evaluation of children with CRS. Fombonne, et al., published on the link between autism and other medical disorders in 1997 . The investigators found the rate of CRS among autistic children to be 0.6%, but the association between autism and CRS was not significant. However, given the low overall prevalence of CRS in the population, the relatively small sample size (174 children with autism) would not have been sufficient to detect an association . In addition, the controls selected by Fombonne and colleagues were children with other "medical and developmental problems" and "intellectual deficits," conditions which are also associated with CRS [1, 2, 22]. The findings of Fombonne and colleagues cannot be directly compared to Chess's estimates because Fombonne's study examined the relationship between CRS and autism following implementation of rubella vaccination.
According to Castillo-Solórzano, et al. , "In the United States the lifetime cost of treating a patient with CRS is estimated at more than $200,000." Using the CRS prevention estimates described in Results, rubella vaccination saved the US between $1.7 billion and $12.5 billion by preventing CRS from 2001 to 2010. These cost savings do not include the prevention of stress and emotional suffering for individuals with CRS and their families. Furthermore, since pre-vaccine era CRS incidence was based on the number of live births, these estimates do not account for stillbirths, miscarriages, or induced abortions related to CRS.
Our study is subject to a number of limitations. Because CRS and ASD prevention were estimated through mathematical modeling, the accuracy of prevention estimates is determined by the model parameters. Of the model parameters, estimates of vaccine era CRS incidence and the number of live births per year are based on recent data from reliable sources and, therefore, are expected to be accurate. Although data on these parameters were only available through 2008 and 2009, respectively, substantial changes in these parameters were not expected to occur before the end of the ten-year period (2001 to 2010).
Our results are limited in that we had to rely on the 1971 Chess study  for estimates of the percentage of CRS cases presenting with an ASD; no comparable data were available elsewhere. However, this limitation should lead to underestimation of ASD prevention, since ASD diagnostic criteria have expanded since the Chess study was conducted . With the expansion of Kanner's classical criteria, we expect that more children with CRS would be diagnosed with an ASD today. Additionally, we used data from other countries to estimate pre-vaccine era CRS incidence in the United States because specific data for the US were not available.
Among parents who choose not to have their children vaccinated, vaccine safety concerns are often cited as the reason for vaccine refusal . In a survey of pediatricians and family physicians in the United States in 2000, 69% of physicians perceived a "substantial increase" in parental concerns about vaccine safety . Regarding MMR vaccination, a prospective cohort study of children in the United Kingdom found that 74% of parents of unvaccinated children made a "conscious decision" not to have their children vaccinated with the combined MMR vaccine . Many of these parents claimed to be concerned about vaccine safety, including the potential association between the vaccine and autism. A smaller fraction cited "negative media attention" as a reason for their decision . In the US, where monovalent rubella vaccine is not available, MMR vaccination is the only means of rubella vaccination.
Despite claims that MMR vaccination causes autism, research does not support this association [27–30]. These claims are also ironic in light of our results, which demonstrate that MMR vaccination (through the rubella component of the vaccine) actually prevents cases of autism and other ASDs through the prevention of CRS. Although a disparity between public opinion and scientific fact is evident, physicians and other healthcare providers are in a prime position to address the issue.