Can autism be identified as young as 2 months old?

Early diagnosis of developmental disorders is of significant interest to parents because of the opportunity to engage in early interventions that may be more effective earlier in development. It is not surprising then that a study, published in the highly reputable journal Nature, that suggests differences can be detected in children as young as 2 months old who go on to develop autism spectrum disorder (ASD), has received such significant attention.

What did the study examine?

The authors, Dr. Warren Jones and Professor Ami Klin from the Marcus Autism Center, Atlanta, Georgia, conducted a longitudinal study of 110 children at high risk of developing ASD on the basis of having a full sibling with the disorder.  They assessed the infant’s preferential attention to the eyes of others using eye-tracking technology at 2, 3, 4, 5, 6, 9, 12, 15, 18 and 24 months. Of this sample, 12 children (11 of whom were included in the analysis), met diagnostic criteria for ASD at 36 months. One of the 11 children came from a comparison ‘low-risk’ group (who did not have any relatives with ASD). At each of the 10 testing sessions the infants viewed videos of a female actor who looked directly into the camera and who played the part of a caregiver (e.g. playing pat-a-cake).

What were the findings?

The authors reported that infants who later developed ASD showed a mean decline in eye fixation from 2 to 6 months. The authors concluded from these findings that in infants who later develop autism, eye looking starts out in the typically expected way, but from 2 months begins to decline. Such a decline is not observed in typically developing children and the authors refer to this decline as “a derailment of processes that would otherwise have a key role in canalizing typical social development”. They are suggesting that children with ASD start out with typical looking patterns but for some reason this development derails fairly early on in development.

What media attention did this study receive?

The study received significant media attention including pieces in the BBC News, Daily Mail and The Guardian.

The BBC coverage included comments from two experts including an academic (Deborah Riby, University of Durham) and an Autistic Society representative (Caroline Hattersley, Head of Information, Advice and Advocacy at the National Autistic Society), providing a very balanced approach. Caroline Hattersley emphasised that the research was “based on a very small sample and needs to be replicated on a far larger scale before any concrete conclusions can be drawn”. She also pointed out the complexity of arriving at a diagnosis of ASD and the need for a comprehensive approach.  The Daily Mail coverage was also very well balanced with the inclusion of several comments from the study authors which set the study findings in context e.g. that these patterns of eye gaze are not visible to the naked eye and require specialist equipment to be assessed.

Caution needed when interpreting the results

Although the study has a very strong experimental design, most notably the use of a prospective longitudinal design (whereby those taking part are followed up over time) of both low and high risk children, the data reported are with a very small sample size (11 children) and replication of the study results is really essential to verify the pattern of data reported. A number of autism experts have critically assessed the study in blogs and within social media and have appraised the study design favourably but have similarly emphasised the need for replication. For example, Jon Brock, an autism researcher from Macquarie University, Sydney, has pointed out on Twitter that the paper does not directly compare high risk infants who do and do not develop ASD, which would be extremely interesting data; the data reported only compare high and low risk infants. He also highlighted that, statistically speaking, the data from 3 to 6 months were just above chance in discriminating children who went on to develop ASD with low risk infants.  Like others, he emphasises that the study data are very preliminary while noting that it is a very interesting study.  Overall, the findings are potentially very informative for the direction of research with young infants at risk of developing autism. The authors note in some of the media coverage that they are currently following up a larger number of children and it will be interesting to see the outcome of that work.

Warren Jones & Ami Kiln (2013). Attention to eyes is present but in decline in 2–6-month-old infants later diagnosed with autism. Nature. doi:10.1038/nature12715