Autism and Vaccines: Are Siblings Affected?

Background: After a 1998 study in England incorrectly showed a link between autism and the measles, mumps and rubella (MMR) vaccine, parents became concerned about vaccinating their children. This study led to a decrease in vaccination rates and an increase in measles, mumps, rubella and other vaccine-preventable diseases in England and other countries. Although the original article has since been invalidated by several studies and subsequently retracted, the decrease in vaccination rates persists. In the United States, parents can easily obtain both medical and non-medical exemptions from school vaccine requirements for their young children. Communities with low vaccination rates continue to be at higher risk for disease outbreaks. Although previous studies have documented that parents with at least one child with autism spectrum disorder (ASD) may delay or decline vaccination for subsequent children, these studies suffer from the limitations of small sample sizes and parent self-report.  

Objectives: To compare the vaccination rates of the immediate younger siblings of children with ASD to those of immediate younger siblings of children without ASD using a large national healthcare claims database. 

Methods: Data for the study came from the MarketScan^® Commercial Claims and Encounters database, a national health insurance claims database of over 40 million privately insured individuals. Immediate younger siblings of children with ASD aged 0 to 6 were identified and matched to immediate younger siblings of children without ASD. Children in both groups were matched based on age, sex and order of birth. Vaccination rates for both groups were determined using ICD-9 codes to identify receipt of any recommended vaccine (any vaccine) and individual vaccines of interest (MMR, influenza, hepatitis A/B, and chicken pox). The proportions of children who received studied vaccines were compared across groups using t-tests. In addition, generalized estimating equations regression was used to determine whether the effect of group (ASD vs. control) remained after controlling for other child and family characteristics.

Results: A total of 5,048 case-control pairs were included in the study. Vaccination rates for all studied vaccines were significantly lower among children with an older sibling with ASD than among matched controls. After controlling for other covariates, immediate younger siblings of children with ASD remained significantly less likely to receive any vaccine (OR=0.86; 0.79-0.93), MMR (OR=0.72; 0.66-0.80), influenza (OR=0.79; 0.73-0.86), hepatitis A/B (OR=0.82; 0.75-0.90), and the chicken pox vaccines (OR=0.80; 0.73-0.87).  

Conclusions: Immediate younger siblings of children with ASD are significantly less likely to receive recommended vaccinations compared to immediate younger siblings of children without ASD. This is especially true of the MMR vaccine, the subject of the invalidated 1998 study. Our results suggest that in addition to their current efforts, public health officials and clinicians need to tailor vaccine education specifically to parents of children with ASD.