Wright 2004 SWZ.

Methods	Generation of allocation sequence: unclear; stratified into adults and children; then, within each group, randomized by type of TB (sputum positive, sputum negative, extrapulmonary, relapse). Allocation concealment: unclear; sealed, sequentially numbered envelopes not stated if opaque. Blinding: assessors of sputum results blinded. Completeness of follow‐up: 98%.
Participants	Number: 1353 randomized; 55% male; most 15+ years. Included: adults and children with smear positive or negative, extrapulmonary TB, or relapse of previously treated TB. Excluded: died before discharge; or too ill to receive outpatient treatment; lived in area without treatment supporter; or referred in after treatment commenced.
Interventions	DOT by community health worker: participants visited for observation daily; community health worker trained to provide daily treatment supervision, record adherence on Treatment Support Card, remind participants who did not report for treatment, and notify diagnostic centre about those who defaulted treatment. DOT by family member: family member or carer chosen by participant trained to provide daily treatment supervision, record adherence on Treatment Support Card, and remind participants who did not report for treatment; participants also required to visit the community health worker weekly to check side effects and adherence and receive health education; defaulters reported to the diagnostic centre.
Outcomes	Cure or treatment completion: cure defined as smear negative at 6 months and on at least 1 previous occasion; treatment completion defined as treatment completed but smear results not available on at least 2 occasions before treatment completion. Death. Treatment failure: remained or became smear positive at ≥ 5 months. Default: failed to collect medication for > 2 consecutive months. Transferred out: formally transferred to another centre.
Notes	Location: Swaziland. Date: 2000 to 2002.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Insufficient information.
Allocation concealment (selection bias)	Unclear risk	Use of sealed envelopes not clear whether opaque. " sealed, sequentially numbered, stratum specific envelopes containing treatment assignments".
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Laboratory assessors were blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Non differential loss to follow‐up (4/664 and 5/662).
Selective reporting (reporting bias)	Unclear risk	All outcomes stated in the methodology are reported.
Other bias	Unclear risk	Not applicable.