Association of elevated neonatal thyroid-stimulating hormone levels with school performance and stimulant prescription for attention deficit hyperactivity disorder in childhood.

Abstract

Untreated severe newborn thyroid deficiency causes neurocognitive impairment; however, the impact of mild thyroid deficiency is not known. This study aimed to examine whether mildly elevated neonatal thyroid-stimulating hormone (TSH) levels are associated with poor school performance or stimulant prescription for attention deficit hyperactivity disorder (ADHD). This record-linkage study included 232,790 term-born infants in Australia with a TSH level below newborn screening threshold (< 15mIU/L). Among our cohort, as TSH levels increased, the proportion of infants born low birthweight via caesarean section and with disadvantaged socioeconomic status increased. Multivariable logistic regression analysis showed that, compared with infants with 'normal' neonatal TSH level (< 5mIU/L), those with neonatal TSH 10-15mIU/L had an increased risk of being exempt from school testing (aOR 1.63 (95% CI 1.06-2.69)) or prescribed a stimulant for ADHD (aOR 1.57 (95% CI 1.10-2.24)), adjusted for perinatal and sociodemographic factors. Among a nested analysis of 460 sibling pairs, siblings with 'mildly elevated' TSH levels were more likely to be exempt from school tests compared with siblings with normal TSH levels (aOR 2.53, 95% CI 1.01-6.33).Conclusion: In this population cohort and sibling analysis, mildly elevated neonatal TSH levels were associated with being exempt from school testing due to significant or complex disability. What is Known: • Newborn screening for severe thyroid hormone deficiency has virtually eliminated congenital hypothyroidism-associated intellectual disability in developed countries. • The impact of mild thyroid hormone deficiency in infants is unclear. What is New: • Children with a mildly elevated neonatal TSH level below current newborn screening cut-offs have an increased likelihood of being exempt from school testing due to significant or complex disability compared with siblings and peers. This study includes a population-based and nested sibling analysis.