As adaptive skills (AS) are dynamic and may indicate the success of an intervention, they are a common domain measured in clinical trials. Typical interview tools for measuring AS are time-consuming, and questionnaire measures often lead to inconsistent information. The present study was designed to evaluate the feasibility, validity and test-retest performance of the Pediatric Evaluation of Disability Inventory Computer Adaptive Test (PEDI-CAT) in Fragile X syndrome (FXS). The PEDI-CAT is administered via tablet and uses the item response theory to efficiently determine the items administered. The PEDI-CAT was administered to 42 individuals with FXS (27 males; 15 females) aged 1.6–50.9 years (M = 14.9; SD = 11.2), followed by the Vineland-3 (VABS-3) interview for comparison. Administration was efficient (M = 21.7 min; SD = 9.5; range 8–45 min; mode = 19). Males and females did not significantly differ on the PEDI-CAT domains, except for daily activities (t(40) = −2.22, p = 0.037). Floor effects were significant for both measures, although the PEDI-CAT showed more floor effects in the mobility (35.7%) and social-cognitive (50%) domains. PEDI-CAT daily activities, mobility, social-cognitive and responsibility domains were all significantly correlated with most of the VABS-3 domains (all rho > 0.5; p < 0.01). Test-rest of the PEDI-CAT was comparable to the VABS-3. Results suggest that the PEDI-CAT is efficient, and minimal training is needed to administer it; however, it lacks specificity and shares a high rate of floor effects with the VABS-3.
An efficient and direct measure of social interactions and autism symptoms is needed for fragile X syndrome (FXS) research and clinical care. The Autism Screening Instrument for Educational Planning-Third Edition (ASIEP-3) Interaction assessment is a brief standardized measure that quantifies social responses under different conditions. The feasibility and validity of the ASIEP-3 was evaluated in 26 males and 13 females with FXS, along with cognitive testing and behavior questionnaires. The videos were scored at 10-second intervals, and the observed behaviors were scored as an interaction, independent play, no response, or aggression. In total, 39/41 participants successfully completed the ASIEP-3 (age M = 14.4 ± 10.2), with a range of cognitive abilities (abbreviated IQ (ABIQ) M = 58.9 ± 17.3, median = 50), behaviors (Aberrant Behavior Checklist (ABC) Total M = 37.00 ± 27.3), and autism diagnoses (N = 22/39). Reliable administration was demonstrated by all team members. The mean coded behaviors included interaction (40.6%), independent play (36.8%), no response (21.1%), and aggressive behavior (<10%). The interaction score was negatively correlated with the Social Communication Questionnaire (SCQ) score (p = 0.037), and the profiles differed by autism spectrum disorder (ASD) diagnosis. The intraclass correlation coefficients (ICCs) ranged from 0.79 to 0.93 for master’s level and above. Administration of the ASIEP-3 was feasible for FXS across sex, age, ability, and behavior ratings by a trained research team. Reliable scoring required advanced training in the assessment of social development and FXS experience. The scores correlated to ratings and diagnoses of ASD. The ASIEP-3 shows promise to reliably index social interactions in FXS.