Caregiver perception of hand function in infants with cerebral palsy: psychometric properties of the Infant Motor Activity Log

AIM To evaluate the properties of the Infant Motor Activity Log (IMAL), a caregiver-report for frequency and quality of use of more affected upper extremity in infants with neurological and functional impairments. METHOD This was a prospective cohort study of 66 children (34 females, 32 males) aged 6 to 24 months (mean age [SD] 13.7mo [5.3]) with neurological and functional impairments and a confirmed cerebral palsy diagnoses after 2 years, and 51 age-matched typically developing children. The IMAL was administered at baseline and 4 weeks later. Typically developing infants were tested with randomly assigned 'more affected' upper extremity. Psychometric properties were evaluated using Spearman's correlation coefficient, Cronbach's alpha, and Jonckheere-Terpstra tests. RESULTS In the children with impairments, the IMAL showed internal consistency (alpha≥0.88) for the How Well Scale (HWS) and How Often Scale (HOS). Test-retest reliability was 0.64 (HOS) and 0.70 (HWS), demonstrating stability over time. Correlation with Bayley Scales of Infant and Toddler Development, Third Edition more affected arm raw scores were 0.70 (HOS) and 0.72 (HWS) (p<0.001) demonstrating construct validity. Both scale scores decreased with increasing Gross Motor Function Classification System and Mini-Manual Ability Classification System (p<0.001) levels, supporting discriminative validity. Discrimination between typically developing infants and infants with impairments was high (HWS: area under the receiver operating characteristic curve [AUC] 0.96, 95% confidence interval [CI] 0.94-0.99 and HOS AUC=0.95, CI 0.92-0.99). INTERPRETATION The IMAL is a valid and reliable discriminative caregiver measure of upper limb performance and may complement measures of capacity in infants with neurological and functional impairments.