Early gross motor development: Agreement between the AIMS and the BSID-III

Abstract

Background: Early gross motor development is a crucial indicator of overall neurodevelopment. In low- and middle-income countries, lack of accessible assessment tools poses challenges for healthcare professionals evaluating infant neurodevelopment.

Objectives: To determine the agreement between the Alberta Infant Motor Scale (AIMS) and Bayley Scales of Infant Development-III (BSID-III) gross motor domain at 6 months and to evaluate the predictive validity of the AIMS at 6 months for identifying severe gross motor delays at 18 months.

Method: This nested subgroup study assessed 112 full-term infants using both AIMS and BSID-III at 6 months and BSID-III at 18 months. Agreement between measures was determined using Bland-Altman plots, while predictive validity was evaluated using receiver operating characteristic (ROC) curves with various cut-off scores.

Results: Bland-Altman analysis showed strong agreement between AIMS and BSID-III in the lower-performance range, with bias only in scores above 33. The traditional 10th percentile AIMS cut-off had low sensitivity (27.3%) but high specificity (98%) for predicting delays at 18 months. A modified 23rd percentile cut-off improved sensitivity to 63.6% while maintaining acceptable specificity (81.6%), with a 95.2% negative predictive value (NPV).

Conclusion: The AIMS demonstrates strong agreement with BSID-III when identifying potential developmental delays. The proposed 23rd percentile cut-off offers a more balanced screening threshold for this population.

Clinical Implications: The AIMS presents a viable alternative to the BSID-III for initial screening in resource-limited settings. The high NPV at the 23rd percentile cut-off makes it useful for ruling out developmental delays.