Can the UK AQ-10 be applicable to Chinese samples with autism spectrum disorder in Hong Kong? Cross-cultural similarities and differences.

The current study aimed at testing and developing alternative short versions of autism spectrum quotient (AQ-10) (adult [self-report], adult [parent-report], adolescent, and child versions) for use in Hong Kong. First, the various versions of AQ-10 developed in the United Kingdom (the AQ-10-UK) were applied to Hong Kong Chinese samples and demonstrated satisfactory discriminative power (AUCs 0.77-0.94). Second, the Hong Kong Chinese versions of AQ-10 (AQ-10-HK) were developed, using the same methodology as in the original UK study. There were some changes in the choice of items and cut-offs. The AQ-10-HK demonstrated slightly greater discriminative power (AUCs 0.88-0.97) to that of the AQ-10-UK, but the differences in AUCs were not statistically significant. Compared to the corresponding full-length versions, both the UK and HK short forms did not seem to lose any significant discriminative power. Yet, the various versions of AQ, be they the full-length or AQ-10, appeared to consistently exhibit slightly smaller AUCs with the Hong Kong Chinese samples than with the UK samples. So, this study found both cross-cultural similarities and differences. The AQ-10-HK was recommended for local practice to maximize the advantage gained. Yet, for international multi-site research collaboration, involving the UK and HK, the original AQ-10-UK can be used for direct comparison of data.

Comparing the Psychometric Properties of the Self- and Parent-Report Versions of Autism-Spectrum Quotient-Adult in Hong Kong (AQ-Adult-HK).

The Autism-Spectrum Quotient-Adult (AQ-Adult) is a screening tool for adults with autism spectrum disorder (ASD). The present study examined the psychometric properties of the Chinese self- and parent-report versions of the AQ-Adult in Hong Kong (AQ-Adult-HK). Participants included adults with ASD (n = 27) and community controls (n = 345). Parents of a subset of adults with ASD (n = 21) and controls (n = 87) also participated as informants. The parent-report version showed significantly stronger psychometric properties, including a larger area under receiver operating characteristic curve (AUC) and higher sensitivity/specificity, than those of the self-report version. The stronger psychometric properties of the former were related to its significantly higher ratings of ASD symptoms in the ASD adults.