Identifying developmental vulnerability through linear growth screening: a UK cross-sectional study
Michael Papasavva, Joanna Orr, Isabella Cordani, Lola Oloko, Celes Hutchinson, Christopher Norton, Kaur Kamaljit, Joseph Freer, Robert Walton, Helen L Storr, Melanie Smuk, Andrew J PrendergastIntroduction
Linear growth in early childhood is routinely measured in the UK, but it is unclear whether indicators of linear growth faltering can flag children with neurodevelopmental vulnerability, especially in ethnically diverse urban settings. We tested whether short stature (stunting) or being short for genetic potential identifies children at risk of failing routine developmental screening at age 2–2.5 years.
Methods
We conducted a cross-sectional analysis among 555 children aged 24–30 months (mean 26.0 months, SD 1.5) attending Healthy Child Programme reviews in East London. The sample was ethnically diverse, with substantial proportions of Bangladeshi, White and other minority ethnic groups. Developmental vulnerability was defined as scoring below the Ages and Stages Questionnaire, Third Edition (ASQ-3) cut-off in ≥1 domain. The exposures were stunting (height-for-age z score <−2 SD) and being short for genetic potential, defined as deviation from mid-parental height (DMPH) <−2 SD. Associations were examined using binomial multivariable logistic regression adjusted for age, sex, prematurity, maternal education, ethnicity and household benefits. A subset (n=94) completed the Griffiths III Mental Development Scales (GMDS-III) to validate ASQ-3 classification.
Results
Overall, 141/555 (25.4%) children screened positive for developmental vulnerability on the ASQ-3. Children with developmental vulnerability had significantly lower GMDS-III scaled scores across multiple domains supporting criterion validity. We did not find evidence of an association between stunting and developmental vulnerability (adjusted OR 1.35, 95% CI 0.47 to 3.49). In contrast, children who were short for genetic potential (DMPH <−2 SD) had higher odds of developmental vulnerability (adjusted OR 4.18, 95% CI 1.13 to 14.78).
Conclusions
In this urban UK cohort, deviation from genetic height potential, rather than stunting, identified children at increased risk of developmental vulnerability. Incorporating parental heights into routine growth surveillance could offer a simple trigger for earlier neurodevelopmental review.