Peltzer 2013.

Study characteristics
Methods	Study design: cluster‐randomized controlled trial Country: South Africa Setting: primary care clinics Eligibility criteria: included were clinics with a high tuberculosis caseload in 3 study districts. Adult (18+ years) patients with active tuberculosis and retreatment patients, males and females who visited the primary health care facility and scored 8 (men) or 7 (women) on the AUDIT after screening were included in the study. Excluded were people less than 18 years of age. Duration of follow‐up: 6‐months Informed consent: informed consent was provided by all patients; consent of the clinics was not described Ethical approvals: Human Sciences Research Council Research Ethics Committee
Participants	Sample size: 40 clinics (20 intervention, 20 control); 1196 patients (741 intervention, 455 control) Description of the target population: patients in primary care with active tuberculosis Age: 36.7 years Sex: 25.7% female Race/Ethnicity: not reported Marital status: not reported Harmful alcohol use (baseline): average AUDIT = 15.7; 73.2% had AUDIT scores between 7‐19, 26.8% had AUDIT scores between 20‐40 Co‐occurring disorders: 54.2% HIV positive, All had active tuberculosis
Interventions	Type: non‐pharmacologic Description: Brief counseling included assessment, treatment as usual, and counseling to identify any alcohol‐related problems, psychoeducation, personalized feedback, information on the relationship between alcohol use and tuberculosis treatment, problem solving, keeping a drinking diary, identifying a helper, planning for follow‐up counseling, and other motivation‐behavioral skills Duration and frequency: Two 15‐30 minute sessions within 1 month of each other Delivery and provider: Trained lay HIV counselors and nurses Comparison group: Health education on responsible drinking, assessment, and treatment as usual for 1 session provided by counselors and nurses
Outcomes	Primary outcome(s): hazardous alcohol use Primary outcome measurement tool(s): AUDIT Secondary outcome(s): alcohol dependence, heavy episodic drinking Secondary outcome measurement tool(s): AUDIT Time points: 0‐, 3‐, and 6‐months
Notes	Study funding and conflicts of interest: Department of Health in South Africa (TNDOH: 21/2010‐2011) Linked study records: None
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: "Randomisation was conducted using a secure remote randomization service. Within each district in each of the three provinces the 14 public primary health clinics with the highest TB caseloads were randomly assigned to the treatment and control arms. A secure remote randomisation service carried out randomisation." Pg. 3
Allocation concealment (selection bias)	Low risk	Quote: "A secure remote randomisation service carried out randomisation. Twenty one allocations were initially generated for each of the possible factorial combinations of clinic type, TB case load, and intervention" Pg. 3
Blinding of participants and personnel (performance bias) All outcomes	High risk	"Participants (clinic staff members and TB patients) were not blinded to their intervention or control status" Pg. 4
Blinding of outcome assessment (detection bias) All outcomes	High risk	Outcomes reported by participant not blind to intervention Quote: "However, to protect against information biases in the reporting of alcohol use and TB treatment adherence behaviour, the data collection team who assessed the outcomes were blinded to the clinic’s status as intervention or control group." Pg. 4
Incomplete outcome data (attrition bias) All outcomes	High risk	Quote: "As illustrated in Figure 1, response rates were higher in the intervention than in the control group at both follow‐ups. At the 3‐month follow‐up, response rates for the control and intervention were 39% and 54%, respectively, and at 6 months, the control and intervention group response rates were 59% and 79%, respectively. In the control group 41% did not complete the last follow‐up survey (i.e., the dropout rate was 41%); in the intervention group, 21% did not complete the last follow‐up survey. The main difference between groups was in the dropout rate (percentage of participants that did not complete the last follow‐up survey): 21% in the intervention group versus 41% in the control group. Attrition analyses were conducted to check for differential attrition by examining the condition by dropout interactions at baseline. Dropout was significantly related to the condition (P < 0.001) and poverty (P = 0.003), indicating a relationship between these variables and dropout status depending on condition: 21% in the intervention group versus 41% in the control group dropped out and the poverty index was 9.9 in the intervention versus 8.9 in the control group. Dropout was not related to gender (P = 0.308), age (P = 0.078), education (P = 0.866), AUDIT score (P = 0.958), AUDIT (7–40) (P = 0.516), AUDIT (7–19) (P = 0.952), AUDIT (20–40) (P = 0.918), heavy episodic drinking (P = 0.192), daily or almost daily tobacco use (P = 0.929), HIV status (P = 0.488), and TB new or retreatment patient (P = 0.082)." Pg. 7
Selective reporting (reporting bias)	Unclear risk	Protocol not available