Cross-Cultural Study of Screening Instruments

In The Politics of Autism, I explain that professionals cannot yet reliably diagnose autism through blood tests or brain scans, so they have to rely on observation of behavior.  There are multiple problems with this approach, including cultural differences.

Moreover, the definition of behavioral symptoms rests on Western – and specifically American – expectations about the behavior of children and adults. But what is “autistic” in this context may be “normal” in another. In rural South Africa, for instance, young children avoid direct eye contact with adults because their culture considers it disrespectful.

Sophie Carruthers and colleagues have an article at Molecular Autism titled  "A Cross-Cultural Study of Autistic Traits Across India, Japan and the UK."  The abstract:

Background: There is a global need for brief screening instruments that can identify key indicators for autism to support frontline professionals in their referral decision-making. Although a universal set of conditions, there may be subtle differences in expression, identification and reporting of autistic traits across cultures. In order to assess the potential for any measure for cross-cultural screening use, it is important to understand the relative performance of such measures in different cultures. Our study aimed to identify the items on the Autism Spectrum Quotient (AQ)-Child that are most predictive of an autism diagnosis among children aged 4–9 years across samples from India, Japan and the UK. 

Methods: We analysed parent-reported AQ-Child data from India (73 children with an autism diagnosis and 81 neurotypical children), Japan (116 children with autism and 190 neurotypical children) and the UK (488 children with autism and 532 neurotypical children). None of the children had a reported existing diagnosis of intellectual disability. Discrimination indices (DI) and positive predictive values (PPV) were used to identify the most predictive items in each country.

Results: Sixteen items in the Indian sample, 15 items in the Japanese sample and 28 items in the UK sample demonstrated excellent discriminatory power (DI ≥ 0.5 and PPV ≥ 0.7), suggesting these items represent the strongest indicators for predicting an autism diagnosis within these countries. Across cultures, good performing items were largely overlapping, with five key indicator items appearing across all three countries (can easily keep track of several different people’s conversations, enjoys social chit-chat, knows how to tell if someone listening to him/her is getting bored, good at social chit-chat, finds it difficult to work out people’s intentions). Four items indicated potential cultural differences. One item was highly discriminative in Japan but poorly discriminative (DI < 0.3) in the UK and India, and a further item had excellent discrimination properties in the UK but poorly discriminated in the Indian and Japanese samples. Two additional items were highly discriminative in two cultures but poor in the third.

Conclusions: Cross-cultural overlap in the items most predictive of an autism diagnosis supports the general notion of universality in autistic traits whilst also highlighting that there can be cultural differences associated with certain autistic traits. These findings have the potential to inform the development of a brief global screening tool for autism. Further development and evaluation work is needed.