National Cancer Institute
Last Modified: February 3, 2012
Note: A separate PDQ® summary on Levels of Evidence for Cancer Screening and Prevention Studies is also available.
The cancer prevention summaries in PDQ® refer to cancer prevention, defined as a reduction in the incidence of cancer. The PDQ® includes summaries generally classified by histological type of cancer, especially when there are known risk factors for the specific types of cancer. This summary addresses a specific risk factor, tobacco use, which is associated with a large number of different cancers (and other chronic diseases) and unequivocally contains human carcinogens. 1 The focus of this summary is on clinical interventions by health professionals that decrease the use of tobacco.
Based on solid evidence, cigarette smoking causes cancers of the lung, oral cavity and pharynx, larynx, esophagus, bladder, kidney, pancreas, stomach, uterine cervix, and acute myeloid leukemia. 2 Smoking avoidance and smoking cessation result in decreased incidence and mortality from cancer.
Counseling and Smoking Cessation
Based on solid evidence, counseling by a health professional improves smoking cessation rates.
Physician Advice and Smoking Cessation
Based on solid evidence, simple advice from a physician to stop smoking improves smoking cessation rates.
Drug Treatment and Smoking Cessation
Based on solid evidence, drug treatments, including nicotine replacement therapies (gum, patch, spray, lozenge, and inhaler), selected antidepressant therapies (e.g., bupropion), and nicotinic receptor agonist therapy (varenicline), result in better smoking cessation rates than placebo.
In the United States, smoking-related illnesses accounted for an estimated 443,000 deaths each year between 2000 and 2004. 1 On average, these deaths occur 12 years earlier than would be expected, so the aggregate annual loss exceeds 5 million life-years. 2 These deaths are primarily due to smoking's role as a major cause of cancer, cardiovascular diseases, and chronic lung diseases. The known adverse health effects also include other respiratory diseases and symptoms, nuclear cataract, hip fractures, reduced female fertility, and diminished health status. Maternal smoking during pregnancy is associated with fetal growth restriction, low birth weight, and complications of pregnancy. 3 It has been estimated that at least 30% of cancer deaths and 20% of all premature deaths in the United States are attributable to smoking. 1
Tobacco products are the single, major avoidable cause of cancer, causing more than 155,000 deaths among smokers in the United States annually due to various cancers. 4 The majority of cancers of the lung, trachea, bronchus, larynx, pharynx, oral cavity, nasal cavity, and esophagus are attributable to tobacco products, particularly cigarettes. Smoking is also causally associated with cancers of the pancreas, kidney, bladder, stomach, and cervix and with myeloid leukemia. 3 5
Smoking also has substantial effects on the health of nonsmokers. Environmental or secondhand tobacco smoke is implicated in causing lung cancer and coronary heart disease. 6 Among children, secondhand smoke exposure is causally associated with sudden infant death syndrome, lower respiratory tract illnesses, otitis media, middle ear effusion, exacerbated asthma, and respiratory effects such as cough, wheeze, and dyspnea. 6
Environmental tobacco smoke has the same components as inhaled mainstream smoke, although in lower absolute concentrations, between 1% and 10%, depending on the constituent. Carcinogenic compounds in tobacco smoke include the polycyclic aromatic hydrocarbons (PAHs), including the carcinogen benzo[a]pyrene (BaP) and the nicotine-derived tobacco-specific nitrosamine, 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK). 7 Elevated biomarkers of tobacco exposure, including urinary cotinine, tobacco-related carcinogen metabolites, and carcinogen-protein adducts, are seen in passive or secondhand smokers. 6 8 9 10
In 2007, 22.3% of adult men and 17.4% of adult women in the United States were current smokers. 11 (Also available online.) Cigarette smoking is particularly common among American Indians and Alaska Natives. The prevalence of smoking also varies inversely with education and was highest among adults who had earned a General Educational Development diploma (44.0%) and high school dropouts (33.3%) and generally decreased with increasing years of education. 11 (Also available online.) Cigarette smoking prevalence among male and female high school students increased substantially during the early 1990s in all ethnic groups but appears to have been declining since approximately 1996. 12 13 (Also available online.)
The effect of tobacco use on population-level health outcomes is illustrated by the example of lung cancer mortality trends. Smoking by women increased between 1940 and the early 1960s, resulting in a greater than 600% increase in female lung cancer mortality since 1950. Lung cancer is now the leading cause of cancer death in women. 12 14 In the last 30 years, prevalence of current cigarette use has generally decreased, though far more rapidly in males. Lung cancer mortality in men peaked in the 1980s, and has been declining since then; this decrease has occurred predominantly in squamous cell and small cell carcinomas, the histologic types most strongly associated with smoking. 12 Variations in lung cancer mortality rates by state also more or less parallel long-standing state-specific differences in tobacco use. Among men, the average annual age-adjusted lung cancer death rates from 2001 to 2005 were highest in Kentucky (111.5 per 100,000), where 29.1% of men were current smokers in 1997, and lowest in Utah (33.7 per 100,000), where only 10.4% of men smoked. Among women, lung cancer death rates were highest in Kentucky (55.9 per 100,000), where 28.0% of women were current smokers, and lowest in Utah (16.9 per 100,000), where only 9.3% of women smoked. 12
Many health risks related to tobacco smoking can be reduced by smoking cessation. Smokers who quit smoking before age 50 years have up to half the risk of dying within 15 years compared with people who continue to smoke, and the risk of dying is reduced substantially even among persons who stop smoking after age 70 years. 1 The risk of lung cancer is 30% to 50% lower than that of continuing smokers after 10 years of abstinence, and the risk of oral and esophageal cancer is halved within 5 years of cessation. 1 Smokers who quit also lower their risk of cervical, gastric, and bladder cancer. 1 2 3
In a randomized trial of heavy smokers, the long-term follow-up results demonstrated that compared with the nonintervention group (n = 1,964), those randomly assigned to a smoking cessation intervention (n = 3,923) experienced a 15% reduction in all-cause mortality rates (8.83 vs. 10.38 per 1,000 person-years; P = .03). 4 The smoking cessation intervention consisted of a strong physician message plus 12 group sessions and nicotine gum administered during a 10-week period. Decreases in the risk of lung and other cancers, and coronary heart disease, cardiovascular disease, and respiratory disease contributed to the benefit in the group randomly assigned to the smoking cessation intervention.
A number of approaches at the policy, legislative, and regulatory levels have been attempted to effect widespread reduction in or prevention of commencement of tobacco use. Various efforts at the community, state, and national level have been credited with reducing the prevalence of smoking over time. These efforts include, reducing minors' access to tobacco products, disseminating effective school-based prevention curricula together with media strategies, raising the cost of tobacco products, using tobacco excise taxes to fund community-level interventions including mass media, providing proven quitting strategies through health care organizations, and adopting smoke-free laws and policies. 5 6 School-based interventions alone have not demonstrated long-term impact for smoking prevention. 7 One of the highest quality studies was a large, randomized trial in which children received a theory-based program that incorporated various social-influence approaches from grade 3 through grade 12, with no difference in smoking outcomes observed either at grade 12 or at 2 years after high school between school districts receiving the intervention and those in the control arm. 8
The Community Intervention Trial for Smoking Cessation (COMMIT) was a National Cancer Institute-funded large-scale study to assess a combination of community-based interventions designed to help smokers cease using tobacco. COMMIT involved 11 matched pairs of communities in North America, which were randomly assigned to an arm offering an active community-wide intervention or a control arm (no active intervention). 9 The 4-year intervention included messaging through existing media channels, major community organizations, and social institutions capable of influencing smoking behavior in large groups of people. The interventions were implemented in each community through a local community board that provided oversight and management of COMMIT activities.
In COMMIT, there was no difference in the mean quit rate of heavy-smokers in the intervention communities (18.0%) compared with the control communities (18.7%). The light-to-moderate smoker quit rates were statistically significantly different: averages of 30.6% and 27.5% for the intervention and control communities, respectively (P = .004). Although no significant differences in quit rates between the sexes were observed, less-educated light-to-moderate smokers were more responsive to the intervention than were college-educated smokers with a light-to-moderate habit. 10 11
Clinical interventions targeted at individuals have shown more promising results. A meta-analysis of randomized controlled trials shows that 6-month cessation rates are significantly improved with use of nicotine replacement therapy (NRT) products compared with placebo or no intervention (summary relative risk [RR], 1.58; 95% confidence interval [CI], 1.501.66). 12 The benefits of nicotine replacement therapy product use have been consistently observed regardless of whether the product used was the patch, gum, nasal spray, inhaler, or lozenge. 12 Smoking cessation counseling alone is also effective; 13 even a brief intervention by a health care professional significantly increases the smoking cessation rate. 14
An important issue is whether pharmacotherapies are more effective in the presence of counseling. A randomized trial compared the following three levels of intervention that combined free pharmacotherapy (nicotine patch or bupropion) with or without counseling: 1) pharmacotherapy alone; 2) pharmacotherapy plus up to two counseling sessions every 6 months; and 3) pharmacotherapy plus up to six counseling sessions every 6 months. During the 24-month study, each group was offered a randomly assigned intervention at baseline, 6 months, 12 months, and 18 months later. For the primary study endpoint of 7-day point prevalence of smoking abstinence after 24 months of follow-up, no statistically significant differences were observed among the interventions. 15 The results of this study suggest that the combination of pharmacotherapy plus counseling is no better than intervention alone.
Promoting smoking cessation among cancer survivors is essential because the deleterious health effects of cigarette smoking may impact prognosis in both the short term and long term. In a randomized controlled trial of a peer-delivered smoking cessation intervention among childhood cancer survivors, a significantly higher 12-month quit rate was observed in the intervention group (15% vs. 9%; P < .01). 16
In 1996, the Agency for Health Care Policy and Research (AHCPR), now the Agency for Healthcare Research and Quality released a landmark set of clinical smoking-cessation guidelines for helping nicotine-dependent patients and healthcare providers. Now sponsored by the Public Health Service, the updated 2008 guidelines, "Treating Tobacco Use and Dependence" are available online. 17 The broad elements of these guidelines are:
For individual interventions, the guidelines 17 suggest a model based on outcomes from six major clinical trials of physician-delivered smoking intervention conducted in the late 1980s: 18 the ASK, ADVISE, ASSESS, ASSIST, and ARRANGE model. The physician provides a brief intervention that entails asking about smoking status at every visit, advising abstinence, assisting by setting a quit date, providing self-help materials and recommending use of nicotine replacement therapy, and arranging for a follow-up visit. See below for brief and expanded intervention outlines. The recommendations also strongly support the value of referral to more intensive counseling.
Pharmacotherapy for Smoking Cessation
Pharmacological agents have been used successfully to aid in the cessation of smoking in the general population. 19 Since the original AHCPR guidelines 20 were published in 1996, various nicotine replacement products have been approved for over-the-counter sale, and additional evidence of the effectiveness of therapies for smoking cessation has been published. 21 22 23 24 Pharmacotherapy of smoking, including nicotine replacement therapies (gum, patch, spray, lozenge, and inhaler) and nonnicotine medications (e.g., bupropion and varenicline) result in statistically significant increases in smoking cessation rates compared with a placebo. Based on a synthesis of the results of 110 randomized trials, nicotine replacement therapy treatments, alone or in combination, improve cessation rates over placebos after 6 months (RR, 1.58; 95% CI, 1.501.66). 12
There are also nonnicotine pharmacotherapies that have been efficacious for smoking cessation, including bupropion and varenicline. Based on the results of 31 randomized trials that compared the antidepressant bupropion to placebo, after 6 months of follow-up, bupropion was associated with a statistically significant increase in the likelihood of quitting smoking (summary odds ratio [OR], 1.94; 95% CI, 1.722.19). 25 There is insufficient evidence to support the idea that combining bupropion plus NRT increases smoking cessation rates compared to NRT alone (summary OR, 1.37; 95% CI, 0.652.91). 25
Varenicline is a selective alpha-4-beta-2 nicotinic acetylcholine receptor partial agonist. In two randomized controlled trials for smoking cessation, varenicline titrated to a dose of 1.0 mg twice a day and was compared with bupropion sustained-release (SR) 150 mg twice a day and with a placebo group. 26 27 Treatment lasted for 12 weeks, with an additional 40 weeks of posttreatment follow-up. In both studies, varenicline was more efficacious than bupropion and placebo for continuous abstinence from smoking at 9 to 12 weeks and at 9 to 24 weeks of follow-up. For 9 to 52 weeks of follow-up, varenicline was more efficacious than placebo in both studies. 26 27 At 52 weeks of follow-up, the 7-day point prevalence of smoking abstinence was 46% higher in the varenicline group than in the bupropion SR group (OR, 1.46; 95% CI, 1.042.06). 26 The other study also showed a 46% higher continuous abstinence in the varenicline group (OR, 1.46; 95% CI, 0.992.17). 27 Approximately 30% of the participants who were randomly assigned to receive varenicline reported nausea, more than double that in the bupropion groups, and triple that seen in the placebo groups. In a randomized trial comparing varenicline with transdermal nicotine, continuous abstinence was greater in the varenicline group than in the transdermal nicotine group at the end of treatment (56% vs. 43%; P < .001) and during posttreatment follow-up (26% vs. 20%; P = .06). 28 The prevalence of nausea in the varenicline group (37%) was more than triple that in the transdermal nicotine group (10%).
Based on postmarketing surveillance, on July 1, 2009, the U.S. Food and Drug Administration (FDA) required additions to the Boxed Warnings for both bupropion and varenicline to describe the risk of serious neuropsychiatric symptoms associated with these products. Symptoms include: changes in behavior, hostility, agitation, depressed mood, suicidal thoughts and behavior, and attempted suicide. 29 The FDA goes on to advise that the important health benefits of quitting smoking should be weighed against the small, but real, risk of serious adverse events with the use of varenicline or bupropion. 29 A meta-analysis of double-blind, placebo-controlled, randomized trials of varenicline administered for at least 1 week (N = 14 trials) indicated that varenicline was associated with an increased risk of serious adverse cardiovascular events (RR, 1.72; 95% CI, 1.092.71). 30 This finding is particularly noteworthy because almost all of the randomized trials included in the meta-analysis had the following in common: they excluded patients with cardiovascular disease (CVD) at baseline; the usual average age of the patient populations (early 40s) was young for CVD; varenicline was usually administered for only 12 weeks or less; and, varenicline is efficacious for smoking cessation, which would act to decrease CVD risk.
There is a growing number of smoking cessation pharmacotherapies that have been shown to be efficacious in significantly increasing rates of smoking cessation. The choice of therapy should be individualized based on a number of factors, including past experience, patient and/or physician preference, and potential agent side effects. As more is learned about specific genetic variants that influence a smoker's response to smoking cessation pharmacotherapiessuch as polymorphisms in genes encoding enzymes involved in nicotine metabolismthis information could eventually be integrated into smoking cessation treatment planning. 31 Presently, the evidence is not yet sufficient to be integrated into clinical practice.
The following sections summarize available pharmacologic interventions to assist in tobacco cessation. More comprehensive information is available from product package inserts.
Pharmacologic interventions to assist in tobacco cessation
These products are designed to aid in the withdrawal symptoms associated with nicotine. Several precautions are warranted before initiating therapy, but it should be noted that these precautions do not constitute absolute contraindications. In particular, special considerations are necessary in some patient groups (e.g., those with medical conditions such as arrhythmias, uncontrolled hypertension, esophagitis, peptic ulcer disease, insulin-treated diabetes, or asthma, pregnant or breast-feeding women, and adolescent smokers). 32
| Brand | Dose | Side Effects | Comments | |
| Rx | Habitrol | 721 mg/d | Erythema | Use for 612 wk |
| OTC | Nicoderm CQ | 721 mg/d | Pruritus | Use for 612 wk |
| OTC | Nicotrol | 515 mg/d | Burning at site | Use for 1420 wk |
| Rx | ProStep | 1122 mg/d | Local irritation | Use for 612 wk |
| d = day; OTC = over the counter; Rx = prescription; wk = week. | ||||
Current guidelines recommend 8 weeks of transdermal nicotine therapy. Findings from two randomized placebo-controlled trials of transdermal therapy are divergent in their findings as to whether extended therapy (2224 weeks vs. 8 weeks) improves quit rates. 33 34
| Brand | Dose | Side Effects | Comments | |
| OTC | Nicorette | 1824 mg/d | Stomatitis, sore throat | Max 30 pieces/d; decrease 1 piece every 47 d |
| OTC | Nicorette DS | 3648 mg/d | Jaw ache | Max 20 pieces/d; decrease 1 piece every 47 d |
| d = day; OTC = over the counter. | ||||
| Brand | Dose | Side Effects | Comments | |
| OTC | Commit | 4080 mg/d | Local irritation (warmth and tingling) | Use for 12 wk; max 20 pieces/d. Wk 16: 12 lozenges every 12 h; wk 79: 1 lozenge every 24 h; wk 1012: 1 lozenge every 48 h |
| d = day; h = hour; OTC = over the counter; wk = week. | ||||
| Brand | Dose | Side Effects | Comments | |
| Rx | Nicotrol Inhaler | Individualized | Local irritation | Use up to 24 wk |
| Rx = prescription; wk = week. | ||||
| Brand | Dose | Side Effects | Comments | |
| Rx | Nicotrol NS | Max 40 mg/d | Nasal irritation | Max use 3 mo |
| d = day; mo = month; Rx = prescription. | ||||
Also used as an antidepressant, bupropion HCl is a nonnicotine aid to smoking cessation. It is a relatively weak inhibitor of the neuronal uptake of norepinephrine, serotonin, and dopamine, and does not inhibit monoamine oxidase. The exact mechanism by which bupropion HCl enhances the ability of patients to abstain from smoking is unknown; however, it is presumed that this action is mediated by noradrenergic or dopaminergic mechanisms.
| Brand | Dose | Side Effects | Warning/Precaution | |
| Rx | Zyban | 150 mg/d X 3 d then increase to 300 mg/d X 712 wk | Insomnia, dry mouth, dizziness, rhinitis | Do not take with Wellbutrin or Wellbutrin SR |
| Higher incidence of seizures in patients treated for bulimia, anorexia | ||||
| Do not prescribe >300 mg/d for patients being treated for bulimia | ||||
| d = day; Rx = prescription; wk = week. | ||||
| Brand | Dose | Side Effects | Warning/Precaution | |
| Rx | Chantix | 0.5 mg/d, d 13; 0.5 mg twice a d, d 47; then 1.0 mg twice a d through wk 12 | Nausea, insomnia | Risk of toxicity greater in patients with impaired renal function |
| Not tested in children and pregnant women | ||||
| d = day; Rx = prescription; wk = week. | ||||
Although Zyban (bupropion HCl) is the only antidepressant approved by the FDA for smoking cessation, Prozac (fluoxetine HCl) has been shown to be effective. 35
| Brand | Dose | Side Effects | Comments | |
| Rx | Prozac | 3060 mg/d | Insomnia, dizziness, anorexia, sexual dysfunction, confusion | Limited data available on its use in combination with cognitive-behavioral therapy |
| d = day; Rx = prescription. | ||||
Lobeline (Bantron) is classified as a category III agent by the FDA, safe but no proven effectiveness. This product is not recommended for use in any smoking cessation program due to its lack of efficacy. 36
Clonidine and nortriptyline have been suggested as possibly useful second-line pharmacotherapies, but are not approved for smoking cessation by the FDA. Nortriptyline is an antidepressant that does not contain nicotine. A meta-analysis of five randomized controlled trials found that smokers who received nortriptyline were 2.4 times more likely (95% CI, 1.73.6) than smokers who received a placebo to remain abstinent from smoking after 6 months. 37
Among dependent smokers, complete abstinence from smoking is the ultimate goal. Even in instances when complete abstinence from smoking is not achieved, smoking cessation pharmacotherapies may benefit individual healthand ultimately the public's healthif the smoker reduces the number of cigarettes smoked. The relationship between cigarette smoking and lung cancer, and other smoking-associated malignancies, is strongly dose-dependent. Thus, an individual smoker who is unable to achieve abstinence or who is not motivated to quit smoking may benefit by using pharmacotherapies (or other means) to reduce the number of cigarettes smoked per day. NRT has thus generated attention as a viable means of harm reduction. In studies that randomly assigned smokers who were not trying to quit to NRT or placebo, a greater proportion of those randomly assigned to NRT compared with placebo reduced the number of cigarettes per day. 38 39 However, the impact of NRT on smoking reduction appears not to be sustained in the long run. 40 Less evidence is available for bupropion, varenicline, and psychosocial interventions as a means of harm reduction. A potential problem with such a harm reduction strategy would be if it prevented cessation among smokers who would have otherwise quit smoking. Evidence shows that smoking reduction is actually associated with increased likelihood of future cessation. 39 41 Another potential negative aspect of harm reduction would be if smokers reduced the number of cigarettes smoked per day but modified the way the cigarettes were smoked in such a way that exposure to tobacco toxins was not actually reduced (e.g., by inhaling more deeply). Compensatory behaviors such as inhaling more deeply or smoking more of a cigarette are attempts by the smoker to try to maintain nicotine levels, so the use of supplemental NRT presumably safeguards against this. Evidence from studies of smoking reduction with NRT that measured smoking biomarkers indicates that compensation occurs, but not to such an extent that it would be expected to outweigh the reduction in exposure from the reduced number of cigarettes per day. 38