Hilberink 2011.

Methods	Design Cluster randomised controlled trial Setting Outpatients
Participants	Inclusion criteria Practices Using 1 out of 4 suitable general practice electronic information systems Participants Age > 35 years; diagnosed as having COPD; recorded medication with ICPC code R95/96; prescription of at least 3 times of bronchodilators in the past year (ATC code R03a/bc); prescription of at least 2 times of inhaled anti‐inflammatory medication in the past year (ATC code R03). GPs were asked to confirm the diagnosis. Exclusion criteria None Number randomised: Practices Total 74, counselling and NRT (CN) 23, counselling, NRT, and bupropion (CNB) 26, usual care (UC) 25 Participants Total 697, CN 252, CNB 291, UC 154 Number followed up Not reported Age mean (SD): CN 58.0 (12.2), CNB 60.7 (11.2), UC 60.1 (11.5) Sex n (%) male: CN 11 (46.5), CNB 12 (47.8), UC 14 (55.4) Cigarettes/day mean (SD): CN 16.9 (10.3), CNB 16.9 (9.1), UC 16.8 (9.7) FTND mean (SD): CN 4.4 (2.4), CNB 4.6 (2.3), UC 4.3 (2.6) Severity COPD Not reported
Interventions	Behavioural treatment A. Counselling plus NRT General practice level General practices received support for implementing a smoking cessation program consisting of a counselling strategy plus the recommendation of NRT Participant level Extended version of the minimal intervention strategy (MIS), specifically aimed at education and support of patients by GP. Participants were invited for control visit in accordance with COPD treatment guidelines. The first control visit focused on symptoms, health status and treatment, smoking behaviour, and motivational stage to quit smoking. Participants were divided into 3 categories: preparers (willing to quit within 1 month): next consultation was scheduled to set a quit date and to plan the follow‐up visits to GP (maximum of 2 follow‐up visits and proactive telephone calls by the practice nurse (maximum of 3 calls)); contemplators (willing to quit within 6 months) received self efficacy enhancing information by discussing how to cope with the various barriers to quit. Depending on their severity of nicotine dependence, they received additional info about NRT. Contemplators were invited again 2 weeks later; pre‐contemplators (not willing to quit) received information about the advantages of quitting. The participant education tools consisted of a booklet especially developed for the COPD population and a videotape. B. Counselling plus NRT plus bupropion General practice level General practices received support for implementing a smoking cessation program consisting of a counselling strategy plus the recommendation of NRT and bupropion Participant level Extended version of the MIS, specifically aimed at education and support of patients by GP. Participants were invited for control visit in accordance with COPD treatment guidelines. The first control visit focused on symptoms, health status and treatment, smoking behaviour, and motivational stage to quit smoking. Participants were divided into 3 categories: preparers (willing to quit within 1 month): next consultation was scheduled to set a quit date and to plan the follow‐up visits to GP (maximum of 2 follow‐up visits and proactive telephone calls by the practice nurse (maximum of 3 calls)); contemplators (willing to quit within 6 months) received self efficacy enhancing information by discussing how to cope with the various barriers to quit. Depending on their severity of nicotine dependence, they received additional info about NRT. Contemplators were invited again 2 weeks later; pre‐contemplators (not willing to quit) received information about the advantages of quitting. The participant education tools consisted of a booklet especially developed for the COPD population and a videotape. C. Usual care Consisted of regular check‐ups and COPD information. Pharmacological treatment A. Counselling plus NRT Information about and advice on using NRT (preparers and contemplators) B. Counselling plus NRT plus bupropion Information about and advice on using NRT, plus information/advice on using bupropion (preparers and contemplators) C. Usual care None
Outcomes	Abstinence Point prevalence (did not smoke in the last 7 days) at 6 and 12 months Validation Urine cotinine ≤ 50 ng/mL (measured 12 months after baseline)
Notes	Funding The present study was financed by the Asthma Foundation Netherlands, The Netherlands Organisation for Health Research and Development (ZonMw), Pharmacia, and GlaxoSmithKline. Conflicts of interest None Correspondence Additional information received after correspondence
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"75 practices were allocated to one of the three arms by way of a stratified randomisation procedure, taking into account experience with the Minimal Intervention Strategy, delegation of tasks to the practice assistant and practice size." It was a computer‐generated randomisation. (Additional information from author)
Allocation concealment (selection bias)	Unclear risk	Allocation was not concealed for the GPs because they had to be trained. Participants were not told to which group they were randomised, but it is presumable that participants saw through if they got the intervention or not. (Additional information from author)
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not possible to blind for behavioural treatment
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Self report of abstinence was blinded for the GPs, not for the researchers. Cotinine validation was blinded for the participants and GPs, not for the researchers. (Additional information from author)
Incomplete outcome data (attrition bias) All outcomes	Low risk	Number of missing values is low, and there was no differential attrition
Selective reporting (reporting bias)	High risk	All prespecified outcomes reported, however in one of the articles the authors did not mention that it was a 3‐armed RCT: "the study is a two armed randomised controlled trial". One arm was entirely ignored
Other bias	High risk	A. GPs' adherence to the protocol was poor: "The majority of the GPs (70–83%) complied with six out of the nine aspects of the protocol. Less than a third advised participants to use pharmacotherapy to help them quit smoking. More than 50% of the GPs restarted the protocol after relapse following initial smoking cessation." B. Use of bupropion or NRT, or both was not only reserved for the participants for whom it was mandatory. (Additional information from author)