OBJECTIVE: To determine the rate of exposure of pregnant women to levothyroxine and to assess changes in these rates before, during and after pregnancy.

DESIGN: Register-based cohort study.

SETTING: Danish nationwide registers.

POPULATION: All women having a live birth in Denmark between 1 January 1997 and 31 December 2010 (n = 912 342).

METHODS: All pregnant women in the study period were identified from the Danish Medical Birth Register. Exposed women were identified from the Danish National Prescription Register, based on redemption of levothyroxine prescriptions before, during or after pregnancy.

MAIN OUTCOME MEASURES: The rate of pregnant women redeeming levothyroxine prescriptions and maternal characteristics.

RESULTS: We identified a fourfold increase in levothyroxine prescription redemption during the study period, from 0.34% in 1997 to 1.39% by 2010. A mean of 0.79% of our cohort received levothyroxine. Most of the women who were using levothyroxine before pregnancy continued the therapy during their pregnancy, but 9.4% stopped redeeming their prescriptions. Overall, 0.28% of our cohort received a levothyroxine prescription for the first time within 9 months after pregnancy.

CONCLUSIONS: Fewer women than expected received levothyroxine treatment during pregnancy even though a fourfold increase was observed during the study period. Furthermore, one of 10 discontinued treatments during pregnancy. These findings all indicate that too few women are treated for hypothyroidism during pregnancy. Further research is needed to determine whether hypothyroid pregnant women are suboptimally treated and the possible consequences for the mother and fetus.