Purpose Many tools are available for assessing autism in an adult population; however, few have been studied for the effects of gender on diagnostic scores. The purpose of this paper is to evaluate the Autism Diagnostic Observation Schedule (ADOS) assessment for gender bias in a clinical population, specifically whether the ADOS favours a “male-type” of autism. Design/methodology/approach The ADOS scores of patients referred to an NHS specialist autism assessment service were retrospectively examined for significant gender differences. The combined ADOS scores and diagnostic outcome were grouped by gender for each participant. The data were analysed in SPSS using independent t-tests to look for significant gender differences between combined ADOS scores and diagnostic outcomes. Findings A significant difference was observed in the mean combined ADOS scores for those participants who later received an autism diagnosis (male=10, female=6, t (13)=3.34, p=10; 0.005). However, no significant difference was observed between mean scores of those who did not receive an autism diagnosis (t (26)=1.21, p=0.237). Originality/value The ADOS is a popular assessment used for autism diagnosis. These results provide support for a male gender bias. This could have clinical implications for autism assessment services, whereby lower diagnostic thresholds could be considered for female patients. This could allow more females with autism to receive a diagnosis, and access support services.

Adult referrals to specialist autism spectrum disorder diagnostic services have increased in recent years, placing strain on existing services. It was proposed that the Ritvo Autism Asperger’s Diagnostic Scale could be used as a screening tool, in order to identify and prioritise patients most likely to receive an ASD diagnosis. This study evaluates the validity of the RAADS-R as a screening tool for ASD in an adult population. Retrospective case note analysis was used to evaluate the efficacy of the RAADS-R as a screening tool to predict ASD diagnostic outcomes in 50 service users of a NHS specialist autism service. Results indicate no association between RAADS-R scores and clinical diagnostic outcome, suggesting the RAADS-R is not an effective screening tool for identifying service users most likely to receive an ASD diagnosis. In conclusion, used as a self-report measure pre-full diagnostic assessment, the RAADS-R lacks predictive validity and is not a suitable screening tool for adults awaiting autism assessments. Future research should aim to identify reliable screening tools for this purpose.