Autism is an extensively studied disorder in which the gender disparity in prevalence has received much attention. In contrast, only a few studies examine gender differences in symptomatology. In a recent systematic review and meta-analysis (Van Wijngaarden-Cremers et al. 2013) of 20 peer-reviewed original publications we examined gender differences in the core triad of impairments in autism. Gender differences were transformed and concatenated using standardised mean differences, and analyses were stratified in five age categories (toddlerhood, preschool children, childhood, adolescence, young adulthood).

Boys showed more repetitive and stereotyped behaviour as from the age of six, but not below the age of six. Males and females did not differ in social behaviour and communication. Girls with ASD and an average to high intelligence are strongly underrepresented this could imply that females present another autistic phenotype than males.

In this presentation we will elaborate on the results of this study. As ASD is now defined according to the male phenotype this could imply that there is an ascertainment bias. We will show which research is needed into the female phenotype of ASD with development of appropriate instruments (eg ASSQ) to detect and ascertain them in clinical practice. Thus preventing women with ASD to be misdiagnosed often for years before getting the right diagnosis and treatment.