Wandwalo 2004 TZA.

Methods	Generation of allocation sequence: coin tossing in each of 5 clinics. Allocation concealment: none. Blinding: none. Completeness of follow‐up: 100% (no losses).
Participants	Number: 587 randomized; 322 smear positive, 182 smear negative, and 83 extrapulmonary TB; 57% male. Included: people with TB (aged 5+); new smear positive, smear negative, and extrapulmonary cases; human immunodeficiency virus (HIV) status not known. Excluded: previously treated for TB; severe illness; transferred from another clinic; previously enrolled in the study.
Interventions	Community‐based DOT: daily treatment supervised at home by 'guardian' (usually a family member) during 2‐month intensive period; supervisors trained to observe drug taking, encourage participants to complete treatment, keep records, collect drugs, and assess drug side effects; during first 2 months participants received 'spot' visits by health workers who conducted treatment card checks and pill counts; during first 2 months participants also requested to attend clinic every 2 weeks for clinical review and progress monitoring. Health facility‐based DOT: daily supervision at clinic by health workers during the 2 month intensive period. Apart from the observation option participants received the same standardized management including drug therapy.
Outcomes	Treatment success: cure plus treatment completion. Cure: smear positive initially and negative at 7 or 8 months and on at least 1 previous occasion. Treatment completion: positive results initially, negative at 2 months and no results at end of treatment; or smear negative initially and received treatment on clinical grounds; or those who completed full course of treatment but had no initial or end‐of‐treatment results. Death: from all causes. Treatment failure: participants who remained or became smear positive or 5 months or later. Default: failed to collect medication for > 2 consecutive months. Transferred out: transferred to a clinic in another area.
Notes	Location: Dar es Salaam, Tanzania. Date: 2001 to 2003.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"Randomly by coin toss".
Allocation concealment (selection bias)	High risk	None.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	No details reported.
Incomplete outcome data (attrition bias) All outcomes	Low risk	No exclusions.
Selective reporting (reporting bias)	Low risk	All outcomes stated in the methodology are reported.
Other bias	Unclear risk	Not applicable.