Psychometric profiles of the Ages and Stages Questionnaires, Japanese translation

BACKGROUND: This study assessed the psychometric profile of 10 questionnaires (every 6 months, from 6 to 60 months) from the Japanese translation of the Ages and Stages Questionnaires, third edition (J-ASQ-3). METHODS: Data from 439 children in a birth cohort were used to identify the J-ASQ-3 score distribution, establish cut-off scores, and calculate the instrument's internal consistency. Data were also collected from 491 outpatients to examine J-ASQ-3 test-retest reliability and concurrent validity, which was examined using the Kyoto Scale of Psychological Development (KSPD) and the Japanese version of the Denver Developmental Screening Test II (J-Denver II). Both the original and the alternative screening criteria of the ASQ-3 were used (failure in at least one and at least two domains, respectively). RESULTS: Cronbach's alpha for each J-ASQ-3 subscale on each questionnaire ranged from 0.45 to 0.89. Test-retest reliability was >0.75 for the subscales on almost all questionnaires. Concurrent validity was also adequate. In comparison with the screening results of the KSPD, the overall sensitivity and specificity were 96.0% and 48.8%, respectively, when the ASQ-3 original criterion was used, and 92.1% and 74.9%, respectively, when the alternative criterion was used. In comparison with the screening results of the J-Denver II, the overall sensitivity and specificity were 75.6% and 74.7%, respectively, when the ASQ-3 original criterion was used, and 56.3% and 93.0%, respectively, when the alternative criterion was used. CONCLUSIONS: This study quantified the psychometric profiles of the Japanese translations of 10 ASQ-3 questionnaires. We demonstrated the validity of the J-ASQ-3 and determined new cut-off scores. Further studies with larger samples from a greater range of locations are required to clarify the suitability of this tool for all Japanese children.