Accuracy of motor assessment in the diagnosis of fetal alcohol spectrum disorder.

BACKGROUND: To evaluate the accuracy of motor assessment tools listed in Fetal alcohol spectrum disorder: a guideline for diagnosis across the lifespan (Canadian Guideline) for the purpose of fetal alcohol spectrum disorder (FASD) diagnosis. Specifically, we aimed to determine: 1) diagnostic accuracy of motor assessment tools and subtests; 2) accuracy of multiple subtests versus total scores; and 3) accuracy of alternate cut-offs. METHODS: Cross-sectional diagnostic study of 63 children aged 6-17 years. Diagnostic accuracy and alternate cut-offs were calculated for the Movement Assessment Battery for Children, 2nd edition (MABC-2), Bruininks-Oseretsky Test of Motor Proficiency, 2nd edition Short Form (BOT-2SF) and Beery-Buktenica Developmental Test of Visual Motor Integration, 6th edition (BeeryVMI-6). RESULTS: The MABC-2 total motor score was more sensitive (0.30; 95% CI 0.17-0.46; p < 0.01) to motor impairment in the presence of FASD than the BOT-2SF (0.02; 95% CI 0.00-0.12) at the 2nd percentile (-2SD). The MABC-2 total motor score was more accurate than any combination of subtest scores. The Motor Coordination subtest of the BeeryVMI-6 (BeeryMC) at the 5th percentile (- 1.5SD) (sensitivity 0.68, specificity 0.90) was the most accurate subtest. CONCLUSIONS: The BOT-2SF was an inaccurate assessment tool for FASD diagnosis. The MABC-2 total motor score was the most accurate using current guidelines, though its sensitivity was still low. Further investigation into inclusion of single subtests and/or using a less conservative cut-off in the Canadian Guideline is warranted.