Pediatric Physical Therapy:
Abstracts: Abstracts of Platform and Poster Presentations for the 2006 Combined Sections Meeting: Platform Presentations
Kolobe, Thubi A.1; Bulanda, Michelle2
1Rehabilitation Science, University of Oklahoma Health Science Center, Oklahoma City, OK, USA (Kolobe)
2Physical Therapy, University of Illinois at Chicago, Chicago, IL, USA (Bulanda)
The purpose of this study was to examine the diagnostic ability of the Alberta Infant Motor Scale (AIMS) when used with infants during the first year of life. More specifically, the study examined whether the cut-off points or diagnostic criteria for the AIMS is consistent in correctly classifying infants at various ages and stages of development.
Number of Subjects:
Sixty-one children who were preschool age participated in a longitudinal validation study of the AIMS. The children were from an original cohort of 90 infants who were assessed on the AIMS at 3, 6, 9, and 12 months adjusted age.
The Peabody Developmental Motor Scale, 2nd Edition (PDMS-2) were used to assess the childrens motor performance between the ages of 4-5 years old. Medical diagnoses confirmed the childrens developmental status. Inter- and intra-tester reliabilities were established among the testers on the PDMS-2. The children were tested in their homes. We used the Pearson product moment correlation coefficient and multiple regressions to assess the relationship between the AIMS and PDMS-2. To determine the diagnostic and predictive ability of the AIMS we calculated sensitivity, specificity and predictive values (diagnostic values) using the 5th and 10th percentile cut-offs on the AIMS.
Correlation coefficients between the AIMS percentile scores at 3, 6, 9, 12 months and the three subscales of the PDMS-2 were statistically significant. The coefficient between the AIMS and the PDMS-2 gross motor quotient (GMQ) was low at three months (r = 0.29), but good at 6, 9, and 13 months (r = 0.59, 0.60, and 0.65, respectively). The AIMS at 3 months had the lowest diagnostic values, however these values increased at 6, 9, and 12 months of age. Sensitivity and positive predictive values ranged from 0.33 at 3 months to 0.82 at 12 months. The 5th percentile cut-off correctly classified a higher number of children with or without disabilities than the 10th percentile cut-off.
The AIMS is more likely to correctly identify an infant with motor disabilities at 6, 9, and 12 than at 3 months. The scores on the AIMS at 3 months misclassified approximately 60% of infants who were later diagnosed with CP. The AIMS 5th percentile cut-off is more likely to correctly classify children at 6, 9, and 12 months who will later be diagnosed with CP, than the 10th percentile cut-off.
Physical therapists play a major role in developmental surveillance of infants and young children. Research findings about the instability of motor development raise questions about the diagnostic accuracy of many of the widely used developmental and screening tests that are based on cross-sectional data. The results of this study provide information that therapists can use at several ages during infants first year to base their decisions about referral.
© 2006 Lippincott Williams & Wilkins, Inc.