Nicotine dependence treatment for patients with cancer

Review and recommendations




Cancer patients who use tobacco demonstrate characteristics of strong nicotine dependence and are at increased risk for future tobacco-related morbidity and mortality. Continued smoking may contribute to poorer cancer treatment outcome and additional illness. In contrast, stopping smoking may improve quality of life and facilitate cancer treatment. Unfortunately, limited attention has been given to addressing tobacco use and treating nicotine dependence in cancer patients.


The authors reviewed the current literature on smoking cessation and nicotine dependence treatment in patients with cancer.


Nicotine dependence treatment in patients with cancer facilitates smoking cessation. Evidence that cancer patients are motivated and able to stop smoking support tobacco use treatment intervention. Research is needed to evaluate effective intervention to aid tobacco abstinence and evaluate related health benefits for this patient population.


Cancer patients are able to stop smoking and benefit from intervention. Treatment of nicotine dependence in cancer patients merits attention from researchers and healthcare providers. The health risks of continued smoking and clear benefits of tobacco abstinence provide an ethical imperative for treating nicotine dependence in cancer patients. Research is needed to evaluate effective intervention with the overall goal of promoting tobacco abstinence to benefit oncology treatment and patient health. Cancer 2003;98:632–44. © 2003 American Cancer Society.

DOI 10.1002/cncr.11538


Cigarette smoking is the most preventable cause of morbidity and mortality in men and women, accounting for one in five deaths in the U.S.1, 2 Tobacco use accounts for at least 30% of all cancer deaths, including carcinoma of the lung, mouth, larynx, pharynx, esophagus, pancreas, uterine cervix, kidney, and bladder. In the U.S., approximately 25% of adults3 and 35% of high school students4 report current cigarette use.

The smoking behavior of patients after a diagnosis of cancer is critical because continued smoking in cancer patients contributes to a number of adverse health effects. Continued smoking increases the number and severity of complications for cancer patients, as well as the risk for tumor progression and the development of a second primary cancer. Cancer patients who continue to smoke also have poorer survival compared with patients who do not continue to smoke.5–14 Because surgery, radiation, and chemotherapy are likely to produce more complications and greater morbidity among smokers than nonsmokers, cancer patients who continue to smoke may experience greater difficulties in cancer treatment.15, 16 Patients who continue to smoke also increase their risk of developing or exacerbating serious illnesses such as chronic obstructive pulmonary disease, coronary heart disease, peripheral artery disease, and stroke.17


Smoking cessation yields multiple health benefits for all smokers. Within hours of stopping smoking, blood pressure and heart rate decrease and reduced levels of carbon monoxide result in increased oxygen transport and availability.17 Within days of smoking cessation, improvements in smell, taste, circulation, and breathing are expected. Within months of stopping smoking, improvements to the immune system and metabolic system, increased energy, and decreased shortness of breath may be experienced.17 Smoking abstinence is associated with increased quality of life.18 For example, smokers who stopped smoking for 6 months experienced improved health status compared with smokers who continued to smoke. These patients reported benefits including improved general mental health, cognitive functioning, psychological well-being, and self-esteem.18

Cancer patients who stop smoking experience similar physical and psychological benefits, which also may enhance oncology treatment. Smoking cessation is associated with increased survival rates in patients with squamous cell carcinoma of the head and neck19 and in patients with small cell lung carcinoma.7 Smoking abstinence is also associated with reduced risk for subsequent malignancies.12 Lung carcinoma patients who stopped smoking reported decreased fatigue and shortness of breath, increased activity level, and improved respiratory capabilities, appetite, sleep, mood, and sense of self-worth.20 Even among advanced cancer patients, smoking cessation has improved pulmonary function10 and decreased some symptoms of toxicity in response to cancer treatment.21 There is also evidence that smoking cessation may enhance psychosocial adaptation to cancer diagnosis and treatment.22 For example, Gritz et al.23 found decreased self-reported anxiety in head and neck carcinoma patients who stopped smoking compared with patients who continued smoking. Although some clinicians may presume that smoking cessation treatment may be an unreasonable burden for patients with cancer,24 stopping smoking may contribute to improved physical functioning, increased quality of life, and enhanced oncology treatment for these patients.16 Consequently, it is important for healthcare providers to encourage patients to stop tobacco use and to provide methods of assistance.


Discussion of the health benefits and importance of stopping smoking for cancer patients raises three central questions. First, is nicotine dependence treatment needed in this group? There may be an assumption that patients who receive a diagnosis of cancer will stop smoking immediately and remain abstinent due to health concerns. However, if patients continue to smoke despite a cancer diagnosis, there would be evidence of nicotine dependence and a clinical need for formal intervention. Second, are cancer patients motivated and able to stop smoking? Unfortunately, some healthcare providers may hold a clinical bias that any patient who continues to smoke after a cancer diagnosis must be either unmotivated or unable to stop smoking. Evidence that cancer patients show an interest in smoking cessation and are able to stop would support clinical intervention for this patient group. Third, do nicotine dependence interventions promote smoking abstinence in cancer patients? The goal would be to determine the most effective methods of intervention. To address these questions, we will review the current literature on the natural history of smoking behavior change in cancer patients and evaluate nicotine dependence interventions within this patient population.

Diagnosis of Cancer and Smoking Behavior Change

Table 1 presents a review of studies that reported rates of smoking abstinence after a cancer diagnosis. Some of these studies were designed to observe the natural history of smoking behavior change in cancer patients25 or to evaluate factors related to smoking behavior,26 whereas other studies reported smoking status within the context of cancer treatment trials.21 Two studies assessed smoking status as a risk factor in cancer treatment outcome.12, 13 None of the studies included nicotine dependence intervention. Estimates of current smoking at the time of cancer diagnosis range from 46% to 75%. Across studies, 14–58% of patients who smoked at the time of diagnosis continued to smoke after cancer treatment. Estimates of spontaneous smoking cessation must be interpreted cautiously because most studies have several methodologic limitations. Specific cancer diagnosis, time of diagnosis in relation to the assessment of smoking status, and type of cancer treatment have not been specified routinely. As a result, it is unclear whether cancer site, diagnostic type or stage of cancer, time of diagnosis, or course of treatment may influence smoking abstinence rates.

Table 1. Smoking Behavior after Cancer Diagnosis
Author(s)Sample/cancer treatmentDesignOutcome/smoking statusComments
Silverman et al. (1993)12n = 160 patients with head and neck carcinoma (n = 117 current smokers at diagnosis, 43 were nontobacco users).Followed patients for 1–15 yrs posttreatment.58% of women and 52% of men continued to smoke.Patients who continued to smoke were more likely to develop second primary oral carcinomas than those who stopped smoking.
Stevens et al. (1983)13n = 200 patients with head and neck carcinoma (n = 154 current smokers at diagnosis, 32 were nonsmokers, and 14 were past smokers who stopped 1 yr before diagnosis).Retrospective analysis.47% of patients continued to smoke postdiagnosis.Patients who continued to smoke were more likely to have cancer recurrence than those who stopped smoking.
Ostroff et al. (1995)29n = 144 patients with head and neck carcinoma postsurgical treatment (n = 123 who reported any lifetime history of tobacco use, 74 reported using tobacco in the 12 mo before diagnosis).Questionnaire/chart review.35% of sample that reported tobacco use in 12 mo before diagnosis continued to smoke after surgery.Patients who continued to use tobacco were less likely to have received postoperative radiotherapy, to have had less extensive disease, to have had oral cavity disease, and to have had higher levels of education.
Vander Ark et al. (1997)26n = 87 patients with head and neck carcinoma (all had a history of smoking on a daily basis either before or at diagnosis).Interview/chart review.40% of patients stopped smoking at time of diagnosis; 19% stopped smoking at time of treatment; 16% stopped smoking after treatment.Physical barriers to continued smoking because of cancer treatment as well as counseling at diagnosis were effective deterrents to continued tobacco use.
Knudsen et al. (1985)27n = 57 nonsmall cell lung carcinoma patients (n = 16 current smokers, 41 former smokers).Retrospective patient survey.66% of the former smokers had stopped smoking at time of diagnosis or immediately after surgery, approximately 33% of those who stopped at diagnosis subsequently had disease recurrence.Patient suggestions for stopping smoking are listed as well as nursing implications.
Gritz et al. (1991)25n = 840 nonsmall cell lung carcinoma patients (n = 314 former smokers, 526 current smokers).Four-yr prospective study following diagnosis.Among 317 current smokers who were followed for 2 yrs, 17% continued smoking; 83% either stopped permanently or for some time period.Being female or healthier at baseline increased chances of ever quitting.
Sarna (1995)32n = 65 women with a recent diagnosis or recurrence of lung carcinoma (n = 9 never smokers, 51 former smokers, 5 current smokers).Postdiagnosis interview and survey.51% of former smokers stopped at diagnosis; 49% stopped before diagnosis.Lack of verification of smoking status Small sample size.
Dresler (1996)28n = 362 lung carcinoma patients (95% had history of smoking).Patients followed postoperatively for 3 mos for the first 2 years and every 6 mos thereafter; smoking questionnaires completed each time.42% stopped 1 yr before surgery, 6% stopped 3–12 mos before, 27% stopped 2 weeks to 3 mos before surgery, 19% smoked up to surgery, 85% reported nonsmoking postoperatively, 13% reported smoking postoperatively.Did not specify which factor (diagnosis surgery, advice) was responsible for cessation.
Cox et al. (2002)21n = 208 patients with Stage IIIA/IIIB nonsmall cell lung carcinoma (all had history of smoking).Patients were part of clinical trial.71% stopped smoking before study, 14% stopped during course of trial, 15% continued smoking throughout trial.Stopping smoking was associated with reduced symptoms of severe toxicity in response to cancer treatment.
Ostroff et al. (2000)n = 224 patients with bladder carcinoma (n = 84 smokers at time of diagnosis).Patients completed survey during follow-up care.69% of active smokers at diagnosis reported smoking postdiagnosis; 45% of active smokers reported smoking at time of assessment.The majority of patients who continued to smoke were interested in stopping ( 93% contemplating or preparing to stop smoking). Patients diagnosed at earlier stages more likely to continue smoking; patients diagnosed at later stage 2.8 times more likely to be continuous abstainers.
Schnoll et al. (2002)30n = 74 cancer patients (n = 51 lung, 23 head/neck).Patients completed survey during medical visit.39% current smokers at assessment.Patients who had completed cancer treatment were more likely to smoke than those in treatment; patients with a smoking family member were more likely to continue smoking; patients not ready to stop smoking reported lower self-efficacy and higher emotional distress.

These data demonstrate that although many smokers stop smoking after a diagnosis of cancer, a subset of patients continue to smoke. One group of head and neck carcinoma patients reported psychologic barriers to smoking cessation that included behavioral habit (67%), anxiety (50%), and stress (33%). Alcohol use was also associated with continued smoking in this group.26 Among patients who achieve initial abstinence, the risk of resuming smoking persists. Knudsen et al.27 found that approximately one-third of patients who stopped smoking at or after diagnosis subsequently relapsed. Similarly, Dresler et al.28 found that 13% of patients who were abstinent after a thoracotomy experienced subsequent relapse. It is noteworthy that even among those who continue to smoke, many remain motivated to stop or report an interest in intervention.29 Ostroff et al.29 found that the majority of patients undergoing surgical treatment for head and neck carcinoma reported an interest in tobacco use treatment and made at least one attempt to stop smoking postsurgery. Schnoll et al.30 reported that patients who were diagnosed with head, neck, or lung carcinoma and who were undergoing treatment were less likely to be smokers than those who had completed treatment. These findings support the need for nicotine dependence interventions that are sensitive to the needs of cancer patients to assist them in promoting initial abstinence and preventing smoking relapse.

Nonrandomized Interventions with Cancer Patients

Table 2 provides a summary of nonrandomized studies that evaluated nicotine dependence interventions among cancer patients. Wewers et al.31 evaluated the efficacy of a nurse-managed smoking cessation intervention with 15 hospitalized lung carcinoma patients. At 6 weeks postintervention, 14 (93%) of the subjects reported that they made at least one stop attempt and 6 (40%) were confirmed biochemically to be abstinent. Because the study did not include a control group, the efficacy of the intervention is difficult to interpret because of the rates of spontaneous abstinence from smoking reported among lung carcinoma patients.25, 27, 28, 32, 33 Moreover, long-term stop rates (e.g., 6 months) were not reported.

Table 2. NonRandomized Intervention Studies with Cancer Patients
Author(s)Sample/cancer treatmentDesignInterventionOutcome/smoking statusComments
  1. a AHCPR: Agency for Health Care Policy and Research; NRT: nicotine replacement therapy.

Wewers et al. (1997)31n = 15 hospitalized lung carcinoma patientsProspective, single group.Nurse-delivered counseling consisting of 3 25-minute daily sessions followed by five weekly calls postdischarge.93% reported making at least one stop attempt at 6 week follow-up, and 40% were abstinent. Continue confirmed abstinence.Small sample size. No control group. Brief follow-up.
Browning et al. (2000)34n = 25 patients diagnosed with nonsmall cell lung carcinoma (daily smokers for 1 yr or longer)QuasiexperimentalNurse-delivered counseling based on 1996 AHCPR guidlines (e.g., “4 A ‘s’), average seven sessions. Usual care (written materials and advice to quit).71% of intervention and 55% of usual care group abstinent at 6 mo postsurgical visit, n.s. Confirmed by expired carbon monoxide.Small sample size. No control group.
Cox et al. (2002)33n = 201 lung carcinoma patients treated for nicotine dependence and n = 201 nonlung carcinoma patientsRetrospective, date of treatment, matched case-control.Tailored clinical intervention including behavioral and pharmacologic treatment.22% of lung carcinoma and 14% of nonlung cancer patients were abstinent at 6-mo follow-up. Self-reported abstinence.No biochemical confirmation. Lung carcinoma patients reported more advanced stage readiness and greater motivation to stop smoking. Early intervention was associated with abstinence.
Smith et al. (2002)35n = 1077 hospitalized smokers. (All had smoked in the month before admission.)Prospective, single group.Nurse-delivered education and behavior modification counseling at bedside, which consisted of take-home materials (videotape, workbook, and relaxation audiotape), NRT if requested or indicated, and four postdischarge phone calls.Of the 509 patients who were enrolled for a year and were reached at 12-mo follow-up, 49% had been smoke free for the past 7 days. Patients hospitalized for cancer, cardiovascular, or pulmonary reasons were most likely to participate and had the highest self-reported cessation rates (63%, 57%, and 46%).No control group. No biochemical confirmation.

A quasiexperimental study of a nurse-delivered tobacco treatment intervention was conducted with 25 patients with nonsmall cell lung carcinoma.34 The usual care group (n = 11) was accrued and assessed at their 6-month visit, whereas the intervention group (n = 14) was accrued at the initial surgical consultation, before surgery. At the 6-month postsurgical visit, the intervention group (71% abstinence) was not significantly more likely to be abstinent relative to the usual care group (55% abstinence). Browning et al.34 concluded that patients with lung carcinoma are interested in quitting smoking at the time of diagnosis, as shown by their willingness to participate in the program. Interpretation of these findings is limited by the lack of randomization to treatment and by the small sample size. The usual care group accrued at 6-month follow-up is limited to smokers who did not stop smoking after the cancer diagnosis and may represent smokers with greater levels of nicotine dependence. Smith et al.35 reported the efficacy of a nurse case-managed smoking cessation program for hospitalized patients. Of 51 cancer patients who completed 12-month follow-up postdischarge, 43% self-reported smoking abstinence compared with 35% of the general patient group (n = 509). Although this study provides a comparison of tobacco use outcomes between patient groups, the absence of a control group limits the interpretation of these data.

Cox et al.33 conducted a retrospective study that evaluated nicotine dependence treatment outcomes for 201 lung carcinoma patients and 201 date-of-treatment matched patients without lung carcinoma. Compared with nonlung carcinoma patients, the lung carcinoma patients reported a higher motivation to stop smoking and were at a more advanced stage of readiness to change smoking behavior. The 6-month tobacco abstinence rate was 22% for the lung carcinoma patients compared with 14% of the control patients. Among lung carcinoma patients, early intervention after diagnosis predicted treatment success.

Randomized Behavioral Interventions with Cancer Patients

Four published studies evaluated randomized smoking interventions among cancer patients (Table 3). In the first study, Wewers et al.36 assessed short-term smoking abstinence rates among hospitalized postoperative patients following a structured, nurse-provided smoking cessation intervention. At 6 weeks after discharge, 37.8% of the experimental and 25.6% of the usual care patients were confirmed biochemically to be abstinent. Experimental cardiovascular and oncology patients had abstinence rates of 40% and 64.3%, respectively, compared with a rate of only 7.7% among general surgery patients. The authors concluded that smokers who are hospitalized with a smoking-related diagnosis may be especially amenable to postoperative intervention.

Table 3. Randomized Intervention Studies with Cancer Patients
Author(s)Sample/cancer treatmentDesignInterventionOutcome/smoking statusComments
  1. a n.s.: not significant

Wewers et al. (1994)36n = 80 smokers hospitalized for surgery (continuous smoking for at least 1 yr; n = 30 with cancer; 25 head/neck, 2 breast, 2 prostate, 1 cervical).Randomized controlled trial.Nurse-delivered counseling consisting of 3 consecutive 20–30-minute sessions followed by five weekly calls postdischarge. Usual care (written materials).64% of cancer patients abstinent at 6-week follow-up, n.s. Continue confirmed.Small sample size. Brief follow-up.
Stanislaw and Wewers (1994)37n = 26 hospitalized cancer patients who smoked (continuous smoking for at least 1 yr); n = 21 head/neck, 2 prostate, 2 breast, 1 cervical).Randomized controlled trial.Nurse-delivered counseling consisting of three consecutive 20–30-minute sessions followed by five weekly calls postdischarge. Usual care (written materials).75% of intervention and 43% of usual care group abstinent at 5-week follow-up, n.s. Cotinine confirmed.Small sample size. Brief follow-up.
Griebel et al. (1998)38n = 28 hospitalized cancer patients who smoked (continuous smoking for at least 1 yr; n = 6 gynecologic, 3 breast, 4 gastrointestinal, 4 thoracic, 4 urologic, 3 neurologic, 2 head/neck).Randomized controlled trial.Nurse-delivered, one-time counseling, plus five weekly calls. Usual care (written materials).21% of intervention and 14% of usual care group abstinent at 6-week follow-up, n.s.Small sample size. Brief follow-up.
Gritz et al. (1993)39n = 186 hospitalized cancer patients who smoked within last year (96 current smokers at randomization, all head/neck cancer). Study completers = 114. Study noncompleters = 72.Randomized controlled trial.Provider-delivered counseling postsurgery followed by six monthly booster sessions posttreatment. Usual care (advice to quit).74% of intervention and 77% of usual care group abstinent at 12 mo, n.s. Continine confirmed.Usual care group included many elements of intervention group.

In the second study, Stanislaw and Wewers37 examined smoking behavior change in 26 hospitalized surgical cancer patients who were assigned randomly to either a nurse-delivered tobacco use treatment intervention or to usual care. At 5 weeks after discharge, 75% of patients who received the intervention were confirmed biochemically to be abstinent compared with 43% of those who received usual care. In the third study, Griebel et al.38 determined the effect of a one-time, inpatient, nurse-provided smoking cessation intervention on hospitalized surgical cancer patients. At 6 weeks postintervention, 21% of the intervention group and 14% of the usual care group were confirmed biochemically to be abstinent. In the fourth study, Gritz et al.39 conducted a study with 186 patients with head and neck carcinoma, all of whom had smoked cigarettes within the last year and 88.2% of whom were current smokers at the time of randomization. Participants were assigned randomly to either the intervention or usual care groups. Findings indicated that at 12 months, 60% of baseline smokers were continuously abstinent as determined by urine cotinine levels.

None of these trials produced significant findings, although their results suggest that smoking cessation interventions can be more effective for oncology patients than usual care advice to stop. Three of the studies consistently found higher abstinence rates in patients receiving the nicotine dependence intervention,36–38 yet the power of the findings may have been limited by small samples of patients and brief follow-up periods. Future studies of smoking cessation interventions with a larger sample of patients and long-term (6-month) follow-up are needed to determine treatment efficacy for cancer patients. Although the Gritz et al. study39 had a larger sample of patients and a longer follow-up period, the authors noted that contamination in the control condition may have been a limitation of the study. Specifically, physician advice was used with the control group and delivered at a time (diagnosis and treatment) when advice could have the greatest impact on smoking behavior change. Although physician advice to stop smoking is used commonly as a usual care control condition, such advice may be regarded as an intervention rather than as a standard of oncology treatment.

Current research has not established which mode of intervention is most effective in helping cancer patients to stop smoking. However, nicotine dependence treatment is needed in this group because a significant subset of patients continue to smoke despite a diagnosis of cancer. The findings show that cancer patients are motivated and are able to stop smoking. The most evident challenge to researchers and clinicians is to develop effective methods of intervention within this patient population.


The efficacy of tobacco use treatment interventions may be affected by the individual characteristics and circumstances of the cancer patient. This section reviews reports that address disease stage and cancer treatment, patient readiness to stop smoking, and timing of nicotine dependence intervention with medical patients.

Disease Stage and Cancer Treatment

It is unclear to what degree disease stage may influence smoking behavior change. Some findings suggest the more severe the disease or invasive the treatment, the higher the cessation rate among cancer patient populations.29, 39 For example, Gritz et al.39 found that patients with head and neck carcinoma who were treated with laryngectomy or other surgery had approximately double the continuous abstinence rates of patients who were treated with radiation only. They postulated that the invasiveness of the treatment affected the likelihood of smoking cessation. Ostroff et al.29 reported similar findings with a study of 144 head and neck carcinoma patients who underwent surgical treatment. In the postsurgical follow-up period, 72% of patients with more extensive disease (Stage II–IV) remained abstinent, compared with 24% of patients with in situ or Stage I disease. The authors suggested that patients with earlier stage disease who undergo less extensive cancer treatment may be more likely to minimize the seriousness of their illness and thereby be less motivated to stop smoking. Alternatively, patients with later stage disease who experience more extensive cancer treatment have greater disability and functional impairment, which may interfere with continued tobacco use.29 In contrast to the above findings, Cox et al.33 examined 201 lung carcinoma patients who received individualized clinical treatment of nicotine dependence. They found no relationship between grade, stage, or histology of lung carcinoma and smoking treatment outcome.

Multiple factors must be considered when evaluating differences in smoking outcomes. For example, findings may be influenced by differences in cancer treatment and disease-related symptoms for head and neck carcinoma patients compared with lung carcinoma patients. Differences in study designs also may account for varied findings. Observational studies may demonstrate an influence of disease-related factors on smoking behavior change in the absence of formal intervention.29 In contrast, an individualized nicotine dependence intervention given universally to cancer patients may enhance smoking cessation across all patients.33 Other factors (e.g., smoking abstinence during hospitalization, length of hospitalization, and frequency of follow-up with healthcare providers) should not be overlooked, as these factors may enhance initial abstinence and relapse prevention. Overall, more research is needed to clarify to what degree smoking behavior change may be related to type of cancer, stage of disease, and treatment invasiveness.

Patient Motivation and Readiness to Stop Smoking

The transtheoretic model conceptualizes behavior change as a process rather than as a discrete event.40 The model describes five stages of readiness to change smoking behavior. During precontemplation, a smoker is not considering smoking cessation within the next 6 months. In the contemplation stage, a smoker is seriously considering smoking cessation within the next 6 months. During preparation, the smoker is seriously considering smoking cessation within the next 30 days and has taken steps to prepare to stop smoking. In the action stage, the smoker has just stopped smoking. Finally, during maintenance, a smoker has achieved abstinence for more than 6 months and is developing skills to prevent relapse. Within this model, the goal of intervention is to facilitate advancement in the stage of readiness with the ultimate goal of maintaining abstinence. Stage-matched interventions may be most effective in advancing readiness to stop smoking.41, 42

Gritz et al.39 found that readiness to stop smoking among newly diagnosed head and neck carcinoma patients at baseline was a significant predictor of continuous abstinence 1 year after smoking cessation treatment. They also observed that almost one-half of the precontemplators became continuous abstainers, indicating a “readiness” phenomenon. Cox et al.33 reported similar findings when examining stage of change among lung carcinoma patients. Lung carcinoma patients who reported a higher stage of change were more likely to be abstinent from tobacco at 6 months (P = 0.012). The tobacco abstinence rate was 7.7% for patients in the precontemplation or contemplation stage, 22.2% for patients in the preparation stage, and 28.1% for patients in the action stage. Future research may evaluate the efficacy of stage-matched interventions for cancer patients. Interventions that follow patients over time may identify the need for continued intervention to match the potentially changing need of the patient to maximize long-term abstinence success.

Timing of Nicotine Dependence Intervention

Periods surrounding significant medical events, such as receiving a diagnosis of cancer, may serve as a “teachable moment” for smoking cessation. During this period, a person is prepared for or open to learning new information or to changing a behavior.43 Cancer patients may have heightened motivation or openness to participate in nicotine dependence interventions at one or more points along the diagnostic/treatment continuum from prediagnostic testing, through oncology treatment, and beyond recovery. However, the optimal timing for intervention remains unknown. Given the limited literature on smoking treatment interventions with cancer patients, we have included research conducted with other medical populations to examine the question of intervention timing.

Before the diagnosis

Wewers et al.31 reported a 40% abstinence rate 6 weeks after a brief intervention with smokers who had a suspected diagnosis of lung carcinoma and were admitted for diagnostic testing (Table 2). These data support using the diagnostic work-up as a teachable moment for promoting smoking abstinence. A similar opportunity exists with persons participating in cancer screening programs44–46 although there have been modest47 or no effects43 of cessation interventions conducted with persons attending noncancer-related screenings.

Preoperative intervention

Evidence from observational studies demonstrates the value of conducting preoperative nicotine dependence interventions with surgery patients.48 Relative to nonsmokers, current smokers are at increased risk for complications during and after surgery,49 especially pulmonary complications,50, 51 and for slower postoperative wound healing.52 It is possible that many patients who are scheduled for surgery are unaware of the negative impact of smoking on surgical recovery. The development of interventions that capitalize on providing this information to increase a person's motivation to stop smoking may result in increased attempts at smoking cessation. Provision of pharmacotherapy to reduce nicotine withdrawal and support abstinence after surgery is warranted.

During hospitalization

In a recent review of smoking cessation interventions conducted with hospitalized patients, Rigotti et al.53 evaluated 15 randomized smoking cessation trials. They concluded that interventions were effective only when the outpatient follow-up period lasted for at least 1 month. Rice and Stead47 conducted a metaanalysis of nurse-based smoking cessation interventions and found a significant effect of inpatient interventions for cardiovascular patients compared with patients with nontobacco-related illnesses. Similarly, Smith et al.35 found that patients hospitalized for cancer, cardiovascular, or pulmonary reasons were more likely to participate in a nurse-delivered smoking cessation program and had higher self-reported abstinence rates compared with other medical patients. Healthcare providers may have a window of opportunity with hospitalized smokers, and this opportunity may be maximized when extended follow-up is used to support long-term abstinence.

Outpatient intervention

Gritz et al.39 conducted a large randomized trial with outpatient head and neck carcinoma patients (Table 3), but did not find a significant intervention effect on smoking abstinence. It is possible that physician advice to stop smoking may be particularly salient to cancer patients and may have produced higher abstinence rates within the control group that received physician advice without adjunct intervention. In a metaanalysis of physician-based interventions, interventions were more effective with outpatients at high risk for tobacco-related diseases, compared with interventions conducted with unselected populations.54 Cox et al.33 found that lung carcinoma patients had a greater motivation to stop smoking and higher abstinence rates than general medical patients.In addition, lung carcinoma patients who received a nicotine dependence intervention within 3 months of their diagnosis were more likely to be abstinent at the 6 month follow-up, compared with those who received the intervention more than 3 months postdiagnosis. Although some healthcare providers may be concerned about the addition of a nicotine dependence intervention immediately after diagnosis, these findings support early intervention.

Intervention after oncology treatment

Children and adults who have been treated successfully for cancer remain at increased risk for developing secondary malignancies and other health complications. Because these conditions may be exacerbated by tobacco use, prevention and intervention efforts are high priorities among cancer survivors. Tobacco research in this population is still in an early phase of development. Findings show that although adult survivors of childhood cancer may be less likely to smoke than their siblings, those who do smoke are not more likely to stop smoking compared with smokers with no history of cancer.55, 56 Demark-Wahnefried et al.57 surveyed almost 1000 adult survivors of prostate and breast carcinomas. Approximately half of the participants had ever smoked and greater than half of current smokers were either thinking of stopping smoking or were actively preparing to stop. This study found a high rate of interest in participating in health promotion programming (80%), suggesting that substantial opportunities exist to effect change in smoking behavior among cancer survivors.

Teenage cancer survivors are as likely to use tobacco as their peers who have never been treated for cancer.58 Among the factors that have been identified as predictors of pediatric cancer survivors' intentions to use tobacco are older age and limited knowledge of the adverse health consequences associated with tobacco use.59 Other factors that may contribute to smoking among adolescent cancer survivors include impaired cognitive functioning that may be a side effect of their cancer treatment, low resiliency, behavioral modeling by parents and peers, and poor decision-making capacity.60 A comprehensive approach that includes a combination of prevention and intervention strategies used simultaneously may hold the greatest promise to reduce tobacco use in young survivors.61


The evidence suggests that there may be multiple teachable moments and supports active intervention with patients who continue to smoke. However, additional research is needed to evaluate the feasibility and efficacy of interventions within the cancer diagnostic/treatment continuum. Individual differences associated with treatment outcome must be evaluated. It is possible that multiple efforts may be needed to educate cancer patients and survivors about the risks of smoking and the benefits of cessation, to motivate and assist smokers in making an attempt to stop smoking, and to support continued abstinence during and after oncology treatment.


The current findings provide an ethical imperative to all healthcare providers to address tobacco use with cancer patients and to provide assistance to facilitate tobacco abstinence. The involvement and support of the entire oncology treatment team may facilitate a successful intervention.15, 16, 62 Current recommendations for treatment of tobacco dependence have been reported by the Agency for Health Care Policy and Research.63 Additional resources on providing nicotine dependence intervention to cancer patients also are available.42, 64 An initial directive for healthcare providers is to ensure a mechanism for identifying all patients who use tobacco (e.g., including tobacco status as a vital sign) so that at least brief interventions can be delivered to all tobacco users or to those who are in the early period of abstinence.63

Interventions as brief as 3 minutes can have a significant impact on promoting smoking abstinence.63 For patients who have stopped smoking recently, providers should reinforce their decision and assist them in resolving any residual problems (e.g., depressed mood and weight gain that accompany smoking cessation), using prescriptive interventions as needed. For patients who are not ready to make a stop attempt, a brief motivational intervention can help them to identify the personal relevance of tobacco use, the risks of continued use, and the rewards of stopping. In addition, patients can be educated that effective treatment options exist and the clinician is available to assist future stop attempts.

The guidelines urge clinicians to help patients set a stop date and develop a cessation strategy, to encourage them to obtain additional support, and to recommend pharmacotherapy, except for those with medical contraindications. In addition, the guidelines provide specific instructions for clinical use of pharmacotherapies, as well as specific brief counseling strategies.63 First-line pharmacotherapies include nicotine replacement therapy (NRT; gum, patch, nasal spray, inhaler) and sustained release bupropion. The extended use of these medications does not present a known health risk and may facilitate long-term abstinence. Second-line pharmacotherapies to consider are clonidine and nortriptyline.63 There is also evidence that combined NRT (e.g., combination patch and gum,63 high-dose NRT,65 or combined bupropion and NRT66) increases long-term abstinence rates. Therefore, these medication options should be considered carefully. Because patients who continue to smoke despite medical consequences demonstrate characteristics of strong nicotine dependence,67 pharmacotherapy may be especially important for successful treatment of patients with cancer.

There is a dose-response relationship between nicotine dependence treatment intensity and long-term abstinence from tobacco.63 Intensive interventions produce higher success rates and are more cost-effective than less intensive interventions. In addition to pharmacotherapy, intensive interventions may combine multiple (four or more), individual, or group sessions (30 minutes or more of total contact), telephone counseling, adjunct self-help materials, and follow-up assessment.63 Practical skills training, problem-solving, and support-building should be emphasized. Intensive interventions are appropriate for any tobacco user willing to participate. However, intensive interventions must be considered for cancer patients, given the critical importance of stopping tobacco use within this group.


Although the efficacy of pharmacotherapy (e.g., NRT and bupropion) has been well established in nicotine dependence treatment,66, 68–70 no published studies have examined specifically the efficacy of pharmacotherapy in clinical populations of cancer patients. Therefore, future studies could investigate the effectiveness of various forms of NRT (e.g., gum, transdermal patch, nasal spray, inhaler), other nonnicotine pharmacotherapeutic agents (e.g., bupropion, clonidine, nortriptyline), and combination therapies in clinical populations of cancer patients.

Research is also needed to evaluate the efficacy of pharmacotherapy in symptom management. Withdrawal from nicotine may produce physiologic, affective, and behavioral changes including irritability, anxiety, depression, impatience, restlessness, difficulty concentrating, nocturnal awakening, decreased heart rate, and hunger. These symptoms may exacerbate the physical side effects of cancer treatment. Because withdrawal is due to nicotine deprivation, nicotine replacement can be used in treatment to effectively decrease most symptoms of nicotine withdrawal.69, 71, 72 Smoking abstinence also may elicit or exacerbate symptoms of depression, particularly in patients with a history of depression.73–75 The U.S. Public Health Service guidelines recommend bupropion SR or nortriptyline for the treatment of nicotine dependence in smokers with current or past history of depression, as both drugs are effective treatments for smoking cessation and depression in the general population.63 Finally, a subset of cancer patients may experience anxiety after cessation.22 Symptoms of nicotine withdrawal, depression, or anxiety may contribute to smoking relapse, tax a patient's physical, cognitive, and emotional resources, and subsequently impede overall oncology treatment. Therefore, careful assessment and clinical management of these symptoms during nicotine dependence treatment are warranted. Further research is needed to evaluate the use of pharmacotherapy among cancer patients to manage effectively adverse symptoms and to examine the impact of symptom management on facilitating tobacco abstinence.

Future research should address multiple methodologic concerns identified in previous research, examine factors predictive of relapse and effective strategies for relapse prevention, and assess the impact of smoking cessation on cancer treatment and quality of life. Given the prominent role of spit tobacco in the development of oral carcinoma,76, 77 research should be extended to evaluate the use of spit tobacco and other forms of tobacco. Finally, attention should be given to smoking behavior within the families of cancer patients. Smoking abstinence among family members supports abstinence in patients who are trying to stop smoking.27, 78, 79 Family members of patients with tobacco-related cancer may be more likely to smoke than the general population80 and may be at greater risk of developing cancer.81 Recommendations for future research are outlined in Table 4. The ultimate objectives of this area of research are to advance the treatment of cancer patients, reduce tobacco use, and subsequently improve health outcomes and quality of life.

Table 4. Recommendations for Future Research with Cancer Patients
Area of researchRecommendation
  1. a NRT: nicotine replacement therapy.

Pharmacotherapy for nicotine dependence treatment.· Evaluate the efficacy of NRT (gum, patch, nasal spray, inhaler), nonnicotine medications (bupropion, clonidine, nortriptyline), and combination therapies (e.g., high-dose NRT; combined NRT and bupropion) in increasing tobacco abstinence rates in oncology patients.
 · Assess efficacy of nicotine and nonnicotine medications in managing symptoms of nicotine withdrawal, craving, stress, and depression among cancer patients.
Methodologic issues· Obtain sample sizes needed to provide adequate statistical power for hypothesis testing.
 · Collect and report demographic and medical data including disease site, type, stage, and treatment.
 · Report timing of intervention relative to time of cancer diagnosis and treatment.
 · Include at least 6-mo outcomes to assess stability of treatment effects.
 · Assess tobacco use status using both 7-day point prevalence and continuous abstinence.82
Prevention of disease recurrence· Examine factors predictive of smoking lapse and relapse in cancer patients.
 · Evaluate relapse prevention strategies within nicotine dependence treatment interventions: cognitive-behavioral strategies, longterm use of pharmacotherapy, support mechanisms, and extended follow-up may increase abstinence maintenance.82–85
Cancer treatment and quality of life· Further determine the type and extent of physical and psychological benefits experienced by cancer patients who stop smoking. This research is needed across cancer sites, as benefits may differ.
 · Clinical trials should include primary outcomes related to quality of life and oncology treatment.
 · Quality of life assessment may include global measures86–89 and site-specific measures.90
Other tobacco use· Assess change in use of spit tobacco (e.g., chewing tobacco, snuff), pipes; cigars, or other forms of tobacco after a cancer diagnosis.
 · Examine nicotine dependence treatment for spit tobacco or other tobacco use in cancer patients.
Family smoking behavior· Examine how family members may most effectively support tobacco abstinence in the cancer patient.
 · Assess how the diagnosis of cancer affects smoking behavior among patient's family members.
 · Evaluate methods of extending nicotine dependence interventions to family members who use tobacco.


The health risks of continued smoking and the benefits of smoking cessation provide an ethical imperative for treating nicotine dependence in cancer patients. Evidence that cancer patients are motivated and able to stop smoking supports the use of tobacco use treatment interventions within this patient population. Resources for educating healthcare professionals in providing nicotine dependence intervention are available.16, 42, 63, 64 Research is needed to evaluate further the needs of cancer patients and to provide effective methods and means of intervention to this patient group. The overall goal is to promote tobacco abstinence to benefit oncology treatment outcome, patient health, and quality of life.