This multicenter prospective observational study evaluated the use of a mutation detection and miRNA expression algorithm to improve the diagnostic yield of molecular testing for thyroid nodules. 109 thyroid nodules with AUS/FLUS or FN/SFN cytology collected from 16 physicians at 12 centers across the United States were tested using a panel of 17 validated gene alterations. Nodules negative for gene mutations were then tested with a 10-miRNA gene expression classifier. Mutations were detected in 69% of nodules with malignant outcomes. Among mutation-negative specimens, miRNA testing correctly identified 64% of malignant cases and 98% of benign cases. The diagnostic sensitivity and specificity of the combined algorithm was 89% (95% confidence intervals (CI): 73%-97%) and 85% (95% CI: 75%-92%), respectively. Independently of variations in cancer prevalence, the test increased the yield of true benign results by 65% relative to mRNA-based gene expression classification, and decreased the rate of avoidable diagnostic surgeries by 69%. It was therefore concluded that this unique combined testing approach can accurately classify benign and malignant thyroid nodules, increase the diagnostic yield of molecular cytology, and further improve the preoperative risk-based management of benign nodules with AUS/FLUS or FN/SFN cytology.