Killen 2006.

Methods	Open‐label cessation and double‐blind extended treatment phase Setting: community cessation clinic, USA Recruitment: community volunteers Study period: December 2001 to March 2004
Participants	N = 362 54% males; average age ˜ 45; average number of cigarettes/d: ˜ 20 Inclusion criteria: smokers; 18 to 65 years of age; smoking at least 10 cigarettes/d or 3.5 packs/week Exclusion criteria: pregnancy, current lactation, epilepsy, bipolar disorder, schizophrenia, receiving active treatment for or reporting current depression or substance abuse, current use of bupropion or NRT, use of medication that could interact with bupropion or NRT
Interventions	Open‐label phase: All participants received bupropion SR 300 mg/d (Zyban) for 11 weeks and nicotine patch therapy for 10 weeks. Double‐blind phase: bupropion SR (150 mg/d) (n = 181) vs placebo (n = 181) for 14 weeks; at week 12, those assigned to placebo had their bupropion SR dose tapered to 150 mg/d for 2 weeks, then were switched to placebo in week 14
Outcomes	Primary outcome: point prevalence abstinence rates at 25‐week and 52‐week follow‐up Secondary outcomes: repeated point prevalence abstinence; continuous abstinence; craving and withdrawal symptoms; physiological measurements; adverse events; medication compliance
Funding source	Support was provided solely by National Cancer Institute Grant CA 090300 awarded to Joel D. Killen. Nicotine patches and bupropion were kindly provided by GlaxoSmithKline.
Declaration of interest	GlaxoSmithKline did not otherwise participate in the design, conduct, analysis, or reporting of this study.
Notes	Analysis of pharmacogenetics was reported in Sarginson 2011.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomisation was conducted before study entry by the method of permuted block (block size = 2 to obtain balance between groups) and was stratified on gender in the order of participant ID numbers.
Allocation concealment (selection bias)	Low risk	When a participant was assigned to the next available ID number in the corresponding gender, he or she was associated with that treatment group.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Both participants and researchers were blinded to treatment at extended treatment phases.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Both participants and researchers were blinded to treatment at extended treatment phases.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Lost at 12 months' follow‐up: 8% bupropion; 13% placebo
Selective reporting (reporting bias)	Low risk	Study protocol was not available, but detailed reporting of outcomes does not suggest selective reporting.
Other bias	Unclear risk	Of 304 trial participants, 270 provided samples for DNA extraction, but the selection process for those 270 is unclear.