To determine the incidence of transient congenital hypothyroidism due to TSH receptor-blocking antibodies, we screened dried blood specimens obtained from 788 neonates identified as having possible congenital hypothyroidism (from a total population of 1,614,166 babies) and 121 controls. A RRA was used. The potency of blood spot TSH binding inhibitory activity was compared with the severity of congenital hypothyroidism to assess the possible etiological relationship. Maternal serum was studied to confirm the presence of blocking antibodies by both RRA and bioassay. Blood spots obtained from 9 infants contained potent TSH receptor-blocking activity. Samples from 2 additional babies, studied because of clinical suspicion of the disease, were also positive. Long term outcome was known in 8 of the 11 babies, and all had transient disease. Neonates with TSH receptor-blocking activity greater than 132 U/L had a significantly lower T4 level (P < 0.05) and higher TSH (P < 0.005) than those in whom TSH binding-inhibitory activity was less than 132 U/L. All 9 mothers had autoimmune thyroid disease, and 3 had more than 1 affected child. Potent blocking activity was present in 7 maternal serum samples as long as 7 yr after the births of their affected babies. We conclude that measurement of TSH binding-inhibitory activity in dried neonatal blood specimens is a simple and effective method to predict the occurrence of transient congenital hypothyroidism. The incidence of this disorder in North America is 1 in 180,000 normal infants, or approximately 2% of babies with congenital hypothyroidism.