Context The number of tuberculosis (TB) patients and multidrug-resistant species of Mycobacterium tuberculosis are increasing worldwide. Effective treatment is still the key factor to control TB in the society. Aims The aim of the study was to compare the efficacy of fixed-dose combination (FDC) and separate drug (SD) in the treatment of TB. Settings and design The study was a randomized, clinical trial carried out on 331 smear-positive TB patients in the Golestan province, Iran. Patients and methods Diagnosis and monitoring method was direct observation of sputum smears under a light microscope, which was used at the end of the second, third, fourth, fifth, or sixth months during the treatment. Statistical analysis used Sputum smear conversion rate at the end of second months and therapy outcome at the end of the treatment period were compared in both groups of the study by using the Poisson regression model and calculation incidence rate ratio. Results Sputum smear conversion rate at the end of the second month of the treatment had no significant difference between two groups. Success rate at the end of the treatment period was 89.36% in the intervention group and 95.65% in the nonintervention group which had significant difference (P=0.03). Adverse event incidence rate was 5.44% in all studied cases during the period of treatment. It had significant difference between the group who received FDC (8.93%) and the other who received SD (1.86%). Conclusion SD had better efficacy than FDC drugs in the treatment of TB patients.