Background Treating tuberculosis (TB) with short-course chemotherapy is recommended by the World Health Organization tuberculosis program and is one of the five packages of directly observed treatment short-course (DOTS) strategy. Objectives To investigate the effectiveness of short-course chemotherapy for treating new and retreatment TB patients and to assess the impact of drug resistance on its outcomes. Search strategy and selection criteria Published studies from the electronic databases such as Cochrane Library, DARE, Medline, Embase, Current Contents, CINAHL, Expanded Academic Index, PsycInfo, AustHealth and unpublished studies from Dissertation Abstract International, Index to These, Cochrane Reviewer Handbook were searched between 1993-2002. Studies addressing the treatment of patients with short-course regimen and the association between drug resistance and treatment outcomes were included. Two reviewers independently assessed study quality and extracted data. Results and discussion Eleven cohort studies met the inclusion criteria. Treatment with 6-month daily regimens was effective for new TB patients with success rate (percentage of cure cases and treatment completed cases) of over 80%. Implementation of 100% directly observed treatment was necessary to achieve over 80% treatment success in new cases treated with the intermittent regimen. The intermittent regimens did not work effectively for retreatment cases since success rate was only 68.5% even in the setting using 100% directly observed treatment in Peru. Short-course regimens were effective against drug-sensitive TB with the success rate of 87%; ineffective for patients with multi-drug resistance (rate of treatment failure in new cases ranged 22.3%-35.5% and rate of treatment failure in retreatment cases ranged 16.3-37.1%). Drug resistance had a negative impact on the outcome of short-course therapy and multi-drug resistance had a huge negative impact on the effectiveness of chemotherapy for TB. Conclusion The evidence suggests that the World Health Organization-targeted cure rate of 85% in new smear-positive TB cases is not achievable using the intermittent regimen and the target rate should be adjusted. DOTS-plus multi-drug resistant TB program of the world Health Organization should be adopted. However, the evidence is limited and the level of evidence was III-2.