Measuring autistic traits in the general population: a systematic review of the Autism-Spectrum Quotient (AQ) in a nonclinical population sample of 6,900 typical adult males and females.

The Autism-Spectrum Quotient (AQ) is a self-report measure of autistic traits. It is frequently cited in diverse fields and has been administered to adults of at least average intelligence with autism and to nonclinical controls, as well as to clinical control groups such as those with schizophrenia, prosopagnosia, anorexia, and depression. However, there has been no empirical systematic review of the AQ since its inception in 2001. The present study reports a comprehensive systematic review of the literature to estimate a reliable mean AQ score in individuals without a diagnosis of an autism spectrum condition (ASC), in order to establish a reference norm for future studies. A systematic search of computerized databases was performed to identify studies that administered the AQ to nonclinical participant samples representing the adult male and female general population. Inclusion was based on a set of formalized criteria that evaluated the quality of the study, the usage of the AQ, and the population being assessed. After selection, 73 articles, detailing 6,934 nonclinical participants, as well as 1,963 matched clinical cases of ASC (from available cohorts within each individual study), were analyzed. Mean AQ score for the nonclinical population was 16.94 (95% CI 11.6, 20.0), while mean AQ score for the clinical population with ASC was found to be 35.19 (95% CI 27.6, 41.1). In addition, in the nonclinical population, a sex difference in autistic traits was found, although no sex difference in AQ score was seen in the clinical ASC population. These findings have implications for the study of autistic traits in the general population. Here, we confirm previous norms with more rigorous data and for the first time establish average AQ scores based on a systematic review, for populations of adult males and females with and without ASC. Finally, we advise future researchers to avoid risk of bias by carefully considering the recruitment strategy for both clinical and nonclinical groups and to demonstrate transparency by reporting recruitment methods for all participants.