SETTING: Primary health care centres in Callao, Peru. OBJECTIVES: To evaluate the effect of universal access to the microscopic-observation drug susceptibility (MODS) assay on treatment outcomes in new and primary multidrug-resistant tuberculosis (MDR-TB) patients and on the process of drug susceptibility testing (DST). DESIGN: Retrospective review of tuberculosis (TB) registers and clinical records before (2007) and after (2009) the introduction of MODS in 2008. RESULTS: There were 281 patients in each cohort. Favourable treatment outcomes for 2007 (81%) and 2009 (77%) cohorts were similar. There was an increase in loss to follow-up (from 6% to 10%, P = 0.04) and a reduction in failure rates (from 4% to 0.4%, P = 0.01) in the 2009 compared with the 2007 cohort. In new MDR-TB cases (n = 22), a favourable treatment outcome was improved (from 46% to 82%, P = 0.183) in the 2009 cohort. DST coverage improved (from 24% to 74%, P < 0.001), and a significant reduction in time to diagnosis of drug-susceptible (from 118 to 33 days, P < 0.001) and MDR-TB (from 158 to 52 days, P =30.003) was observed in the 2009 cohort. CONCLUSION: Universal access to MODS increased DST coverage, reduced the time required to obtain DST results and was associated with reduced failure rates. MODS can make an important contribution to TB management and control in Peru.