The aim of the study is to evaluate the accuracy of triple diagnostic test on thyroid nodules. The data from patients’ medical records who came to Cipto Mangunkusumo General Hospital for the first time or for evaluation of thyroid nodule and patients who underwent thyroidectomy during 2010 to 2011. Clinical examination was scored by McGill Thyroid Nodule Score. ROC procedure was performed to obtain clinical cut-off scores of diagnosis of malignant. Ultrasonography (USG) result was considered malignant for TIRADS 4, 5, and 6. If clinical, USG and histopathology examinations of triple diagnostic give positive results, it will be classified as concordant malignant whereas if all those three show benign results, the classification is benign. Thyroid carcinoma was found in 134 out of 161 patients with thyroid nodule. There were 84 patients with concordant results for all three elements of the triple test. Out of 84 patients with concordant triple diagnostic results, there were 53 malignant cases (32.9%) and 31 benign cases (19.3%). Main histopathological findings among patients with thyroid carcinoma was papillary (90.3%), follicular (3%), medullary (0.7%), and anaplastic (6%). The sensitivity and specificity of triple diagnostic was 77% and 94%, with positive predictive value of 98%, negative predictive value of 51,6% and accuracy of 80.9%. Combination of clinical findings, USG, and FNAB gave malignant probability of 92%, better than combination of clinical findings and USG (81.6%) or clinical findings and FNAB (87%). Triple diagnostic cannot be used as an ideal test to replace frozen section examination in managing thyroid nodule. However, in cases with concordant results of each triple diagnostic’s element, the positive predictive value (98%) and malignant probability (92%) is high.