ABSTRACT Objective Maternal hypothyroidism during pregnancy may lead to adverse outcomes. Recently published guidelines by the American Thyroid Association (ATA) do not advocate for universal screening but recommend a case-finding approach in high-risk pregnant women. The present study aims to evaluate the accuracy of this approach in identifying women with thyroid dysfunction during early pregnancy. Subjects and methods This is a multiple-center, cross-sectional study. Three hundred and one pregnant women were enrolled. Anamnesis and a physical examination were performed to detect which women fulfilled the criteria to undergo laboratory screening of thyroid dysfunction, according to the ATA's 2017 guidelines. The Zulewski's validated clinical score was applied to assess signs and symptoms of hypothyroidism. Serum levels of thyrotropin (TSH), free thyroxine (FT4), anti-thyroperoxidase (TPO-Ab), and anti-thyroglobulin (Tg-Ab) antibodies were determined. Results Two hundred and thirty one women (78%) were classified as high risk, and 65 (22%) were classified as low risk for thyroid dysfunction. Regarding the clinical score, 75 patients (31.2%) presented mild symptoms that were compatible with SCH, of which 22 (7.4%) had symptoms as the only risk factor for thyroid disease. 17 patients (5.7%) had SCH, of which 10 (58.8%) belonged to the high-risk group, and 7 (41.2%) belonged to the low-risk group. OH was found in 4 patients (1.4%): 3 (75%) in the high-risk group and 1 (25%) in the low-risk group. Conclusions The ATA's proposed screening criteria were not accurate in the diagnosis of thyroid dysfunction in pregnancy. Testing only the high-risk pregnant women would miss approximately 40% of all hypothyroid patients.