People like to have a diagnosis and a reason for their medical problems. If one is not offered, they will find one no matter what the evidence demonstrates.

Autism is a good example. Medicine cannot point to a single reason why people get autism and as a result people will look for a cause. Vaccines remain a scapegoat, despite numerous studies that cannot find a relationship between vaccines and autism. But if it is not vaccines, then what causes autism?

Sweden has a single health care system and allowed researchers to look at over 2 million children to tease out the risk of autism among relatives, expressed as the relative recurrence risk (RRR),

the risk of an autism diagnosis in a sibling of a child with autism compared with a sibling of a child without autism.

The closer the relationship to a case, the more likely there is a diagnosis of autism: 10.3 for full siblings, 3.3 for maternal half siblings, 2.9 for paternal half siblings, and 2.0 for cousins.

For comparison, for schizophrenia they give the RRR as 8.3 for full siblings, 2.5 for half siblings, and 2.3 for cousins.

It points to a strong genetic component to autism, but how much? Overall they estimate the inheritability for autism at 50%, which leaves another 50% for as of yet identified environmental influences.

It will be good when they can point to the reasons for autism, although I suspect for some an explanation will not sway them from a belief in vaccines as the reason.


The Familial Risk of Autism Sven Sandin, MSc1,; Paul Lichtenstein, PhD1; Ralf Kuja-Halkola, MScns JAMA. 2014;311(17):1770–1777. doi:10.1001/jama.2014.4144.

Points of Interest: 5/08/2014