Gender differences in autism spectrum disorders: Divergence among specific core symptoms. Issue 4 (3rd November 2016)

• Record Type: Journal Article
• Authors: Beggiato, Anita; Peyre, Hugo; Maruani, Anna; Scheid, Isabelle; Rastam, Maria; Amsellem, Frederique; Gillberg, Carina I.; Leboyer, Marion; Bourgeron, Thomas; Gillberg, Christopher; Delorme, Richard
• Is Part Of: Autism research. Volume 10:Issue 4(2017)
• Page Start: 680
• Page End: 689
• Publication Date: 2016-11-03
• Subjects: autism spectrum disorders -- sex -- gender -- autism diagnosis interview‐revised; Autism -- Periodicals; Autism -- Research -- Periodicals; 616.85882005
• DOI: 10.1002/aur.1715
• Languages: English
• ISSNs: 1939-3792

Abstract

Abstract : Community‐based studies have consistently shown a sex ratio heavily skewed towards males in autism spectrum disorders (ASD). The factors underlying this predominance of males are largely unknown, but the way girls score on standardized categorical diagnostic tools might account for the underrecognition of ASD in girls. Despite the existence of different norms for boys and girls with ASD on several major screening tests, the algorithm of the Autism Diagnosis Interview‐Revised (ADI‐R) has not been reformulated. The aim of our study was to investigate which ADI‐R items discriminate between males and females, and to evaluate their weighting in the final diagnosis of autism. We then conducted discriminant analysis (DA) on a sample of 594 probands including 129 females with ASD, recruited by the Paris Autism Research International Sibpair (PARIS) Study. A replication analysis was run on an independent sample of 1716 probands including 338 females with ASD, recruited through the Autism Genetics Resource Exchange (AGRE) program. Entering the raw scores for all ADI‐R items as independent variables, the DA correctly classified 78.9% of males and 72.9% of females ( P < 0.001) in the PARIS cohort, and 72.2% of males and 68.3% of females ( P < 0.0001) in the AGRE cohort. Among the items extracted by the stepwise DA, four belonged to the ADI‐R algorithm used for the final diagnosis of ASD. In conclusion, several items of the ADI‐R that are taken into account in the diagnosis of autism significantly differentiates between males and females. The potential gender bias thus induced may participate in the underestimation of the prevalence of ASD in females. Autism Res 2016, . © 2016 International Society for Autism Research, Wiley Periodicals, Inc. Autism Res 2017, 10: 680–689 . © 2016 International Society for Autism Research, Wiley Periodicals, Inc.