The Quantitative Checklist for Autism in Toddlers (Q-CHAT) and Q-CHAT-10: A psychometric study in Chilean Toddlers

The aim of this study was to examine the psychometric properties of a culturally adapted version for Chile of the Quantitative Checklist for Autism in Toddlers (Q-CHAT Full item) and its abbreviated version (the Q-CHAT-10) in a community sample of Chilean children aged 18 to 24 months. An intentional non-probabilistic sampling was used. One hundred eighty-eight young children were examined. Evidence of construct validity was determined through Exploratory Factor Analysis (EFA), evidence of convergent validity with the Modified Checklist for Autism in Toddlers, Revised with Follow-up (M-CHAT-R/F), evidence of reliability through internal consistency and test-retest and finally, evidence of predictive validity through ROC curve analysis. The EFA showed a two-factor structure: "Communication and social interaction" (factor 1) and "Restrictive and repetitive patterns" (factor 2) and explained 37.7% of the variance. This factor structure excluded item 18, thus structuring a narrower version: the Q-CHAT-24. The total scores of the Q-CHAT Full Item, the Q-CHAT-24 and the Q-CHAT-10 correlated positively and significantly with the total scores of the M-CHAT-R/F (r=0.59; r=0.63; and r=0.79, respectively). The internal consistency was acceptable for the Q-CHAT complete item (α=0.74), for the Q-CHAT-24 (α=0.76) and for the Q-CHAT-10 (α=0.75). The internal consistencies of the Q-CHAT-24 factors were good for factor 1 (α=0.86) and acceptable for factor 2 (Cronbach ́s α=0.76). Regarding test-retest stability, the intraclass correlation coefficient was good (ICC=0.86, p<0.001). The cut-off points were established according to the optimal levels of Sensitivity/Specificity and Youden Index and were: 41 for the Q-CHAT Full item version (84.3/81.8, 0.70), 38 for the Q-CHAT-24 (93.7/81.8, 0.76), and 2 for the Q-CHAT-10 (87.5/90.7, 0.78). In accordance with the objectives of this study, evidence of reliability and validity was demonstrated for three versions of the Q-CHAT. The Q-CHAT-24, the version obtained by EFA from the Q-CHAT Full item, is phenomenologically concordant with the DSM-5 dimensional criteria for autism and is psychometrically adequate in terms of its predictive ability to be considered as a screening instrument in the general population. In the same vein, the Q-CHAT-10, a smaller version of the Q-CHAT Full item, stands out as a useful version for a general population setting, given its psychometric characteristics and length.