International Meeting for Autism Research: Sex Differences in Siblings of Children with ASD at 3 Years

Sex Differences in Siblings of Children with ASD at 3 Years

Saturday, May 22, 2010: 3:00 PM
Grand Ballroom E Level 5 (Philadelphia Marriott Downtown)
1:15 PM
W. Roberts , Department of Pediatrics, Hospital for Sick Children & Bloorview Kids Rehab, University of Toronto, Toronto, ON, Canada
L. Zwaigenbaum , Department of Pediatrics, University of Alberta, Edmonton, AB, Canada
J. Brian , Autism Research Unit & Bloorview Research Institute, Hospital for Sick Children & Bloorview Kids Rehab, Toronto, ON, Canada
C. Roncadin , Psychology, Peel Children's Centre, Mississauga, ON, Canada
I. M. Smith , Pediatrics & Psychology, Dalhousie University & IWK Health Centre, Halifax, NS, Canada
P. Szatmari , The Offord Centre for Child Studies, McMaster University, Hamilton, ON, Canada
T. Vaillancourt , University of Ottawa, Ottawa, ON, Canada
S. E. Bryson , Pediatrics and Psychology, Dalhousie University/IWK Health Centre, Halifax, NS, Canada
Background: Most epidemiological studies have shown a preponderance of males among children with autism spectrum disorder (ASD), with male-to-female ratios ranging from 3.5:1 to 4:1. However, some authors have proposed that milder social and behavioural impairments may impede identification of ASD in girls. Objectives: 1)To compare rates of ASD in boys and girls in a high-risk cohort of infant siblings of children with ASD (hereafter, ‘high-risk infants'), and 2) by looking at the entire cohort, to understand the sex differences in light of their respective developmental trajectories of ASD symptoms and cognitive function.

Methods: ASD diagnoses at age 3 years were based on ADI-R, ADOS and DSM-IV criteria applied by expert clinicians blind to prior study data. Early ASD symptoms were assessed from ages 6 to 18 months using total scores from the Autism Observational Scale for Infants (AOSI). Cognitive development was assessed from ages 6 months to 3 years using the Mullen Scales of Early Learning, Early Learning Composite scores (Mullen-ELC). Developmental trajectories of ASD symptoms and cognitive function were identified independently by semi-parametric group-based modeling using SAS (PROC TRAJ) in our total sample of 401 high-risk and 160 low-risk infants. Sex differences in trajectory membership were then assessed among the 225 high-risk and 91 low-risk infants followed to age 3 years.

Results: ASD rates at age 3 years were similar in boys and girls. Seventeen of 112 boys (15.2%) and 12 of 99 girls (12.1%) received a diagnosis of ASD at 36-42 months. Relative odds of ASD in boys compared to girls was 1.25 (95% CI = 0.63-2.49), robust to age of recruitment and sex of the older affected sibling (in whom the sex ratio was 4:1). Between 6 and 18 months, boys and girls had similar distributions in AOSI symptom trajectories. However, significant sex differences were found between 6 months and 3 years in Mullen-ELC trajectory membership. Three distinct trajectories were identified: two stable trajectories characterized by average and above average scores respectively, and a third (10.2 % of the total sample including 13.4% of the high-risk group), characterized by marked decline from average to intellectually disabled by age 3. High-risk boys were over-represented in the third trajectory (15.7% vs. 5%; χ2 = 21.5, p<.001), as well as among those diagnosed with ASD (52.6% boys vs. 27.8% girls, p = .07). Findings were similar for verbal and non-verbal cognitive trajectories when examined separately.

Conclusions: In our high-risk cohort, preliminary evidence suggests that 1) there are similar rates of ASD in boys and girls at age 3; and 2) girls with ASD tend to show more stable cognitive trajectories than boys. The lack of clear early sex differences in rates and cognitive levels raises interesting questions about whether trajectories diverge after age 3, with some girls no longer meeting ASD criteria as they get older.