Objective: To examine the association between the smoking status of general practitioners (GPs) and abstinence rates among patients receiving GP-delivered brief advice for smoking cessation.
Methods: A quasi-experimental multilevel study with follow-up assessments at 6, 12, 18, and 24 months after baseline was conducted using a random sample of 39 general practices in a defined area (participation rate=87.2%). Patients aged 18-70 were consecutively screened for smoking status (n=11,560) over the course of 3 weeks and were assigned to a control group (week 1), a computer expert system intervention (week 2), or a personal counselling intervention with the GP (week 3). For the current analysis, patients participating in study week 2 were excluded. A total of 1260 patients fulfilled the inclusion criteria and 80.2% took part: 609 patients in study week 1 and 402 patients from study week 3. GPs participated in a training session concerning smoking counselling, which was held between study weeks 2 and 3. Self-reported 4-week and 6-month prolonged abstinence measures at the 6-, 12-, 18-, and 24-month follow-ups were assessed.
Results: The smoking status of the GP was neither significantly related to 4-week prolonged abstinence nor 6-month prolonged abstinence among patients in a main effects model. Further modelling revealed that the intervention group modified the effect of the non-smoking status of the GP on the likelihood to quit smoking. A significant interactive effect was found between the non-smoking status of the GP and the intervention group on both abstinence measures.
Conclusion: The non-smoking status of the GP had a positive effect among counselled patients.
Practice implications: The consideration of lifestyle behavioural variables such as the smoking status of the GP will be essential for further research concerning the efficacy of smoking interventions.