RESEARCH REPORT: PDF OnlyFetters Linda PhD PT; Tronick, Edward Z. PhDPediatric Physical Therapy: Spring 2000 - p 16-23 Free Abstract In this study, we compared the predictive power of the Alberta Infant Motor Scale (AIMS) and the Movement Assessment of Infants (MAI) administered at four and seven months for detection of gross motor delays at 15 months in a group of infants exposed in utero to multiple drugs, including cocaine. Gross motor delay was defined at 15 months as —1 standard deviation (12 subjects) or —2 standard deviations (seven subjects) below the mean on the Gross Motor Scale of the Peabody Developmental Motor Scales (PDMS-GM). The sample comprised 39 infants: 21 exposed and 18 unexposed. Sensitivity, specificity, and positive and negative predictive values were calculated for a range of percentiles on the AIMS and for more than four and more than nine risk points on the MAI. For both the MAI and the AIMS, the best combination of sensitivity and specificity values was at seven months. A cutoff value of more than nine risk points on the MAI and the second percentile on the AIMS offered the best combination of predictive values when predicting scores at —2 standard deviations below the mean on the PDMS-GM. Use of either the MAI or the AIMS alone is insufficient for correct prediction. The MAI overidentifies infants with motor problems in comparison to the AIMS, but neither test adequately identifies infants who go on to have poor motor scores on the PDMS-GM. The most prudent clinical course may be repeated testing over time if referral is not desired or if the criteria for referral are not clear. © 2000 Lippincott Williams & Wilkins, Inc.