Table 1. General information for included studies for Community Based intervention and Treatment (17studies).

Studies	Design	Region	Population (Age, type of TB)	Type and period of DOT and sampling size	Additional measures to promote treatment	Outcome	Main results
Wright, 2004[4]	RCT	Swaziland	Adult and children patients, new SS+ PTB	DOT by CHWs; Entire standard course of treatment; 290 cases with DOTS by CHWs /296 cases with DOT by Family member	Both CHWs and family members reminded them if they forgot and recorded adherence on a Treatment Supporter Card. Defaulters were notified to the diagnostic centre.	Completed treatment, cured treatment, death, failure, default, transferred out	The overall cure rate was 65% among new SS+ PTB cases; this increased to 70% when cure and treatment completed rates were combined. There was a non-significant difference in cure rate between the two groups [7% difference (0–15%), exact P = 0.10]. The difference between groups when cure and treatment-completed rates were combined approached significance [7% difference (0–15%), exact P = 0.06]. Death: 11% CHWs DOT vs.18% family member DOT Failure: 0% CHWs DOT vs.0% family member DOT; Default: 12% CHWs DOT vs.13% family member DOT; Transfer out: 3% CHWs DOT vs.3% family member DOT
Clarke, 2005[15]	Cluster RCT	South Africa	≥15 years patients, new SS+ PTB	DOT by LHWs (trained farmers or LHWs played a mentoring role; Entire standard course of treatment;75 cases in intervention group /89 cases in control	Visiting the patient and encouraging and monitoring treatment adherence regularly for self-supervision	Successful treatment failure, interrupted treatment, transfer out, death	The successful treatment completion rate in NSP adult TB patients was 18.7% higher [95%CI (0.9, 36.4)] on farms in the intervention group than on farms in the control group. There were no significant differences on the rest treatment outcomes.Failure: 7% LHW DOT vs.3% Clinic DOTInterrupted treatment: 8% LHW DOTS vs.16% Clinic DOT; Transfer out: 3% LHW DOT vs.3% Clinic DOT;Death:1% LHW DOT vs.3% Clinic DOT
Wandwalo, 2004[16]	Un-blinded RCT	Tanzania	≥5 years patients, new SS+ PTB	CB-DOT by guardians and former TB patients; 2 months of intensive treatment phase; 143 cases with CB-DOT /179 cases with health facility-based DOT	Recording drug taking, encouraging patients during treatment.	Successful treatment, cured treatment, Completed treatment, death, failure, default, transferred out	Treatment success rates were 85% and 83% among patients under CB-DOT and health facility-based DOT, respectively [OR (95% CI): 1.17(0.75, 1.83)]. There were no significant differences on the rest treatment outcomes. The cure rate: 78% CB-DOT vs.79% health facility-based DOT. Completed treatment: 9.8% CB-DOT/3.4% health facility DOT. Death: 7.0% CB-DOT vs.11.2% health facility DOT. Failure: 0% CB-DOT vs. 1.1% health facility DOT Default: .2.8% CB-DOT vs.3.4% health facility DOT Transferred out: 2.8% CB-DOT vs. 2.1% health facility DOT
Lwilla, 2003[17]	Cluster RCT	Tanzania	Patients with mean age(years) (SD): IBDOT 35.7(14.1) and CBDOT 35.5(13.3), SS+ PTB	CB-DOT by a community member, living in the same village as that of the patient; 2 months of intensive treatment phase; 221 cases with CB-DOT/301 cases with IB DOT	checking treatment card and counting pills during house visit; a sputum check follow-up	Cured treatment, transferred out, death, failure,	There was no significant difference in cure rates between the two strategies [M-H pooled OR (95%CI):1.58(0.32, 7.88)]. Fewer patients randomized to CBDOT died than under IBDOT (OR (95%CI):0.13(0.03, 0.65) at 2 months and 0.12(0.02, 0.89) at7 months]. Transfer out: 14.75% IBDOT vs. 4.98% CBDOT [OR (95%CI):0.29(0.07,1.28)]. Failure: 1.33% IBDOT vs. 5.29% CBDOT [OR(95%CI):0.63(0.13, 3.15)]
Kamolratanakul,1999[18]	RCT	Thailand	≥15 years old, SS+ PTB	CB DOT by village health volunteers or other community leaders; Entire standard course of treatment; 24 cases with DOTS by HW/ 34 DOT by community members /352 cases with DOTS by family members	Treatment monitoring card, home visit.	Successful treatment, completed treatment, default, failure, transfer out and death	No significant differences in outcomes could be observed between patient groups receiving DOT under the various options for treatment supervisors. Cure:79% Health workers vs.74% community members vs.77% family members Complete: 88%Health workers vs.79% community members VS. 84%family members. Death: 8%Health workers vs.3%community members/6% family members Failure: 0% Health workers vs.0% community members/2% family members Default:4% Health workers vs.15% community members /6% family members Transfer out: 0% Health workers vs.3% community members /2% family members
Zvavamwe, 2009[19]	Cohort study	Namibia	no description about ages of patients, SS+ PTB	Community-based TB treatment by family members of TB patients and other selected members of the community; Entire standard course of treatment; 308 cases with CB-DOT /24 cases with clinic-based DOT or self-administrated treatment	encouraging observing, documenting daily treatment and providing information about TB to the communities	Cured treatment and death	However, there was a statistically significant difference in cure rates between these two groups (x2 = 11.78, p≤0.05, and RR = 1.35, p = 0.05). There was no difference in death rate between community-based and the clinic/self-administered TB treatment groups (x2 = 3.01, p>0.05)
Cavalcante,2007 [26]	Longitudinal cohort study	Brazil	Adult patients, new SS+ PTB	CB DOT by CHWs who underwent a training program designed to teach them how to provide TB care in the community, focusing on TB control and DOT administration. Entire standard course of treatment; 489 cases with DOTS by CHWs /726 cases with clinic DOT/ 596 cases with SAT	Encouraging non-adherent patients to continue treatment, monthly clinical evaluation and microscopy	Successful treatment	Treatment success rates for new smear-positive and retreatment TB cases were significantly higher among those treated with CB-DOT compared to clinic-based DOT. 81.4% SAT vs. 84.6% Clinic-based DOT vs. 90.1% CB-DOT
Adatu, 2003[27]	Pre-post cohort study	Uganda	Patients with median age (years): pre-CB-DOTS 32 and CB-DOTS 30, new SS+ PTB	CB-DOT by community volunteers (always neighbors, rather than family members); 2 months of intensive phase); 540 cases in pre-CB-DOTS groups/450 cases in CB-DOTS groups	Administering drugs, noting the treatment cards, monitoring adverse reactions and reminding patients to return to health unit for sputum examination	Successful treatment, cured treatment, completed treatment, death, Failure, transfer out, interrupted treatment	Treatment success among new SS+ PTB cases increased from 56% to 74% [RR (95% CI):1.3(1.2, 1.5)] and interrupted treatment decreased from 23% to 1% [RR (95% CI):16.5 (6.1, 44.7)]. There were no significant differences for the rest outcome: Death: 15% pre-CB-DOT vs. 14% post CB-DOT; Cured: 45.3% pre-CB-DOT vs.62.2% post CB-DOT; Completed: 10.9% pre-CB-DOT vs.11.6% post CB-DOT; Failure: 0.9% pre-CB-DOT vs.0% post CB-DOT Transfer out: 5.2% pre-CB-DOT vs.11.2% post CB-DOT; Interrupted: 22.5% pre-CB-DOT vs.1.4% post CB-DOT
Newell, 2006[31]	Cluster RCT	Nepal	≥15 years patients, new SS+ PTB	CB-DOT by a female community health volunteer or a village health worker. Family-member DOT was defined as a strategy with drug taking supervised daily by a household member selected by the patient, with drugs provided to the patient’s supervisor every week. Entire standard course of treatment; 549 cases with CB-DOT /358 cases with family-based DOT	The supervisor also traced patients who discontinued treatment	Successful treatment	Community DOT and family-member DOT achieved success rates of 85% and 89%, respectively, but no significant difference [OR (95% CI): 0.67 (0.41, 1.10)].
Sinanovic, 2006[39]	Cohort study	South Africa	no description about ages of patients, new PTB (no description about smear positive or negative)	CB DOT by community health workers (known as ‘treatment supporters’) in the community. No mention period of DOT implementation; 305 cases with public clinics based DOT/ 518 cases with workplace-based DOT / 445 cases with CB-DOTS by CHWs	No mention of additional measures.	Successful treatment, cured treatment, interrupted treatment, failure, transferred out and death	Patients supervised in public clinics generally had lower treatment completion rates than those supervised in the occupational health clinics in the workplace and in the community. Successful treatment rate: 67.2% public clinics based DOT/ 87.1% workplace-based DOT / 72.8% CB-DOT; Cure rate: 63.9% public clinics based DOT vs.77.2% workplace-based DOT vs. 61.1% CB-DOT; Failed: 0.33% public clinics based DOT/0.97% workplace-based DOT vs. 0.90% CB- DOT; Died: 1.64% public clinics based DOT/11.20% workplace-based DOT vs. 3.82% CB- DOT; Interrupted: 14.10% public clinics based DOT/0.00% workplace-based DOT vs. 13.48% CB-DOT; Transfer out: 16.72% public clinics based DOT/0.97% workplace-based DOT vs. 8.99% CB- DOT
Sinanovic, 2003[40]	Cohort study	South Africa	No description about ages of patients, new and retreatment SS+ PTB	CB DOT lay-person ‘treatment supporter’; Entire standard course of treatment; NSP: 261 cases with CB-DOT /248 cases with clinic-based DOT/ 16 cases with Workplace-based supervision; Retreatment: 86 cases with CB-DOTS by lay-person ‘treatment supporter’/150 cases with clinic-based DOT/ 4 cases with Workplace-based DOT	No mention of additional measures.	Successful treatment	Community based care had higher successful treatment rate for both new and retreatment TB patients: NSP: 54% clinic-based DOT vs. 80% Community supervision by lay-person ‘treatment supporter’/81% Workplace-based supervision; Retreatment: 49% clinic-based DOT vs. 73% Community supervision by lay-person ‘treatment supporter’/ 75%Workplace-based supervision
Dudley, 2003[41]	Proohort study	South Africa	≥15 years patients, new and retreatment SS+ PTB	CB DOT by treatment supporters. Entire standard course of treatment. NSP: 369 cases with CB-DOTS /360 cases with Clinic based DOT. Retreatment: 123 cases with CB-DOTS /203 cases with Clinic based DOT	To note patients’ adherence card and visited patients who did not attend within a week; To recall the patients of default from treatment in 24 hours	Cured treatment, completed treatment, death, failure, transfer out, interrupted treatment	Higher cure rate for the patients with CB-DOTS, treatment success rates were similar. NSP: Cure rate: 72% CB-DOT vs. 46% Clinic based DOT; Complete: 9% CB-DOT vs. 7% Clinic based DOT; Failure:0.3% CB-DOT vs. 0% Clinic based DOT; Died: 1% CB-DOT vs. 3% Clinic based DOT; Interrupted:13% CB-DOT vs. 25% Clinic based DOT;Transfer:5% CB-DOT vs.19% Clinic based DOT; Retreatment: Cure rate: 63% CB-DOT vs. 35% Clinic based DOT; Completed:11% CB-DOT vs.12% Clinic based DOT; Failure:1% CB-DOT vs. 1% Clinic based DOT; Died:2% CB-DOT vs. 10% Clinic based DOT; Interrupted:18% CB-DOT vs. 30% Clinic based DOT; Transfer out:5% CB-DOT vs. 13% Clinic based DOT
Zwarenstein, 2000[42]	RCT	South Africa	≥15 years, new and retreatment PTB	DOT by LHW (took their drugs several times per week at their LHW’s home and under the LHW’s direct supervision); 5 days per week for the first 8 weeks for new patients,12 weeks for re-treatment patients, followed by 3 days per week for the continuation phase; 58 cases with DOT by clinic HW / 44 Self /54 cases with DOT by LHW	To note the adherence card in each visit to the patients, to visited the patients who failed to attend to treatment	Successful treatment, Completed treatment, cured treatment, death, failure, transfer out, interrupted treatment	Successful treatment rates: There were no statistically significant differences across the three supervision options (P = 0.136), but new patients benefit from LHW supervision (compared with clinic nurse and self-supervision, risk difference (95%CI): 24.2% (6, 42.5) and 39.1% (17.8, 60.3) respectively] as do female patients [compared with clinic nurse and self-supervision, risk difference (95%CI): (48.3%(22.8, 73.8) and 32.6% (6.4, 58.7) respectively]; Cured: 41% clinic DOT vs.57% LHW vs.41% Self; Completed: 16% clinic DOT vs.17% LHW vs.18% Self; Failure: 2% clinic DOT vs.6% LHW vs.5% Self; Interrupted: 26% clinic DOT vs.15% LHW vs.25% Self; Transferred out: 16% clinic DOT vs.2% LHW vs.9% Self; Death: 0% clinic DOT vs.4% LHW vs.2% Self.
Singh,2004[43]	Cohort study	India	no description about ages of patients, new SS+ PTB	DOT by CV. Entire standard course of treatment;141 cases with DOTS by CVs /476 cases with DOTS by GHWs	To using treatment cards with treatment details during the program, to visit the CVs or patients at least once every 2–3 days when CVs is deviated from standard practice	Successful treatment, cured treatment, default, death, failure, transfer out	There were no significant differences in all outcomes: Cure rate: 70% CVs vs. 75% GHWs [RR (95% CI): 0.95 (0.8, 1.2)]. Successful treatment: 78% CVs vs. 77% GHWs [RR (95% CI): 1.0 (0.8, 1.3)]. Default:15% CVs vs.15% GHWs [RR 1.03(0.7–1.5)]; Death:5% CVs vs.4% GHWs [RR 1.18(0.6–2.3)]; Failure: 2% CVs vs. 4% GHWs [RR 0.57(0.2–1.7)]; Transfer out: 0% CVs vs. 1% g GHWs
Pungrassami,2002[44]	Cohort study	Thailand	Adult and children patients, new and retreatment SS+ PTB	CB-DOT by village health volunteers, community Leaders or friends;.Entire standard course of treatment 21 CB-DOT/177 HP DOT / 181 FM/32 SA DOT	No mention of additional measures.	Successful treatment, cured treatment, completed treatment, Death, Failure, Default	There were no significant differences in treatment success between different types of main observers. AOR (95% CI) of treatment non-success were 1.1 (0.3, 4.7). 0.7 (0.2, 3.3), and 0.5(0.2, 1.1) for HP, CM, FM, and SA groups, respectively.
Walley, 2001[45]	RCT	Pakistan	≥15 years patients, new SS+ PTB	CB-DOT by a CHW at or near the patient’s home. 2 month of intensive phase; 66 cases with DOT by Health facility staff /104 cases with DOT by CHWs	To record drug-taking form and encouraged patients to complete treatment	Completed treatment, cured treatment, death, failure, transfer out, default	Cure rates: it is higher among TB patients supervised by CHWs (67%) than patients supervised by health facility staff (58%); Completed: 58%Health-facility staff vs. 67% CHWs; Death: 3%Health-facility staff vs. 4% CHWs; Failure: 0%Health-facility staff vs. 1% CHWs; Default: 30%Health-facility staff vs. 25% CHWs; Transfer out: 6%Health-facility staff vs. 0% CHWs
Becx-Bleumink, 2001[46]	Cohort study	Republic of Indonesia	no description about ages of patients, SS+ /SS- PTB	CB-DOT by sub-center health workers and village midwives. Entire standard course of treatment; 951 cases with DOT by CBTP /1402 cases with DOT by Non-CBTP	To select tuberculosis suspects, deliver treatment	Successful treatment rate	Treatment success rate (cure and treatment completion): CBTP villages: 90.4%, 89.5% and 93.7% in 1996, 1997, and 1998; Non-CBTP villages: 85.4%, 86.8% and 85.9% in 1996, 1997, and 1998

Notes

DOT refers to direct observation of treatment

NPTB refers to new cases of new cases of smear positive or culture-positive pulmonary tuberculosis

CB refers to community-based

HW refers to health worker

CHW refers to community health workers

LHW refers to lay health workers

NSP refers to new smear positive

NSN refers to new smear negative

NSP refers to new smear positive

CV refers to community volunteers

GHW refers to government health workers

TM refers to traditional hospital-based model of care

CBTP refers to Community based tuberculosis program