OBJECTIVES: The aim of this study was to enhance tuberculosis (TB) treatment outcome monitoring by linking diverse surveillance systems and estimating treatment outcomes including relapse. METHODS: Tuberculosis treatment was surveyed in the Lazio region (Italy) from 1999 to 2001; a six-year follow-up of notified cases was undertaken to detect relapses. The results were analyzed as a population-based case-control study comparing each unsuccessful outcome and relapse with eligible controls. RESULTS: Of the 974 patients who entered the survey, 805 (82.6%) had complete treatment evaluations; 398 (49.4%) had a successful outcome, 401 (49.8%) had an unsuccessful outcome, and six developed chronic TB. Death was associated with age >64 years (OR 5.9; 95% CI 3.1-11.2), male gender (OR 2.1; 95% CI 1.0-4.4), and using second-line drugs (OR 2.3; 95% CI 1.0-5.4). Treatment failure was associated with previous treatment (OR 3.0; 95% CI 1.4-6.7) and being male, being foreign born (OR 6.6; 95% CI 2.1-21.2), receiving second-line drugs (OR 7.4; 95% CI 1.8-29.5), and receiving modified therapy (OR 5.1; 95% CI 1.7-14.9). Relapses after successful outcomes were detected in 5.5%, for which the strongest predictor was having extrapulmonary lesions (OR 22.8; 95% CI 1.8-287.3). CONCLUSIONS: Linking our survey data to other surveillance systems improved the mortality estimates and detected a high rate of relapse. Having received previous treatment and being a foreigner were independent determinants of treatment failure, suggesting that both acquired and primary drug resistance affect TB patients in Lazio.