Autism spectrum disorder (ASD) affects almost one percent of all children born in the United States. It can vary in severity from a complete inability to communicate and interact socially to a minimal lack of social and communicative skills, restricted interests (and even intellectual brilliance in the interest) and repetitive behaviors. How much of this is genetic and how worried about heredity do families have to be if one child is already affected by ASD?
A recent study in Sweden of all children born between 1982 through 2006 assessed this genetic concern and the findings were reported in the May 7 issue of JAMA. In Sweden, all infants and preschool children regularly undergo medical and development examinations. And at age 4, every child goes through a mandatory development assessment of his or her motor, language, cognitive and social development. Any child with suspected developmental disorders is referred for further assessment by a specialized team in a child psychiatry unit or habilitation service. (Ideally, this should be done here in the US; alas it is not…so much depends on parents’ recognizing and reporting initial symptoms and follow up by pediatricians. A similar comprehensive study will not be forthcoming nationally, unless, of course, we embrace universal health care…but I am supposed to leave that subject alone for a while.)
The investigators followed the records of a total of 2,049,973 children; 37,570 were twins, 2,642,064 were full siblings (sisters and brothers with the same parents), 432,281 were maternal half siblings (with the same mother) and 445,531 were paternal half siblings (with the same father). Please don’t try to remember these numbers, I just thought they were meticulously fabulous and had to include them!…To continue, they found that 14,516 children had ASD. The male to female ratio was 2.74.
What this means is that for children with a full sibling who has ASD, the probability of an ASD diagnosis by age 20 is estimated to be 12.9% compared with 1.2% for individuals without am affected sibling. The probability of an ASD diagnosis at age 20 is 59.2% for identical twins, 12.9% for non-identical twins, 8.6% for maternal half siblings, 6.8% and for paternal half siblings, 2.6%.
This was a study of more than 2 million families and indeed is the largest population-based study evaluating familial risk of ASD. After the epidemiologists and statisticians did their analyses they came to the conclusion that heritability of ASD is 50%, and that genetic factors explain half of the risk of autism-spectrum disorder. An individual’s risk increases with the degree of relatedness to the person who has ASD. In statistical terms: the relative risk is 10.3 for full siblings (versus 1 for non relatives… i.e.. more than 10 times more likely to occur), 3.3 for maternal half siblings, 2.9 for paternal siblings, and 2.0 for cousins. (My take on the difference between maternal and paternal half siblings may have to do with the fact that the stats may not always represent the true father!)
The authors concluded with the statement that “These findings may help inform the counseling of families with affected children.” I’m not sure it helps to know that heritability of ASD is 50% but I guess it’s a lot better than being told it’s 100%… we definitely need more research and methods for predicting ASD at or before the onset of a pregnancy. The good news is that we’re getting better at early diagnosis and cognitive therapies.