Infants at High Risk for Autism Spectrum Disorder: Social-Communication and Language Skills at 12 Months

Background: Infants who have an older sibling with Autism Spectrum Disorder (ASD) are at risk for early social and communication difficulties, including language delays and ASD.

Objectives: To compare 12-month-old infants at high risk for ASD (HR-SIB) and infants without genetic liability for ASD (low risk, LR) on measures of verbal and nonverbal development as well as their social-communicative functioning.  The toddler version of the ADOS (ADOS-T; Lord et al., 2008) is designed for toddlers ages 12 to 30 months and quantifies delays and abnormalities in communication, social interaction, and presence of repetitive behaviors. 

Methods: 35 HR-SIB infants and 26 LR infants were assessed at 12 months with the Mullen Scales of Early Learning and the ADOS-T by psychologists blind to infants’ risk status.  Verbal and Nonverbal DQ scores were derived from the Mullen Scales.  Two indices were obtained from the ADOS-T: A subset of 14 items comprises the Social-Affective/Restricted Repetitive Behavior (SA/RRB) algorithm (range=0-28), and the Joint Attention (JA) algorithm (Gotham et al., 2007) includes 5 items (range=0-10).  The diagnostic status of HR-SIB infants was still unknown at the time of this submission. 

Results: Language abilities of the HR group were significantly lower (Mullen VDQ mean=80.4, SD=18.0) than those of the LR group (VDQ mean=91.4, SD 17.5), but there were no group differences on nonverbal scales. VDQ was negatively correlated with ADOS-T scores both in the LR group (r=.55, p<.001) and in the HR group (r=.40, p<.05).  ADOS-T SA/RRB algorithm scores were significantly higher (denoting more atypical presentation) in the HR-SIB group (mean=12.0, SD=5.7) compared to the LR group (mean=8.5, SD=4.1).  Furthermore, on the JA algorithm, the HR-SIB group also scored higher (mean=5.9, SD=2.8) than LR infants (mean=3.9, SD=1.7). The group differences on algorithm scores were not attributable to a few infants with extremely high scores. Analysis of algorithm items showed that HR-SIBs engaged in significantly fewer showing behaviors, were less likely to respond to joint attention bids, and directed fewer vocalizations and facial expressions to others than LR infants.  Exploration of other ADOS-T items revealed that infants in the HR-SIB group were less likely to respond to a playful tease, spent less time maintaining/directing parents’ attention, and had less developed functional play skills than the LR infants.  There were no group differences on restricted and repetitive behaviors.

Conclusions: Results suggest that compared to low-risk infants, infants at high risk for ASD demonstrate less robust early-emerging social-communication skills.  Delays in both nonverbal social communication and language skills were evident by the first birthday. In both low-risk and high-risk infants, lower verbal DQ scores were associated with social/communication difficulties measured by the ADOS-T.  The findings suggest that, as a group, high-risk 12-month-old infants demonstrate vulnerabilities in domains of social interaction, communication, and verbal ability.  We are currently in the process of collecting data at 24 months on the entire cohort at which point we will examine the predictive validity of performance at 12 months with regard to diagnostic outcome and overall levels of social, communication, and language development.