To identify lung and head-and-neck cancer patients who will have difficulty stopping smoking it is necessary to measure the severity of their nicotine dependence. In this study, we compiled a Japanese version of the Fagerstrom test for nicotine dependence (FTND) and examined its reliability and validity. One hundred and fifty-one cancer patients participated in this study and took our Japanese version of the FTND. Socio-demographic and medical data and information about smoking habits were obtained from a semi-structured interview, and the patients' nicotine dependence was evaluated according to the Diagnostic and Statistical Manual of Mental Disorders, 3rd Ed., Rev. (DSM-III-R). The mean FTND scores+/-SD of the group with nicotine dependence and the group without nicotine dependence were 6.85+/-2.00 and 3.70+/-2.13 respectively, and the difference was significant (P<0.001, Mann-Whitney's U-test). The test-retest correlation was 0.75. Cronbach's alpha of the FTND was 0.66. The FTND score correlated significantly with the number of satisfied criteria of nicotine dependence (r=0.70; P<0.001, Pearson's correlation). By using a receiver-operating-characteristic curve, we determined a score of 5/6 as a suitable cut-off point for nicotine dependence; this point gave high sensitivity and specificity (0.75 and 0.80, respectively). These results suggest that our Japanese version of FTND is a reliable and valid measure of nicotine dependence in patients with smoking-related cancers.