The purpose of this study was to determine the health outcome benefits and cost-effectiveness of molecular testing in nodules with AUS/FLUS or FN/SFN cytology. The initial diagnosis and treatment of a hypothetical cohort of adult U.S. patients with solitary thyroid nodules ≥1 cm was simulated by decision analytic modelling using Medicare cost estimates for three management strategies: standard of care without molecular testing (StC), gene expression classifier (GEC), and mutation and miRNA testing (MMT). Test specificity had to be >68% for molecular testing to be cost-effective and decrease by >50% the rate of unnecessary surgeries performed on benign nodules. GEC decreased the rate of unnecessary surgeries by 32% relative to StC, yielding incremental costs of $1008 per patient or $5070 per unnecessary surgery avoided. MMT decreased the surgery rate by 67%, yielding incremental savings of -$1384 per patient or -$3170 per unnecessary surgery avoided. Results remained robust in deterministic sensitivity analyses; MMT was dominant for every variable tested. Independent of cancer prevalence, MMT yielded 52% fewer unnecessary surgeries relative to GEC and 70% fewer two-stage thyroidectomies, and correctly identified 70% more benign nodules.