Incidence of congenital hypothyroidism: retrospective study of neonatal laboratory screening versus clinical symptoms as indicators leading to diagnosis.

Filter paper blood samples taken routinely from 100 239 newborn infants were radioimmunoassayed five years later for plasma thyrotrophin concentrations. In 32 cases (0.03%) these were found to be raised. Thirty one of these children were traced and subjected to follow up examination by a paediatrician and a psychologist. Of the 31 children, 15 were found to have been receiving treatment for congenital hypothyroidism since a median of 5 months of age (diagnosed group). Of the 16 others, seven children were found to have raised serum thyrotrophin concentrations and were classified as hypothyroid (undiagnosed group). The remaining nine children were euthyroid. Children in the diagnosed group had a mean Griffiths developmental quotient of 87 (control value 103; p less than 0.01), and five out of 13 showed impaired neurological development. Of the remainder, those in the undiagnosed group had a mean developmental quotient of 100 and those in the euthyroid group a mean developmental quotient of 107. In this study achieving a detection rate of congenital hypothyroidism of one in 3000 in a neonatal screening programme resulted in overdiagnosis of about a quarter of patients considered to have true positive findings. This was outweighed, however, by the early identification of all infants with the disease.