AIMS:Lack of adherence to smoking cessation medication regimens is assumed to play a significant role in limiting their effectiveness. This study aimed to assess evidence for this assumption. METHODS:A systematic search was conducted, supplemented by expert consultation, of papers reporting on randomized trials and observational studies examining the association between adherence to cessation medication and the success of quit attempts. To rule out reverse causality, only studies where adherence was assessed prior to relapse were included. Five studies met the inclusion criteria and results were extracted independently by two researchers. Heterogeneity between studies precluded a pooled analysis of the data. RESULTS:Studies varied widely with regard to both the definition of adherence and outcome measures. The included studies only addressed adherence to nicotine replacement therapy. One study of lozenge use found that the amount of medication used between 1 and 2 weeks after the quit date predicted abstinence at 6 weeks [adjusted odds ratio (OR) for 'high' versus 'low' lozenge use 1.25; 95% confidence interval (CI) = 1.05-1.50; P < 0.02]. Similarly, one study found a significant impact of oral nicotine consumption during the first week on abstinence at 4 weeks (adjusted OR per additional mg/day = 1.05; CI = 1.01-1.10). Another study found that participants using nicotine replacement therapy for at least 5 weeks were significantly more likely to self-report continuous abstinence at 6 months. The remaining two studies failed to find a significant effect of treatment duration on outcome at 1 and 2 years but had very low power to detect such an effect. CONCLUSIONS:There is modest evidence to support the assumption that lack of adherence to nicotine replacement therapy regimens undermines effectiveness in clinical studies.