Predictability of the Social Communication Questionnaire (SCQ) on Autism Diagnoses from a Community Study in Taiwan

Background: The Social Communication Questionnaire (SCQ) is a screening instrument for Autism Spectrum Disorder (ASD).  The screener is designed to match the Autism Diagnostic Interview-Revised (ADI-R).  Studies conducted in Western countries that were designed as clinic-based or for high-risk populations (e.g., children in special education) reported strong indices of validity against the ADI-R.  Because the ADI-R is subjective to a respondent’s cultural bias, especially respondents who are from cultures different from that of Western populations, it may not be the best choice for diagnostic instrument in international studies.  In light of the need for information from both caregiver’s report and direct observation of the child in order to make an accurate diagnosis, a combination of the SCQ and ADOS can be considered for studies implemented in international settings. 

Objectives: The purpose of this study is to examine the predictive ability of individual items on the SCQ, and different cut-offs of the SCQ, on ADI-R, ADOS, and clinician-determined ASD diagnoses.

Methods: Study participants were drawn from a community-based autism study in Taiwan, where school children aged 6-8 were screened using the SCQ, followed by diagnostic evaluation using the ADOS, ADI-R, and a comprehensive evaluation by a child psychiatrist based on DSM-IV criteria if the SCQ is >=7.  The study sample comprised 27 boys and 19 girls, and their primary caregivers, who completed the SCQ, ADOS, ADI-R, and who were evaluated by a child psychiatrist.  Three study groups were classified based on the ADOS: ASD (n=12), subclinical (n=8), and unaffected (n=25).  Similarly, 3 groups were defined based on the ADI-R: autism (n=4), subclinical (n=6), and unaffected (n=25).  The 3 groups based on a child psychiatrist’s clinical judgment were: ASD (n=7), other DD (n=24), and unaffected with ASD or other DD (n=15). Three cut-offs for the SCQ were used to define higher versus lower risk for ASD: 11, 15, and 22.

Results: Overall, the preliminary findings showed individual items of the SCQ, and the three SCQ cut-offs (11, 15, and 22) predict ADI-R diagnoses better than predicting ADOS diagnoses and clinical diagnosis made by a child psychiatrist. 

Conclusions: It is not surprising that the SCQ better predicts the ADI-R classification than the ADOS or child psychiatrist’s clinical judgment, as SCQ items were chosen from the ADI-R.  Although some items significantly predict diagnostic classification, the majority of items and different cut-offs are not good predictors and do not discriminate diagnoses within each of the diagnostic evaluations (i.e., ADOS, ADI-R, and DSM-IV-based clinical judgment).  This finding implies that the SCQ, when used alone, may not be a great screener for community-based or population-based autism studies.  However, these results need to be interpreted with caution as the sample sizes for many subgroups are too small to provide sufficient statistical power to detect a statistically significant difference.