The Evolving Pharmacotherapy of Smoking Cessation

Over a decade has passed since the first nicotine-replacement product appeared on the U.S. market. In 1984, nicotine gum became the first medication of any type approved for smoking cessation by the Food and Drug Administration, and it revolutionized therapy. Over the next 13 years, other nicotine products followed—transdermal patches in 1992, a nasal spray in 1996, and an inhaler in 1997. Even newer ways to deliver nicotine into the bloodstream, including a sublingual tablet, a lozenge, and a nicotine “lollipop,” are in development. In the meantime, the nicotine patch and gum have become mainstays of treatment. Their efficacy is supported by the results of multiple randomized controlled trials and, more recently, by a series of meta-analyses which demonstrate that the patch more than doubles the likelihood of long-term success in a smoking cessation attempt, while the gum increases cessation rates by 50% to 60%.^1–3 In 1996, an evidence-based clinical guideline for smoking cessation from the Agency for Health Care Policy and Research established nicotine replacement as the standard of care, concluding that both the nicotine patch and gum were effective and recommending that, “except in the presence of serious medical precautions, the clinician should encourage the use of nicotine replacement with patients who smoke.”⁴ Nonetheless, unanswered questions remain about the use of these products. Two articles in this issue of JGIM address questions about the nicotine patch: the generalizability of its benefit to disadvantaged populations and its value in blunting the weight gain that follows smoking cessation.

Ahluwalia and colleagues ⁵ consider whether the patch is as effective in poor African-American smokers as it has been shown to be in the predominantly middle-class white smokers in whom it has been tested. The prevalence of smoking—and many smoking-related diseases—is higher among individuals with less education and lower incomes,⁶ yet these smokers have less access to treatment, especially nicotine-replacement therapy, which is expensive and not covered by most Medicaid plans. Demonstrating the patch's efficacy in a low-income population would strengthen the case for Medicaid reimbursement, thereby removing a barrier to good preventive care for smokers who are poor.

Ahluwalia and colleagues tested the efficacy of the nicotine patch in low-income, urban African Americans with limited health insurance coverage. Both the study design, a double-blind, randomized, placebo-controlled trial, and the intervention, a 10-week course of free nicotine patches accompanied by a moderately intense behavioral program, were similar to those in previous nicotine patch trials. At the end of treatment, the tobacco abstinence rate was 57% higher among patients receiving nicotine than among those receiving placebo. The difference was statistically significant with the one-tailed test used by the investigators. However, the difference between groups diminished and failed to reach statistical significance at the 6-month follow-up, nearly 4 months after the end of treatment. Furthermore, the actual smoking cessation rates achieved with nicotine replacement (22% at the end of treatment and 17% at 6 months) were lower than those in most previous trials, and they would probably have been even lower had they been biochemically validated to correct for the overreporting of abstinence that occurs in intervention trials.

Thus, despite a strong record in earlier clinical trials, the nicotine patch had no long-term benefit among the poor African-American smokers in this study. Why not? The smoking patterns of whites and African Americans do differ,⁶ but there is no a priori reason to expect that nicotine dependence varies by ethnicity or income. The investigators note that the participants' level of nicotine dependence, as measured by a standard scale, was lower than in other patch studies, but the patch has been successful even in smokers with lower levels of dependence.⁴ A more likely explanation for the failure of the patch in this study was participants' poor adherence to the treatment program, which meant that most patients received a substantially weaker intervention than intended. This occurred despite the removal of financial barriers to treatment, an intervention manual tailored to the concerns of African Americans, and aggressive efforts by the investigators to maintain contact with participants who said that they wanted to quit. What explains the poor compliance of smokers who want to quit and have no financial barriers to treatment? Clearly, other obstacles to success exist in low-income communities. Substance abuse is only one example of a comorbidity that reduces the success of smoking treatment. It was not measured in this study, but it is likely to have been present at a higher rate than in other patch trials, which generally excluded smokers with this problem.

Nonetheless, it would be incorrect to conclude from this one study that the nicotine patch is not effective in poor African Americans. Rather, the study reminds us that the context in which nicotine replacement is provided strongly affects its success, perhaps especially so in low-income smokers. Changing behavior in disadvantaged populations may require not only a greater intensity of treatment but also different treatment approaches, such as linking with existing community structures (e.g., churches), reliance on audiovisual rather than written material for low-literacy populations, outreach efforts, and perhaps even offering financial incentives to motivate smokers to stop. The lesson of the study is that more needs to be learned about how to work effectively with the urban poor—regardless of color—to reduce smoking.

The other study on the nicotine patch in this issue, by Dale and coworkers,⁷ explores its value for avoiding the troubling weight gain that follows smoking cessation. Identifying a way to blunt, delay, or ideally prevent this weight gain could remove this often-cited barrier to attempting cessation, especially for women. The weight gain is partly attributable to quitters' loss of nicotine, which increases resting metabolic rate and thus energy expenditure.⁸ There was hope that nicotine replacement therapy, which maintains an individual's exposure to nicotine while smoking stops, might help smokers avoid weight gain, but the evidence has been disappointing.⁹ Most patch trials found no statistically significant effect on weight gain.

Dale and colleagues hypothesized that the failure of nicotine patches to alter postcessation weight gain might be attributed to the fact that standard patch doses produce lower blood nicotine levels than smoking does. They tested the effect of different nicotine patch doses, including higher than normal doses, on the amount of weight gained after cessation. They measured the degree to which the smokers' baseline blood level of cotinine (a nicotine metabolite) was replaced by their patch dose. Among patients who were abstinent for 8 weeks of nicotine patch therapy, weight gain was independently and inversely associated with the degree to which blood cotinine levels were replaced, but not, interestingly, with the actual patch dose. The investigators did not simultaneously measure patients' caloric intake or physical activity levels and could therefore not exclude the possibility of confounding by these factors, although that seems unlikely. Once nicotine therapy stopped, the relation disappeared, suggesting that better replacement of blood nicotine levels delays, but does not prevent, postcessation weight gain.

In short, there is still no easy answer for the thorny problem of weight gain after smoking cessation. No drug does more than delay weight gain, and neither behavioral programs nor diets have demonstrated efficacy.⁸ Dale and colleagues suggest the use of higher nicotine patch doses in selected smokers concerned with weight gain. Combining the patch with nicotine gum is a better idea, in my view, because the gum, unlike the patch, has shown some benefit in delaying weight gain,⁸ probably because it not only provides nicotine but also keeps the smoker's mouth busy with something other than food. Newer products such as the nicotine inhaler might have a similar effect, though they have not been studied. A better alternative is probably the use of sustained-release bupropion (Zyban), an antidepressant newly approved for smoking cessation. In a recent study, the drug at recommended doses reduced the amount of weight gained, although the effect did not last after drug treatment stopped.¹⁰ This finding needs to be replicated. Future studies could also examine the efficacy of a combination of bupropion and nicotine replacement for preventing weight gain.

It is possible, however, that no drug will be able to blunt postcessation weight gain once the quitter stops taking it. Even delaying weight gain may help, if it allows a smoker to break this most deadly habit and manage the extra pounds later. In the meantime, encouraging smokers to increase their physical activity when quitting is probably the best strategy because exercise is the best long-term approach to weight control and preliminary evidence supports its value in smoking cessation. Physicians should also realize that what many smokers fear is a large weight gain. Smokers may be comforted to learn that the average weight gain is only 6 to 10 pounds—less than they fear—and that only about 10% gain more than 25 pounds.¹¹

These studies highlight some of the limits of existing smoking cessation treatment. Meanwhile, the pharmacotherapy of smoking cessation has continued to evolve. Nicotine gum and patches are now sold without a prescription, newer nicotine replacement products are on the horizon, and nonnicotine agents are beginning to appear. It will take time for the role of each of these agents to be clearly defined. Which should be first-line agents? Will combinations of agents achieve better results than single agents? Are these drugs, especially the newer ones, effective in low-income populations in which smoking is more common? If so, will cost be a barrier to their use? And how will the rapidly changing public policy landscape affect patients' interest in quitting or their ability to do so? Future studies will need to address these and other questions if we are to reduce tobacco use, the leading preventable cause of death in the United States. Stay tuned.