Book Part
Technologies for child development screening: The case of SIMEDID and its potential to trigger early interventions against disability risk
Sánchez-Vincitore, Laura V.
|
Castro, A.
|
Valdez, M. E.
|
Camilo, M.
|
Peterson, P.
|
Cubilla-Bonnetier, Daniel
| Fecha de publicación | Revista | DOI |
2 de julio de 2026 | Biosystems and Biorobotics |
Resumen
Early identification of developmental delays is critical for timely intervention, particularly in low- and middle-income countries where access to diagnostic services is limited. The Dominican Republic developed the Sistema de Medición de Desarrollo Infantil Dominicano (SIMEDID), a digital screening tool integrated into national early childhood services. We examined SIMEDID’s capacity to predict caregiver-reported disability using data from a probabilistic national sample of 2,060 children aged 2–60 months. Developmental outcomes (gross motor, fine motor, language, socioemotional, and overall scores) were compared by disability status through ANCOVAs controlling for age and sex. Children with disabilities (n = 16; 0.8%) scored significantly lower across all domains (η2 ≈ .001–.002, small effects). Logistic regression models adjusted for age and sex assessed predictive accuracy. Language scores showed the strongest performance, with 100% sensitivity (recall) but very low precision (~2%). Socioemotional scores reached 80% sensitivity with ~1% precision, while other domains had lower sensitivity (40–60%) and similarly low precision (1%). Precision was low because predictions were contrasted with caregiver-reported disability status at enrollment, which may underestimate true cases. Despite these limitations, SIMEDID demonstrates potential as a first-level alert system for developmental risk, supporting scalable surveillance and referral pathways in resource-limited contexts.