INT J TUBERC LUNG DIS 18(9):1092–1098 Q 2014 The Union http://dx.doi.org/10.5588/ijtld.13.0867 Effectiveness of directly observed treatment of tuberculosis: a systematic review of controlled studies J-H. Tian,* Z-X. Lu, † M. O. Bachmann, ‡ F-J. Song ‡ *Evidence-Based Medicine Centre, Lanzhou University, Lanzhou, Gansu, † Department of Social Medicine and Health Management, School of Public Health, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, Hubei, China; ‡ Norwich Medical School, University of East Anglia, Norwich, UK SUMMARY BACKGROUND: There is controversy about the effec- tiveness of directly observed treatment (DOT) for anti- tuberculosis treatment. This systematic review aimed to synthesise evidence from studies that compared DOT and self-administered treatment (SAT) or different types of DOT for anti-tuberculosis treatment. METHODS: Multiple databases were searched by two independent reviewers to identify relevant randomised (RCTs) and non-randomised studies. The risk of bias was independently assessed by two reviewers, and studies at high risk of bias were excluded. Data extraction was conducted by one reviewer and checked by a second reviewer. Primary outcome measures were cure and treatment success. RESULTS: We included eight RCTs and 15 non- randomised studies that were predominantly conducted in low- and middle-income countries. There was no convincing evidence that clinic DOT was more effective than SAT. Evidence from both RCTs and non-rando- mised studies suggested that community DOT was more effective than SAT. Community DOTwas as effective as, or more effective than, clinic DOT. There was no statistically significant difference in results between family and non-family community DOT. CONCLUSIONS: Community DOT by non-family members might be the best option if it is more convenient to patients and less costly to health services than clinic DOT. KEY WORDS: directly observed treatment; self-admin- istered treatment; tuberculosis; community-based DOT THE MAIN PILLAR of global tuberculosis (TB) control is the diagnosis and treatment of persons with active disease. To this end, the World Health Organization (WHO) recommended directly ob- served treatment (DOT) in 1993, and then the DOTS strategy in 1997. The DOTS strategy contains five major elements: political commitment, improved laboratory testing, adequate and free supply of anti- tuberculosis drugs, a reporting system to document progress and DOT. Considerable progress has been made towards global TB control since the recom- mendation of the DOTS strategy. 1 A complete course of anti-tuberculosis treatment usually lasts 6–9 months. Inadequate adherence to treatment is common, causing treatment failure, relapse, the development of drug resistance and transmission of infection. It has been hoped that treatment adherence could be improved if health workers or other individuals are present as observers when patients are taking their anti-tuberculosis medications. Although DOT was initially designed to be carried out by health workers in health facilities, community-based DOT has also been developed, particularly in low- and middle-income countries (LMICs). 2 A Cochrane systematic review of randomised controlled trials (RCTs), which concluded in 2007 that DOT was not more effective than self-adminis- tration of treatment (SAT) for TB patients, 3 triggered considerable debate among researchers and profes- sionals in terms of whether DOT is necessary, by whom, and in what settings. 4,5 The controversy regarding this issue is partly due to the scarce evidence from RCTs. We searched the WHO Inter- national Clinical Trials Registry (March 2013) and found no relevant ongoing studies. Given the current policies and ethical considerations, it is unlikely that new evidence from RCTs comparing DOT and SAT for anti-tuberculosis treatment will be available in the near future. On the other hand, there is abundant evidence from non-randomised observational studies. Although there are challenges in using non-rando- mised studies for evaluating health interventions, such evidence is relevant and should be appropriately assessed and considered. 6 We conducted a systematic review of both rando- Correspondence to: Fujian Song, Norwich Medical School, University of East Anglia, Norwich, NR4 7TJ, UK. Tel: ( þ 44) 16 0359 1253. e-mail: Fujian.song@uea.ac.uk Article submitted 2 December 2013. Final version accepted 16 May 2014.