Abstract
AIM: To determine the feasibility of combining the Hammersmith Infant Neurological Examination (HINE) and General Movements Assessment (GMA) within a standard follow-up schedule to predict developmental outcomes in infants at risk in low- and middle-income countries (LMICs). METHOD: A total of 201 Sri Lankan infants (128 male, 73 female) were prospectively assessed with the GMA before 44 weeks (writhing movements) and at 3 to 4 months (fidgeting movements), followed by the HINE at 5 to 6 months. Developmental outcomes were assessed using the Bayley Scales of Infant and Toddler Development, Fourth Edition and clinical assessment after 24 months. RESULTS: The sensitivity of predicting cerebral palsy (CP) was lower with a single GMA assessment (writhing 89.5%, fidgeting 94.7%) or HINE (89.5%) compared to all three assessments combined (sensitivity 100%, 95% confidence interval [CI] = 82.4-100.0). The GMA and HINE were less predictive of non-CP-related developmental delays, particularly when single assessments were used (< 65% for all domains) compared to all three assessments combined (motor sensitivity > 86.9%, 95% CI = 66.4-97.2; cognitive sensitivity > 86.7%, 95% CI = 69.3-96.2; social-emotional sensitivity > 83.3%, 95% CI = 65.3-94.4). Specificity was lower for the prediction of CP-related (40.1%) and non-CP-related developmental delays (< 46.0% for all). INTERPRETATION: In an LMIC such as Sri Lanka, with limited access to specialist care and neuroimaging, combining two GMA measures and the HINE identified most infants with CP-related and non-CP-related developmental delay, thereby allowing targeted early intervention therapies.