Karl Fagerström

Around 70% of smokers say they want to quit smoking. However, if they are asked to set a quit date ‘right now’, the figure comes down to around 5–10%. In practice, we need to realise that there is a range of motivation for change. Smokers can be grouped into those wanting and not wanting to quit. Among those wanting to quit, some may have tried and failed so many times that they have more or less given up trying. Among those not wanting to quit, there are those who want to manage their smoking, get it under better control and reduce it. There is also a small group that do not want any change.

We explored the relationship between the Fagerström Test for Nicotine Dependence (FTND) and smoking abstinence rates in 10 randomized, double-blind placebo-controlled Phase 2–4 varenicline studies. Participants were adult smokers (≥10 cigarettes/day) who were motivated to quit. Efficacy end points included continuous abstinence rate (CAR) for weeks 9–24 analyzed, by baseline FTND and Heaviness of Smoking Index (HSI) scores, and treatment. Data were analyzed using logistic regression models. Overall, 2,763 varenicline (M [SD] FTND score: 5.6 [2.2]) and 2,229 placebo subjects (5.5 [2.1]) were included in the analysis. An increase of one unit in baseline FTND or HSI score decreased the odds of abstinence at Week 24 by 11% (odds ratio [OR] 0.89, 95% CI 0.86–0.92, p < .0001) and 18% (OR 0.82, 95% CI 0.79–0.87, p < .0001), respectively. Treatment had a significant impact on CAR 9–24: odds of abstinence were increased threefold for varenicline versus placebo (OR 3.3, 95% CI 2.8–3.8, p < .0001). There was no interaction between treatment and FTND (p = .98) or HSI score (p = .97) for CAR 9–24. The HSI score predicted abstinence outcome as effectively as the FTND score. Abstinence rates decreased with increasing dependence scores. There was no interaction between treatment and baseline FTND or HSI score, suggesting that they have no effect on the efficacy of varenicline versus placebo. These results also suggest that the HSI may be as effective at predicting smoking cessation outcome as the whole FTND questionnaire.

The International Classification of Diseases and the Diagnostic and Statistical Manual for diagnosing tobacco/nicotine dependence emphasize the dependence-producing drug nicotine. These diagnostic tools have been challenged on grounds of poor predictive validity, and they do not differentiate across various forms of nicotine-containing products. In fact, nicotine-containing products (e.g., tobacco cigarettes, smokeless tobacco [ST], waterpipe, electronic cigarettes [ECIGs], and nicotine replacement [NR] products) have very different characteristics both in terms of sensory and behavioral involvement and also in pharmacokinetic and pharmacodynamic effects. For example, a cigarette and a nicotine patch are very different on almost every one of these dimensions. When ability to stop using a nicotine/tobacco product is used as a criterion for dependence, success rates vary considerably across products: Tobacco cigarette cessation is more difficult than ST cessation that in turn is more difficult than NR product cessation. Based on these results, we hypothesize that there is a continuum of dependence as much as there is a continuum of harm, with tobacco cigarettes and NR products on opposite ends of both continua and other products (waterpipe and ECIGs) somewhere in between. In order to capture more precisely the dependence produced by both nicotine and its administration forms, product-specific instruments may be required. The pros and cons of this approach are discussed.