Validation of neurodevelopmental assessments for early detection of high-risk infants in Zimbabwe: protocol for the Child Development Observation study (ChiDO study).

INTRODUCTION: Appropriate neurodevelopmental assessment tools are often unavailable in low- and middle-income countries (LMICs) due to lack of validation, cost and cultural-linguistic challenges. This reduces the availability of high-quality developmental evaluation, resulting in clinical gaps that impact individual child outcomes and modifiable risk detection for prevention. The objective of this study is to validate three neurodevelopmental assessment tools for early detection in low-resource settings through recruitment of infants across the neurodevelopmental impairment (NDI) risk spectrum related to perinatal asphyxia and neonatal encephalopathy (NE). METHODS: This prospective, longitudinal cohort study in Harare, Zimbabwe will collect data at 3 months, 6 months, 12 months, 18 months and 24 months. The sample of approximately 600 caregiver-infant dyads will include: (1) infants without clinical concern for perinatal asphyxia or other complications (low risk); (2) infants who do not cry at birth and require resuscitation but do not have NE (moderate risk) and (3) infants who experience NE (high risk). In addition to demographic, antenatal and perinatal data collected at enrolment, neurodevelopmental assessments to be conducted include the Prechtl General Movement Assessment (GMA); Hammersmith Infant Neurological Examination (HINE); Global Scales for Early Development (GSED); Mullen Scales of Early Learning (MSEL) and caregiver-reported Ages and Stages Questionnaire. At the 24-month visit, a comprehensive diagnostic evaluation for NDI will be conducted adhering to international guidance. Predictive ability of all tools will be assessed by receiver operating characteristic (ROC) analyses, including area under the ROC curve. Construct validity of the GSED will be assessed by comparing to the MSEL. The GSED's ability to differentiate subgroups defined by correlates of NDI risk will be assessed using generalised estimating equations. ETHICS AND DISSEMINATION: Ethical approvals have been obtained. If shown to be valid, the GMA and HINE could provide earlier, feasible and low-cost detection of children at elevated risk for NDI.