Objective: To describe treatment outcomes (cure, noncompliance or death) after the implementation of the Directly Observed Treatment, Short-course (DOTS) strategy for tuberculosis control in the city of São José do Rio Preto, Brazil, between 1998 and 2003. Methods: A descriptive study, based on secondary data (National Case Registry database, Tuberculosis Epidemiology database, and the 'Black Book' Registry), was conducted using a specific instrument. The data were analyzed using descriptive statistics. Results: After the implementation of the DOTS strategy, there was a decrease in noncompliance and case detection rates as well as an increase in cure and death rates. The increase in the number of tuberculosis-related deaths might be attributable to three factors: the predominance of the disease in individuals over 50 years of age, tuberculosis/HIV co-infection, and the presence of accompanying diseases. Conclusion: The implementation of the DOTS strategy strengthened the decentralization of the tuberculosis control plans as well as the integration of the Basic Health Care Clinic teams with the Tuberculosis Control Program team. Political commitment of the administrator to tuberculosis control, in conjunction with the policy of benefits and incentives, is essential for the sustainability of the DOTS strategy.

Keywords: Tuberculosis; Epidemiology; Health services; Politics; Directly observed therapy.