Mertens 2014.

Methods	Parallel RCT: brief motivational intervention plus referral list versus usual care plus referral list. ITT: no.
Participants	Setting: South Africa, primary health care clinic. Participants: patients aged 18 to 24, who screened positive to heavy alcohol or illicit and non medical drug use with a single question: "In the past year how many times have you had three or more drinks on one occasion" (for women) or "In the past year, how many times have you had five or more drinks on one occasion" (for men); excluded if too ill to participate or no phone. Number randomised = 403; (in assessed population) mean age = 21 years; 49% black, 51% mixed race; 8% education grade six or less, 79% some high school, 13% completed high school, 74% unemployed.
Interventions	BMI group (N = 206) received brief motivational intervention for alcohol and drug misuse at the Cape Town Drug Counselling Center (CTDCC). The training manual was Rollnick’s Health Behavior Change: A Guide for Practitioners (Rollnick et al 1999). Average intervention length was 10 minutes plus referral resource list for drinking and drug use. Usual Care group (N = 197) received minimally enhanced usual care plus resource list.
Outcomes	ASSIST score, prevalence of at risk use of alcohol, cannabis, methamphetamine, sedatives, methaqualone and heavy drinking. Assessed at 3 months.
Funding source	Funded by the National Institute on Drug Abuse R21DA022557.
Declaration of interests	Conflict of interest statement: none declared.
Notes	Loss to follow‐up: BMI group: 16/206 (7.8%). Usual Care group: 24/197 (12.2%). Payment: voucher for cell phone time or a local supermarket, worth ZAR 50 (approximately USD 6).
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Method of sequence generation not described.
Allocation concealment (selection bias)	Unclear risk	Sealed envelope but not described as numbered or opaque (p. 431).
Blinding of treatment providers	High risk	Not blinded.
Blinding of participants	High risk	Not blinded.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	"Research interviewers conducting follow‐up were blinded to randomisation status" (p. 431).
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Loss to follow‐up reported but only reason given is "unable to be contacted".
Selective reporting (reporting bias)	Low risk	Outcomes specified in methods are reported.