Rohsenow 2017.

Study characteristics
Methods	Country: USA  Setting: university offices in Rhode Island, USA Aim: to compare varenicline with transdermal nicotine (NRT) for smokers with current substance use disorders (SUD) for effects on smoking abstinence Study design: double‐blind double‐placebo‐controlled randomised design, stratifying by MDD history
Participants	137 adult smokers in substance‐use disorder treatment, who were substance abstinent < 12 months, and smoked 10+ CPD for the past 6 months. 47% female, mean age 39.6, baseline CPD 19.5 Participants did not need to motivated to quit Exclusion criteria: hallucinations/delusions, current SC treatment, contraindications for either medication, using medications affected by SC (antipsychotics, warfarin, theophylline and insulin), suicidal ideation, not willing to try to quit smoking and, substance use reported on the day of or before recruitment or positive breath alcohol at recruitment
Interventions	Varenicline: 12 weeks of 2 x 1 mg/day (titrated for first week) NRT: 21 mg/day, decreasing to 7 mg/day
Outcomes	Primary: 7‐day smoking PPA at 3 months Secondary: 7‐day PPA at 6 months; quantity and frequency of smoking and substance use at 3 and 6 months; within‐treatment abstinence; medication adherence; depressive symptoms
Notes	New for 2022 update  Funding by a personal grant 1R01DA024652 from the National Institute on Drug Abuse and Department of Veterans Affairs Author declaration of interests: "All authors report no financial interests or potential conflicts of interest except for the following: R.M.S. has received travel and honorarium from D&A Pharma and has received consultant fees from CT Laboratorie"
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"Urn randomization [32] to medication condition (1: 1 assignment) was stratified by gender, median score on the Fagerström Test for Nicotine Dependence [33] from a previous study [28] and history of MDD. The randomization program was run by the Project Manager, who concealed the assignment from all other staff."
Allocation concealment (selection bias)	Low risk	"Urn randomization [32] to medication condition (1: 1 assignment) was stratified by gender, median score on the Fagerström Test for Nicotine Dependence [33] from a previous study [28] and history of MDD. The randomization program was run by the Project Manager, who concealed the assignment from all other staff."
Blinding (performance bias and detection bias) All outcomes	Low risk	Double‐blind trial with placebo patch and placebo tablets used (double placebo). Abstinence biochemically validated
Incomplete outcome data (attrition bias) All outcomes	Low risk	23/60 and 25/77 did not complete follow‐ups for primary outcome at 3 months but information not stated for 6 months; however multiple imputation sensitivity analysis has been done (supplementary materials) which the study authors claim did not change the conclusion
Selective reporting (reporting bias)	Low risk	Prespecified outcomes from trial registry reported