Joint Attention and Social-Communicative Abilities of Siblings of Children with ASD

Background: Research on the broader autism phenotype (BAP) can benefit from studying early social and communicative skills of younger siblings of children with an autism spectrum disorder (ASD) (Tager-Flusberg, 2010). These siblings have a 10 to 20 times higher risk of developing ASD themselves and they are likely to share some behavioural characteristics with their older sibling (Szatmari, Zweigenbaum, & Bryson, 2004).

Objectives: High-risk siblings of children with ASD were followed between the ages of 7 and 36 months. For the current study, we focused on the ages of 12 and 18 months. Joint attention, language, and cognitive ability were investigated in both the siblings and a low-risk control group. Three research questions were formulated. First, do high-risk siblings show a generalized impairment in joint attention abilities as a component of the BAP? Second, can joint attention be an early marker for a later diagnosis of ASD in high-risk siblings? And third, can early joint attention predict the development of later language and cognitive abilities in both high-risk siblings and low-risk control children?

Methods: 33 infant siblings of children with ASD (of whom 6 with an ASD diagnosis at the age of 3) and 32 typically developing infants participated. At the age of 12 months, two aspects of joint attention were measured: initiation of joint attention (IJA) and response to joint attention (RJA), as well as cognitive ability with the Mullen Scales of Early learning. At 18 months the joint attention measures were repeated, together with parent report on the M-CHAT and the Dutch version of the CDI. At 18 months the ADOS was administered as well.

Results: Unaffected siblings and typically developing children did not differ with regard to joint attention ability at both ages: both groups showed the same degree of IJA and RJA. However, siblings with a later ASD diagnosis had lower scores than unaffected siblings for both IJA and RJA. Furthermore, in the combined sibling group, RJA at 12 months was predictive of ADOS and M-CHAT scores at 18 months. Furthermore, developmental index at 12 months was predictive of the social interaction score on the ADOS, M-CHAT and receptive language at 18 months. In the control group RJA only predicted the social interaction score on the ADOS.

Conclusions: Unaffected siblings of children with ASD seem to have no generalized joint attention deficits at 12 and 18 months of age, while siblings with a later ASD diagnosis do show problems with both IJA and RJA. Social-communicative and language problems of the siblings, as measured with the ADOS, M-CHAT and NCDI, were predicted by earlier RJA and developmental index. These predictive variables are thus important for follow-up of infants at risk for ASD.