Objective: Within the framework of a randomized, active treatment controlled trial, we used a mediation analysis to understand the mechanisms by which an intervention that uses confrontation with spirometry for smoking cessation achieves its effects. Methods: Participants were 228 smokers from the general population with previously undetected chronic obstructive pulmonary disease (COPD), who were detected with airflow limitation by means of spirometry. They received two equally intensive behavioural treatments by a respiratory nurse combined with nortriptyline for smoking cessation: confrontational counselling with spirometry versus conventional health education and promotion (excluding confrontation with spirometry and COPD). Results: Cotinine validated abstinence rates from smoking at 5 weeks after the target quit date were 43.1% in the confrontational counselling group versus 31.3% in the control group (OR = 1.67, 95%CI = 0.97-2.87). The effect of confrontational counselling on abstinence was independently mediated by the expectation of getting a serious smoking related disease in the future (OR = 1.76, 95%CI = 1.03-3.00), self-exempting beliefs (OR = 0.42, 95%CI = 0.21-0.84), and self-efficacy (OR = 1.38, 95%CI = 1.11-1.73). Conclusion: We conclude that confrontational counselling increases risk perceptions and self-efficacy, and decreases self-exempting beliefs (risk denial) in smokers with previously undetected COPD. These changes in mediators are associated with a higher likelihood of smoking cessation. Practice implications: Apart from the intensity, the content of smoking cessation counselling may be an important factor of success. A confrontational counselling approach as we applied may have the potential to alter smoking-related cognitions in Such a way that smokers are more successful in quitting. Nurses can be trained to deliver this treatment.