Thailand's implementation of the Directly Observed Treatment, Short course (DOTS) strategy to increase tuberculosis (TB) control program efficacy has not achieved the World Health Organization (WHO) TB key targets. We defined two TB control models in the study. Patients in Model 1 were treated with a conventional DOTS strategy and in Model 2, patients were treated the same as Model method 1 but were given a phone call reminder to take their medication. Multi-drug resistant tuberculosis (MDR-TB) and non-MDR-TB patients were randomized into either Model 1 or 2. Treatment outcomes were given as cure rates, completion rates, failure rates or success rates at 18 months in the MDR-TB group and 6 months in the non-MDR-TB group. The sputum conversion rate at 1 month were evaluated for both groups. In the MDR-TB group, the sputum conversion rate was 20% (95% CI 8-45) in Model 1 and 90% (95% CI 73-98) in Model 2 (p < 0.001). In the non-MDR-TB group, the sputum conversion rate was 52% (95% CI 36-70) in Model 1 and 37% (95% CI 22-56) in Model 2 although the difference was not significant (p = 0.221). The Model 2 success rates were significantly higher (73.7%, 96.7%) in both the MDR-TB and non-MDR-TB groups (p < 0.001, p = 0.047). The MDR-TB rate in northern Thailand decreased from 4.1% during April-September 2008 to 1.8% during April-September 2009. Further study of the association between implementation of Model 2 and MDR-TB incidence reduction needs to be carried out.