Drug-resistant tuberculosis has been reported since the early days of the introduction of chemotherapy. However, most of the evidence was limited to developed countries. In 1992, the Third World Congress on Tuberculosis concluded that there was little recent information on the global magnitude of multidrug-resistant tuberculosis (MDR-TB), defined as resistance to at least isoniazid and rifampicin. Through the WHO/IUATLD Global Project on Drug-Resistance Surveillance launched in 1994, a large number of reliable and accurate data have allowed us to understand the magnitude of the problem of MDR-TB. The data available suggest that globally MDR-TB is not a problem (median = 1% in 64 countries/geographical sites surveyed) of the same magnitude as that of drug-susceptible tuberculosis. However, MDR-TB is at critical levels in specific regions of the world. Hot spots for MDR-TB include Estonia, Latvia, the Oblasts of Ivanovo and Tomsk in Russia, and the provinces of Henan and Zhejiang Provinces in China. Trends confirm that MDR-TB is limited to local epidemics but the evidence is not yet irrefutable, as many countries have only provided short-term data. Two-thirds of the world's countries and, more importantly, half of the 22 tuberculosis high-burden countries, have not yet provided data. Mathematical modelling suggests that 3.2% (or 273,000) of the world's estimated new tuberculosis cases (95% confidence intervals: 185,000 and 414,000) were MDR-TB in 2000. Adoption of DOTS to prevent the generation of resistant strains and careful introduction of second-line drugs to treat patients with MDR are the top priorities for proper control/containment of MDR-TB.