Background: Guidelines recommend that smoking cessation interventions are offered in all clinical settings to all smokers willing to make a quit attempt. Since the effectiveness of routine provision of behavioural counselling and nicotine replacement therapy (NRT) to smokers admitted to hospital has not been established, a randomised controlled trial of these interventions given together compared with counselling alone or minimal intervention was performed in hospital inpatients. Methods: Medical and surgical inpatients who were current smokers at the time of admission were randomised to receive either usual care (no additional advice at admission), counselling alone (20 minute intervention with written materials), or NRT plus counselling (counselling intervention with a 6 week course of NRT). Continuous and point prevalence abstinence from smoking (validated by exhaled carbon monoxide <10 ppm) was measured at discharge from hospital and at 3 and 12 months, and self-reported reduction in cigarette consumption in smokers was assessed at 3 and 12 months. Results: 274 inpatient smokers were enrolled. Abstinence was higher in the NRT plus counselling group (n=91) than in the counselling alone (n=91) or usual care (n=92) groups. The difference between the groups was significant for validated point prevalence abstinence at discharge (55%, 43%, 37% respectively, p=0.045) and at 12 months (17%, 6%, 8%, p=0.03). The respective differences in continuous validated abstinence at 12 months were 11%, 4%, 8% (p=0.25). There was no significant difference between counselling alone and usual care, or in reduction in cigarette consumption between the treatment groups. Conclusions: NRT given with brief counselling to hospital inpatients is an effective routine smoking cessation intervention.