By: Dr. Geoffrey Modest
Yet another study on the relationship (or lack thereof) between MMR vaccination and autism. I bring this up since this is a large study but includes kids at higher risk of autism (see JAMA. 2015;313(15):1534-1540). This is a retrospective cohort study from a large health insurer (Optum) database, using claims data (the insurer is mostly in the south and Midwest of the US, though does have people in all regions. 75% white, 9% Hispanic, 3.5% black and 3.5% Asian).
Details:
–95,727 kids who had older siblings. Autism spectrum disorder (ASD) was defined as 2 claims with a diagnosis of autistic disorder or pervasive developmental disorder, including Asperger. All kids were continuously enrolled in the health plan from birth to age 5 during 2001-2012.
–944 (1.04%) were diagnosed with ASD. 1929 had an older sibling with ASD, of which 134 (6.9%) were diagnosed with ASD, vs 860 (0.9%) of those without an older sib having ASD
–MMR vaccination rates were 84% (n= 78,564) at 2 years of age and 92% (86,063) at 5 years in those without an affected older sib, and 73% (n=1409) at 2 years and 86% (n=1660) at 5 years in those with an older sib with ASD
–for children with an older sib with ASD: at age 2, the adjusted relative risk of ASD for 1 dose of MMR vs no vaccine was 0.76 (0.49-1.18); at age 5 the RR for 2 doses of MMR vs no vaccine was 0.56 (0.31-1.01) — ie, a trend to lower likelihood!!
–for children with no older sib with ASD: at age 2, the adjusted relative risk of ASD for 1 dose of MMR vs no vaccine was 0.91 (0.67-1.20); at age 5 the RR for 2 doses of MMR vs no vaccine was 1.12 (0.78-1.59)
–risk was adjusted for birth year, sex, region, race/ethnicity, maternal/paternal highest education level, mother’s and father’s age at birth of index infant, Childhood Chronic Condition score, seizure, allergies, preterm birth
So, bottom line: another study not finding a link between MMR and ASD, even in kids at higher risk of ASD because of family history (older sib with diagnosis of ASD). Not a great study, since based on insurance claims, but seems to me quite likely that ASD would have been coded as a diagnosis, especially with the concern about the linkage with MMR. Although the insured are a select population (eg working), the prevalence of ASD reported as 1.04% is in line with the nationally-reported 1.5%, and the % of younger sibs developing ASD who have an affected older sib (6.9%) is also within the range of published estimates of 6.4-24.7%. I am actually impressed in this study that 86% of 5 year olds with an affected sib actually got 2 MMR vaccines, given the hype in the popular press!! and, maybe MMR vaccination is even protective, given the trend noted above.