, 1991, 2006; Shiffman protein inhibitor et al., 1997). Unlike the FTCQ sample, these participants had committed to quitting smoking. Factor scale means were higher on the FTCQ than on the FTCQ-12, especially for purposefulness (5.4 vs. 4.5). Additionally, there were small to substantial positive relationships among all factors on the FTCQ, yet there was an almost negligible relationship between purposefulness and expectancy, and no significant relationship for purposefulness with either compulsivity or emotionality on the FTCQ-12. There was also a negative loading for craving coupled with intention and planning to smoke with FTCQ-12 Item 7 (I would smoke as soon as I had an occasion). Desire was positively correlated with craving coupled with purposefulness in the TCQ validation study (Heishman et al., 2003).

Taken together, our findings are consistent with cognitive explanations of drug use (Tiffany, 1990; Tiffany and Drobes, 1991), which state that desire and intention to smoke cigarettes (i.e., purposefulness) are uncoupled in individuals attempting to quit smoking (West & Hardy, 2006). This might also explain why the variance accounted for by purposefulness in the FTCQ validation study was more than twice that of the FTCQ-12. There is evidence that the FTCQ-12 yields a measure of specific factors and general tobacco craving as well. Our results indicated that highly dependent smokers experienced heightened craving. We could rule in with 94.7% certainty (analysis not shown) that a participant with a General Craving Score ��6 was highly dependent on nicotine.

These findings also indicate that craving is an important element of nicotine dependence and smoking-related health outcomes (Berlin & Singleton, 2008). Findings are based on a sample of French smokers with smoking-related diseases, which limits generalizability. Another shortcoming of the study is the cross-sectional design (assessment only at baseline). Like the FTCQ, TCQ, and TCQ-SF (Heishman et al. 2008), the FTCQ-12 was developed to assess state-level changes in craving (Berlin et al., 2005; Heishman, Saha, & Singleton, 2004; Heishman et al., 2008; Singleton, Anderson, & Heishman, 2003) as opposed to trait-like concepts, such as situational outcome expectancies. Whether increases in craving found in repeated measure designs using the FTCQ-12 indicate actual increases in craving intensity or whether the factor structure of craving varies with this increase is unknown.

The most common clinical endpoint in tobacco cessation is abstinence, but we did not evaluate the predictive Cilengitide utility of the FTCQ-12 for successful quitting. The aim of this paper was to determine the psychometric characteristics of the FTCQ-12, then use it in the next phase of analyses in the ADONIS trial as a valid measure to potentially predict abstinence.