INTRODUCTION: Etiology remains unknown in 30% to 50% cases of children with global developmental delay (GDD). A selective approach has been recommended to increase etiologic yield.
OBJECTIVE: Our aim was to identify clinical markers that enhance diagnostic yield of GDD at initial assessment.
METHODS: The charts of all patients with GDD (N = 577) followed up at the Duchess of Kent Child Assessment Centre were reviewed. GDD was defined as significant delay (<2 SD) in ≥2 developmental domains. Nine clinical items at initial assessment (gender, severity of delay, parental consanguinity, family history, behavioral disturbance, head size, facial dysmorphism, malformations, and neurologic deficits) were correlated with the likelihood of finding an etiology for GDD.
RESULTS: A significant threshold effect was found between mild and moderate GDD (positive likelihood ratio (LR+]: 1.9; negative likelihood ratio [LR−]: 0.72). Other items that significantly affected diagnostic yield were (1) female gender (LR+: 1.62; LR−: 0.79), (2) behavioral trait (LR+: 0.24; LR−: 1.67), (3) microcephaly (LR+: 2.78; LR−: 0.79), (4) presence of facial dysmorphism (LR+: 2.65; LR−: 0.65), (5) malformation (LR+: 1.49; LR−: 0.50), and (6) neurologic deficits (LR+: 2.86; LR−: 0.32). A dose-response relationship was found between LR+ and the number of facial dysmorphisms and malformations.
CONCLUSIONS: Most checklists used for GDD are syndrome specific (eg, fragile X syndrome checklists). These 7 clinical markers are useful in the initial assessment, even if no specific diagnosis is suspected. Unique statistical characteristics of LRs allow for a wide application in different clinical settings.
Submitted by B. H. Y. Chung
- Copyright © 2008 by the American Academy of Pediatrics