BACKGROUND: Smokers who use nicotine replacement therapy (NRT) to aid smoking reduction (SR) are more likely to quit smoking than those who try to reduce without NRT. This could be because NRT enhances the likelihood of quitting or because those who are motivated to quit choose to use NRT (i.e., selection bias). METHODS: 'Propensity score matching' was used to assess whether the increased likelihood of a quit attempt in those using NRT for SR would remain in a subsample of smokers paired on variables indicative of the likelihood of making a quit attempt and using NRT (i.e., when selection bias is reduced). Measures were obtained on a range of smoking and socio-demographic variables at baseline, then after 3 and 6 months. Fifty-eight smokers who were attempting SR at 3 months were matched on baseline measures to 58 smokers not using NRT for SR. The odds of their going on to make a quit attempt in the following 3 months were then compared. RESULTS: In smokers matched on motivational and other variables for their propensity to use NRT to aid smoking reduction, those using NRT for SR had three times greater odds of reporting a quit attempt than those not using NRT (OR 3.23; CI 1.49-7.01; p