Screening Children with Autism Spectrum Disorders (ASD) Using the Japanese Version of the Strengths and Difficulties Questionnaire (J-SDQ)

Background:  The Japanese version of the Strengths and Difficulties Questionnaire (J-SDQ) is a brief measure of behavioral and emotional problems in children and adolescents rated by parents. The J-SDQ has 25 items and can be allocated to five subscales of five items each: the Emotional Symptoms subscale, the Conduct Problems subscale, the Hyperactivity-Inattention subscale, the Peer Problems subscale, and the Prosocial Behavior subscale.

Objectives: This study included one main research question and two additional research questions. The main research question was: "Does the J-SDQ have sufficient power to discriminate children with Autism Spectrum Disorders (ASD) from those who are typically developing?" The additional research questions were: (a) What is the impact of cognitive function (i.e., IQ) on a power of discriminating children with ASD from those who are typically developing? and (b) What is the impact of age on a power of discrimination?

Methods: Participants were 180 children with ASD and 6,586 typically developing children in elementary and middle schools (i.e., 1st to 9th grade). All the J-SDQ scores were compared between the two groups by five subscales, and powers of discrimination were calculated by receiver operating characteristic (ROC) analyses. For (a) the first additional question, children whose IQ was below 70 was excluded from the analysis because more children with ASD usually function significantly lower than normally developing children, which might result in higher spurious discrimination power. For (b) the second research question, analysis was conducted with additional data of individuals who were over 17-year-old to examine the effect of age on the discrimination power.

Results: For the main research question, the Peer Problems subscale exhibited a significant and the largest discrimination power. Among five subcategory scores, there was a significant differences between the ASD group and the typically developing group in the area of the Peer Problems only. Furthermore, for the first and second additional  questions, there were no effects of cognitive function and age on powers of discriminating children with ASD from typically developing peers.

Conclusions: The efficiency of the Peer Problems subscale of the J-SDQ, which consist of only five items, to screen children with ASD among those who are typically developing was supported.