Results at 30 months of a randomised trial of FDCs and separate drugs for the treatment of tuberculosis.

Study C was an open-label, non-inferiority, randomised controlled trial of fixed-dose combination (FDC) or separate drugs given during the intensive phase of treatment to 1585 patients with smear-positive pulmonary tuberculosis conducted at 11 sites in Africa, Asia and Latin America. Thirty months post-randomisation, the failure/relapse rates in the per protocol population were 7.4% of 591 patients on FDCs and 6.5% of 587 patients on separate drugs; the site-adjusted difference was 0.3% (90%CI -1.8 to 2.3). In the modified intention-to-treat analysis, the corresponding results were respectively 17.9% of 683 and 16.1% of 671; the site-adjusted difference was 2.0% (90%CI -1.2 to 5.2).