Hong 2015.

Study characteristics
Methods	Country: China Setting: hospital Study design: parallel RCT
Participants	300 COPD inpatients/perioperative patients, all smokers. 3% female, mean age 58
Interventions	5As (Ask, Advise, Assess, Assist, and Arrange) 5As + varenicline All participants received interactive behavioural support, either in person or on the telephone
Outcomes	Self‐reported CA at 24 weeks
Notes	Study funding: not reported Author declaration of interests: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Stated randomised but methods not reported
Allocation concealment (selection bias)	Unclear risk	Concealment not reported
Blinding (performance bias and detection bias) All outcomes	High risk	Not placebo‐controlled
Incomplete outcome data (attrition bias) All outcomes	Low risk	Study reported no participants lost to follow‐up
Selective reporting (reporting bias)	Unclear risk	Protocol unavailable