Penetar 2012.

Methods	Randomised, double‐blind, placebo‐controlled trial. Participants used cannabis as usual for 7 days prior to randomisation.
Participants	Setting: outpatient with daily clinic attendance Monday to Friday, Harvard Medical School, USA. Scheduled duration 21 days Participants: 22 adults, seeking treatment, cannabis abuse or dependence by DSM‐IV, ≥ 3 years of heavy use (smoking on ≥ 5 days a week or > 25 times per month) and with ≥ 2 negative symptoms in previous quit attempts Group sizes: group 1, 10; group 2: 12 Demographic data provided only for 9 who completed the study (5 male, mean age 31.2 years, 7 met criteria for dependence) Exclusion criterion: abuse or dependence on any other drug (including nicotine)
Interventions	Group 1: oral bupropion SR 150 mg/day for days 1–3, then 150 mg twice a day Group 2: placebo Riboflavin added to medication capsules to measure compliance Weekly individual MET (3 sessions) as adjunct intervention
Outcomes	Number completing study, change in withdrawal discomfort and change in craving Data reported as graphs and results of statistical tests Withdrawal by Marijuana Withdrawal Checklist (29 items each rated 0–3). Withdrawal discomfort score calculated from 10 items (maximum score 30) Drug use, sleep and withdrawal recorded by participants in daily diary. Urine testing to confirm cannabis use
Notes	Funding: research grant (NIDA). No conflicts of interest reported Disclosures of interests according to ICMJE criteria were a requirement of publication.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Random allocation to treatment group stated, but method of sequence generation not reported.
Allocation concealment (selection bias)	Unclear risk	Method of allocation concealment not reported.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Quote: "Bupropion tablets were repackaged into gelatin capsules … Placebo consisted of identical appearing gelatin capsules." Comment: double‐blind stated
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Double‐blind stated as indicated above. Use of urine screening to verify self‐report expected to reduce risk of bias.
Incomplete outcome data (attrition bias) All outcomes	High risk	High rate of dropout and demographics reported only for those who completed treatment. Unclear whether there were differences between the groups, or between those who did and did not complete the study. Unclear how missing data were handled.
Selective reporting (reporting bias)	Unclear risk	Data on adverse effects not reported.
Other bias	Low risk	None apparent