Tobacco use outcomes among patients with head and neck carcinoma treated for nicotine dependence

A matched-pair analysis

Authors

  • Yolanda I. Garces M.D.,

    Corresponding author
    1. Department of Oncology, Division of Radiation Oncology, Mayo Clinic, Rochester, Minnesota
    • Division of Radiation Oncology, Mayo Clinic, Charlton Building, Desk R, 200 First Street SW, Rochester, MN 55905
    Search for more papers by this author
    • Fax: (507) 284-0079

  • Darrell R. Schroeder M.S.,

    1. Department of Health Sciences Research, Division of Biostatistics, Mayo Clinic, Rochester, Minnesota
    Search for more papers by this author
  • Liza M. Nirelli B.S.,

    1. Department of Health Sciences Research, Division of Biostatistics, Mayo Clinic, Rochester, Minnesota
    Search for more papers by this author
  • Gary A. Croghan M.D., Ph.D.,

    1. Department of Oncology, Division of Medical Oncology, Mayo Clinic, Rochester, Minnesota
    2. Department of Internal Medicine, Nicotine Research Center, Mayo Clinic, Rochester, Minnesota
    Search for more papers by this author
  • Ivana T. Croghan Ph.D.,

    1. Department of Internal Medicine, Nicotine Research Center, Mayo Clinic, Rochester, Minnesota
    Search for more papers by this author
  • Robert L. Foote M.D.,

    1. Department of Oncology, Division of Radiation Oncology, Mayo Clinic, Rochester, Minnesota
    Search for more papers by this author
  • Richard D. Hurt M.D.

    1. Department of Internal Medicine, Nicotine Research Center, Mayo Clinic, Rochester, Minnesota
    2. Department of Internal Medicine, Division of Community Internal Medicine, Mayo Clinic, Rochester, Minnesota
    Search for more papers by this author

  • Presented at the 4th European Conference of the Society for Research on Nicotine and Tobacco, Santander, Spain, October 3–5, 2002, and at the 39th Annual Meeting of the American Society of Clinical Oncology, Chicago, Illinois, May 31–June 3, 2003.

Abstract

BACKGROUND

The current study described tobacco use outcomes among patients with head and neck carcinoma who underwent treatment for nicotine dependence at the Mayo Clinic Nicotine Dependence Center (NDC; Rochester, MN).

METHODS

Using a 1:1 matched-pair design, conditional logistic regression was employed to compare the 6-month tobacco abstinence outcomes of patients with head and neck carcinoma (n = 101) with controls (n = 101) from the general patient population treated for nicotine dependence between 1988 and 2001. The two groups were matched with regard to age, gender, date of treatment, and type of NDC treatment service.

RESULTS

Baseline demographics were similar between both groups. However, patients with head and neck carcinoma smoked significantly more cigarettes per day (cpd) than controls (P = 0.003). The self-reported tobacco abstinence rate at the 6-month follow-up was 33% for patients with head and neck carcinoma compared with 26% for matched controls (P = 0.279; after adjusting for baseline cpd and stage of change, P = 0.205). Among patients with head and neck carcinoma, the tobacco abstinence rates were 47%, 22%, and 19%, respectively, for those receiving an NDC consult within 3 months, between 3 months and 5 years, and > 5 years after their diagnosis (P = 0.021). Furthermore, the patients with head and neck carcinoma treated within 3 months of diagnosis who received surgery (with or without radiation therapy) were more likely to be tobacco abstinent than those who received primary radiation therapy (P = 0.042).

CONCLUSIONS

These findings suggested that nicotine dependence treatments were effective among patients with head and neck carcinoma, particularly when delivered shortly after initial diagnosis and for those who received surgery as their primary treatment. Cancer 2004. © 2004 American Cancer Society.

Overall, 30% of malignancies are due to smoking.1 In 2003, approximately 37,000 new cases of head and neck carcinoma (carcinomas of the larynx, pharynx, mouth, tongue, and other oral cavity) were diagnosed in the U.S. In 2004, 11,000 persons are expected to die of head and neck carcinoma.2 The majority of these malignancies are due to smoking and alcohol.3 Among survivors of head and neck carcinoma, approximately 20–50% will continue to smoke after they have completed treatment.4, 5 Continued smoking during or after cancer treatments can lead to increased morbidity and mortality,6 as well as increased rates of second primary tumors,7 among patients with head and neck carcinoma. More recently, current cigarette smoking while receiving radiation therapy and/or chemotherapy also has been found to be detrimental to survival8, 9 and to increase toxicity related to treatment10 for patients with lung, breast, and gynecologic carcinomas. Furthermore, patients with cancer who quit smoking could have all of the health benefits observed in the general population (i.e., a longer life span, improved lung function, and a reduced risk of heart disease). Therefore, effective interventions for patients with head and neck carcinoma who continue to smoke are needed.11, 12

Because tobacco use is a chronic disorder, effective interventions should combine behavioral therapy with pharmacologic therapy, when medically indicated, as has been demonstrated in the general population of tobacco users.13 Tobacco use interventions have received limited empirical investigation among oncology patients as a subgroup. Some physicians believe that a patient should not be overwhelmed with additional concerns at the time of a cancer diagnosis or that it is simply “too late.”14 Smokers who quit at age 65 years will gain an additional 1.4–3.7 years of life whereas younger smokers will achieve even greater gains.15 Intuitively, it could be speculated that patients with cancer would have similar benefits, but this remains to be shown. To our knowledge, only a few studies have been performed specifically to aid patients with cancer in their efforts to become tobacco free.

Tobacco intervention studies for patients with cancer have typically included a variety of cancer types,16–18 and only one of these studies specifically addressed patients with head and neck carcinoma.4 Some studies utilized a nurse-managed intervention during hospitalization,19–21 whereas other studies utilized a physician-based intervention with self-help guides from the National Institutes of Health or the American Cancer Society.4, 18 The results of these interventions have had extremely variable tobacco abstinence outcomes, ranging from 13% to 70% of treated smokers demonstrating continuous tobacco abstinence at a 1-year follow-up.18 Some of the limitations of these studies included limited power (due to small sample sizes; N ≤ 30) and short-term tobacco cessation rates (≤ 6 weeks).20–22 Furthermore, the majority of these studies did not report the type of pharmacotherapy utilized or the patient's compliance with the treatment. Future studies will likely need to incorporate pharmacotherapy, a theoretical basis for behavior change, and continued follow-up to prevent relapse during or shortly after completing cancer therapies to assist patients with cancer with nicotine dependence.17, 18

The primary goal of our retrospective study was to examine tobacco use history and tobacco abstinence outcomes among patients with head and neck carcinoma treated for nicotine dependence compared with the general patient population of treated tobacco users by means of a matched-pair control group. All patients with head and neck carcinoma and controls received a consultation for nicotine dependence at the Mayo Clinic Nicotine Dependence Center (NDC; Rochester, MN). We hypothesized that patients with a diagnosis of head and neck carcinoma who were treated at the NDC would be more motivated to stop using tobacco than the general population of tobacco users and, as a result, would have a higher rate of tobacco abstinence at 6 months. The patients with head and neck carcinoma would be more motivated because they would be treated during a “teachable moment” at the time they received their diagnosis.23 Twenty-seven of the patients with head and neck carcinoma were treated at the NDC > 5 years after their head and neck carcinoma diagnosis, which did limit the strength of our hypothesis. However, despite this limitation, these results will add to our knowledge concerning interventions for tobacco use and may facilitate the future development of tailored interventions among patients with head and neck carcinoma.

MATERIALS AND METHODS

Study Design

The current study was designed as a retrospective matched-pair study. The patients with head and neck carcinoma were matched to the controls by gender, age, and date and type of NDC treatment program. The primary aim was to compare the rates of 7-day point prevalence of self-reported tobacco abstinence at 6 months between the patients with head and neck carcinoma and the controls, all of whom had received an NDC consultation between 1988 and 2001. The Mayo Foundation institutional review board approved the current study. For all cases, subjects were excluded from consideration if they had declined use of their medical records for research purposes.

Patients

Potential patients for the current study were identified using a cross referencing procedure conducted among the general Mayo Clinic medical index, tumor registry, and the NDC database. The cases were comprised of all patients with head and neck carcinoma (including carcinomas of the larynx, tongue, pharynx, oral cavity, and parotid glands) who received an initial consultation for treatment of nicotine dependence between April 1988 and June 2001 (n = 101). Only patients with head and neck carcinoma who received their diagnosis before (or within 3 weeks after) their NDC consultation were included.

Once the patients with head and neck carcinoma were identified, the controls were selected from a pool of approximately 20,000 patients who had been treated during the same time period. The controls were matched to the patients with head and neck carcinoma by gender, age (± 5 years), date of treatment at NDC (± 6 months), and type of tobacco treatment program including type of tobacco product. Among the potential matches for a given patient with head and neck carcinoma, the control patient with the date of treatment closest to that of the patient with head and neck carcinoma was selected as the matched control. The date-of-treatment matched control method of selection accounts for secular changes in pharmacotherapy options available over the 13-year period of clinical care, as well as for other temporal changes in the approach and clinical experience of the counselor staff.

Treatment Model

The treatment model utilized by the NDC is a comprehensive intervention that combines behavioral, addiction, and pharmacologic approaches, as well as a structured relapse prevention program. The intervention used in the NDC residential and nonresidential consultations involves a 45–60-minute consultation with a nicotine dependence counselor. The counselor assesses the patient's degree of nicotine dependence and stage of readiness to change.24 An individualized treatment plan, tailored to the patient's specific needs, is then developed. The NDC residential program is an 8-day inpatient group treatment program that provides additional didactics, individual and group therapy, exercise, and tailored pharmacotherapy.25 The nonresidential program is delivered in either the inpatient or outpatient setting depending on where the patient is located at the time the consultation is requested. Pharmacotherapy options for both the residential and nonresidential consultations include nicotine replacement (e.g., nicotine patch, gum, inhaler, lozenges, and nasal spray) and/or bupropion SR. These medications are often used in combination. A structured relapse prevention program based on telephone follow-up (at 1 month, 3 months, and 6 months), patient contact via letter (at 1 year), and a tobacco telephone counseling service are available during follow-up. As part of the routine scheduled follow-up, information on the patient's 7-day point prevalence of self-reported tobacco use status 6 months after the nicotine dependence consultation is collected by telephone by a trained interviewer. Because these data are drawn from a clinical service provided to patients, biochemical confirmation of tobacco abstinence is not obtained routinely in this clinical setting. A few patients falsely report smoking status especially among adults who receive an interviewer-administered questionnaire.26

Measures

Baseline demographics and tobacco use history

Data for the current study were gathered from an existing NDC database of patients treated by the NDC staff and a medical record review. All patients treated by the NDC staff completed a baseline questionnaire. This questionnaire contains information regarding the patients' tobacco use, education level, age, gender, marital status, and alcohol and drug use.

Medical record review

Medical records were reviewed by a physician abstractor (Y.I.G.) for information on patients' baseline characteristics that were missing from the NDC database. In addition, the medical records were reviewed for medical information that is not included in the NDC database. This information included the patients' medical and psychiatric diagnoses that may be associated with tobacco use. For the purpose of the current investigation, information specific to patients with head and neck carcinoma was also collected, including the type and stage of the patients' initial head and neck carcinoma,27 the date of diagnosis, and the type of cancer treatment received (e.g., surgery, radiation therapy, and/or chemotherapy).

Tobacco use status

The World Health Organization definitions were utilized to define a tobacco user and an ex-tobacco user.28 Furthermore, the patient's stage of readiness to change was also determined by the counselors.24 Patients in the precontemplation (not seriously thinking of quitting in the next 6 months), contemplation (seriously thinking of quitting in the next 6 months), preparation (seriously thinking of quitting in the next 30 days and a 24-hour quit attempt made in the last year), and action (tobacco abstinent at the time of the consultation or within the last 6 months) stage of change were all included. Those in maintenance (abstinent from tobacco for > 6 months) were not included in the study as they were not referred to the NDC. Patients were asked to self-report tobacco (e.g., cigarettes, cigars, pipes, snuff, and/or spit tobacco) use including a single puff or the use of any smokeless tobacco product in the 7 days before their telephone interview at the 6-month follow-up. This definition of abstinence is consistent with a recent consensus statement from the Society for Research on Nicotine and Tobacco work group on treatment outcome methodology that recommended as one option using a 7-day point prevalence estimate of tobacco abstinence.29 If tobacco use outcome at 6 months was missing from the NDC database for reasons including refusal to respond or inability to be contacted, and this information could not be ascertained from the patient's medical record, we classified the patient as a tobacco user at that time point using an intent-to-treat analysis. Those who died before the gathering of the 6-month follow-up data were also classified as using tobacco.

Statistical Analysis

Baseline characteristics, tobacco use history, and medical diagnoses of the patients with head and neck carcinoma and controls were compared using the rank sum test for continuous variables and the chi-square test for categoric variables. Tobacco abstinence rates 6 months after the nicotine dependence consultation were summarized and compared between patients with head and neck carcinoma and controls using conditional logistic regression and taking into account the 1:1 matched-pair study design.30 Conditional logistic regression was also utilized to assess whether the likelihood of abstinence after nicotine dependence treatment differed between patients with head and neck carcinoma and controls after adjusting for previously determined covariates predictive of tobacco abstinence, stage of change, and mean cigarettes per day (cpd).31, 32 Age and gender were not included as covariates in the model because these variables were included in the matched-pair design. For the Head and neck carcinoma patients, logistic regression was used to assess whether 6-month tobacco abstinence outcome was associated with cancer stage at diagnosis, treatment, or the length of time from the head and neck carcinoma diagnosis to the NDC consult (categorized as < 3 months, 3 months–5 years, and > 5 years). For all cases, two-sided tests were performed with P ≤ 0.05 used to denote statistical significance.

RESULTS

Patients

Baseline characteristics at the time of the NDC consult for the patients with head and neck carcinoma and their matched controls are summarized in Table 1. The 101 patients with head and neck carcinoma (66 males and 35 females) in the current study had a mean age ± standard deviation at the time of the NDC consult of 58.7 ± 10.1 years (range, 36–79 years). Nearly all (100 of 101) of the patients with head and neck carcinoma were treated for cigarette smoking (1 spit tobacco user) and only 2 patients were treated in the residential nicotine dependence treatment program (1 cigarette smoker and 1 spit tobacco user). All patients with head and neck carcinomawere treated at the NDC after receiving their initial diagnosis. The median time from the initial head and neck carcinoma diagnosis to the NDC consult was 7.4 months. Forty-seven patients received an NDC consult within 3 months of their initial head and neck carcinoma diagnosis (median, 8 days; range, 15–72 days), 27 patients received an NDC consult 3 months–5 years after their initial diagnosis (median, 1.6 years; range, 100 days–3.5 years), and 27 patients received an NDC consult > 5 years after their initial diagnosis (median, 11.2 years; range, 5.0–24.3 years). Eight patients were treated at the NDC before their tissue diagnosis (days −15 to −1), 2 patients were treated 1 day before tissue diagnosis, and 1 patient each was treated 2 days, 3 days, 4 days, 9 days, 13 days, and 15 days before tissue diagnosis. We chose not to exclude these patients from our study population because these patients were in the process of being evaluated and staged. When diagnosed, most of the patients with head and neck carcinoma had early-stage disease (58% had AJCC Stage 0, Stage I, or Stage II disease) and 91% had squamous cell carcinomas. The majority of patients had carcinomas of the oral cavity (e.g., carcinomas of the oral tongue, retromolar trigone, alveolar ridge, buccal mucosa, floor of mouth, hard palate, or lip) and laryngeal carcinoma (e.g., true vocal cords or supraglottic larynx). Most patients with head and neck carcinoma underwent surgery as their initial treatment (69% surgery only, 16% radiation therapy only, 13% surgery and radiation therapy, and 2% other). Of the 82 patients who underwent surgery (69 patients received surgery only and 13 received surgery plus postoperative radiation therapy), 37 patients underwent wide local excisions or carbon dioxide laser excisions. We considered laryngectomies and mandibulectomies to be the more “extreme” surgical procedures. Four patients received total or near-total laryngectomies (three patients in the surgery only group and one patient in the surgery and radiation therapy group). Seven patients received mandibulectomies (five in the surgery only group and two in the surgery and radiation therapy group). Only one of these patients had a myocutaneous free flap placed in the surgical defect. The remaining surgical patients underwent surgery that was believed to be intermediate in severity and included a variety of procedures due to the different head and neck carcinoma locations (e.g., tonsillectomy, supraglottic laryngectomy, parotidectomy, or partial glossectomy). We did not adjust our analyses based on surgical procedures.

Table 1. Baseline Characteristics of Patients with Head and Neck Carcinoma (n = 101) and Matched Controls (n = 101)
CharacteristicsPatients with head and neck carcinoma (%)aControls (%)aPc
  • NDC: Mayo Clinic Nicotine Dependence Center; SD: standard deviation; NS: not significant; COPD: chronic obstructive pulmonary disease.

  • a

    Because of patients having missing data the numbers of patients within categories of a given characteristic do not always sum to 101. In all cases, percentages are based on those with data available.

  • b

    Fourteen patients with head and neck carcinoma had other non head and neck carcinomas diagnosed before the Mayo Clinic Nicotine Dependence Center consults (11 had one other cancer diagnosis and 3 had two other cancer diagnoses). Seven control patients had other non-head and neck carcinoma diasnoses before the Mayo Clinic Nicotine Dependence Center consult (all had only one cancer diagnosis).

  • c

    Two-tailed P value from chi-square test for categoric variables or rank sum test for continuous variables.

  • d

    Of the 101 patients with head and neck carcinoma, 10 experienced nine second primary tumor (SPM) that were a head and neck carcinoma before the Mayo Clinic Nicotine Dependence Center consult. In all cases, characteristics are presented for the initial head and neck carcinoma.

Demographics   
 Age (yrs)  NS
  Mean ± SD58.7 ± 10.158.3 ± 10.6 
  Median (range)59 (36–79)59 (35–80) 
 Gender  NS
  Female35 (34.7)35 (34.7) 
  Male66 (65.4)66 (65.4) 
 Race  NS
  White97 (98.0)95 (96.0) 
  Other 2 (2.0) 4 (4.0) 
 Education  NS
  Less than high school41 (42.3)41 (41.0) 
  High school graduate34 (35.1)26 (26.0) 
  Some post-high school13 (13.4)19 (19.0) 
  College graduate/more 9 (9.3)14 (14.0) 
 Marital status  NS
  Single/never married 8 (8.1) 9 (9.0) 
  Married69 (69.7)70 (70.0) 
  Divorced/separated14 (14.1)11 (11.0) 
  Widowed 8 (8.1)10 (10.0) 
 Patient type  NS
  Nonresidential outpatient69 (68.3)66 (65.4) 
  Nonresidential inpatient30 (29.7)33 (32.7) 
  Residential 2 (2.0) 2 (2.0) 
Tobacco use history   
 Product treated for by NDC  NS
  Cigarettes100 (99.0)100 (99.0) 
  Spit tobacco  1 (1.0)  1 (1.0) 
 Average cigarettes per day, past 6 mos  0.003
  1–1917 (17.0)34 (34) 
  20–3947 (47.0)48 (48) 
  ≥ 4036 (36.0)18 (18.0) 
 Age of first regular tobacco use (yrs)  NS
  ≤ 1751 (50.5)42 (42.0) 
  ≥ 1850 (49.5)58 (58.0) 
 No. of serious previous stop attempts  NS
  011 (12.8) 8 (9.0) 
  114 (16.3)11 (12.5) 
  2–539 (45.4)47 (53.4) 
  ≥ 622 (25.6)22 (25.0) 
 Longest previous stop attempt  NS
  Not at all/< 1 day20 (22.5)10 (11.5) 
  1–30 days28 (31.5)27 (31.0) 
  1–12 mos30 (33.7)32 (36.8) 
  ≥ 1 yr11 (12.4)18 (20.7) 
 Stage of readiness to change tobacco use  NS
  Precontemplation 5 (5.0) 6 (6.1) 
  Contemplation25 (24.8)30 (30.3) 
  Preparation16 (15.8)18 (18.2) 
  Action55 (54.5)45 (45.5) 
  Maintenance 0 (0.0) 0 (0.0) 
 Recommendation for quitting made by physician   
  No 2 (2.0)11 (11.0)0.010
  Yes98 (98.0)89 (89.0) 
 Other smokers in household  NS
  No53 (62.4)46 (56.8) 
  Yes32 (37.7)35 (43.2) 
Medical diagnoses before NDC consult   
 Asthma 3 (3.0) 5 (5.0)NS
 COPD35 (34.7)33 (32.7)NS
 Coronary artery disease22 (21.8)25 (24.8)NS
 Other cardiovascular disease 8 (7.9) 6 (5.9)NS
 Cerebrovascular disease 5 (5.0)12 (11.9)NS
 Peripheral vascular disease 5 (5.0)21 (20.8)< 0.001
 Cancer other than head and neckb14 (13.9) 7 (6.9)NS
 Major Depressive disorder13 (12.9)21 (20.8)NS
 Alcohol abuse/dependence38 (37.6)20 (19.8)0.005
 Drug abuse/dependence 1 (1.0) 4 (4.0)NS
Head and neck carcinoma characteristicsd   
 Time from head and neck carcinoma diagnosis to NDC consult   
  < 3 mos47 (46.5)  
  3 mos to 5 yrs27 (26.7)  
  > 5 yrs27 (26.7)  
 Stage at diagnosis   
  0 6 (5.9)  
  I38 (37.6)  
  II15 (14.9)  
  III15 (14.9)  
  IV22 (21.8)  
  Unknown 5 (5.0)  
 Histology   
  Squamous cell carcinoma90 (90.9)  
  Adenocarcinoma 2 (2.0)  
  Other 7 (7.1)  
 Primary site   
  Oral cavity37 (36.6)  
  Larynx37 (36.6)  
  Oropharynx19 (18.8)  
  Major salivary gland 5 (5.0)  
  Hypopharynx 3 (3.0)  
 Treatment   
  Surgery only69 (69.0)  
  Radiotherapy only16 (16.0)  
  Surgery and radiotherapy13 (13.0)  
  Other 2 (2.0)  

Compared with controls, the patients with head and neck carcinoma reported a higher rate of cigarette use during the 6 months before the consult. We found that 17%, 47%, and 36% of patients with head and neck carcinoma were smoking ≤ 19, 20–39, and ≥ 40 cpd compared with 34%, 48%, and 18% for the control group, respectively (P = 0.003). Patients with head and neck carcinoma were also more likely than controls to report that a physician recommended tobacco use cessation (98% vs. 89%; P = 0.010). In addition, patients with head and neck carcinoma had a higher frequency of alcohol abuse/dependence (38% vs. 20%; P=0.003) and a lower frequency of peripheral vascular disease (PVD; 5% vs. 21%; P = 0.002).

Tobacco Abstinence Rates

The tobacco abstinence rate 6 months after the NDC consult was 33% for patients with head and neck carcinoma and 26% for controls. Using conditional logistic regression and taking into account the matched-pair study design, the likelihood of tobacco abstinence 6 months after the NDC consult was not significantly different for patients with head and neck carcinoma relative to controls (odds ratio [OR] = 1.4, 95% confidence interval [95% CI], 0.8–2.4; P = 0.279). Similar findings were obtained from an analysis assessing tobacco abstinence at 6 months for patients with head and neck carcinoma relative to controls after adjusting for stage of change and average cpd during the 6 months before the consult (OR = 1.6, 95% CI, 0.8–3.2; P = 0.205). Similar findings were also obtained when the analysis was repeated with patients who died before the 6-month follow-up (3 patients with head and neck carcinoma, no controls) were classified as tobacco abstinent (data not shown).

Univariate Association of Baseline Characteristics and Tobacco Abstinence among Patients with Head and Neck Carcinoma

For the patients with head and neck carcinoma, additional univariate logistic regression analyses were performed to assess the association between baseline patient characteristics and tobacco abstinence after the NDC consult. Cancer stage at diagnosis was not associated with tobacco use outcomes (data not shown). However, the likelihood of tobacco abstinence 6 months after the NDC consult was significantly (P = 0.021) associated with the duration of time between the head and neck carcinoma diagnosis and the NDC consult (Fig. 1). The tobacco abstinence rate was 47% (22 of 47patients) for patients treated within 3 months of their diagnosis, 22% (6 of 27 patients) for patients treated between 3 months and 5 years of their diagnosis, and 19% (5 of 27 patients) for patients treated > 5 years after their diagnosis. In addition, among patients who had an NDC consult within 3 months of their diagnosis, the type of initial treatment received was predictive of tobacco use outcome. The tobacco abstinence rates at 6 months were 10%, 55%, and 50% for patients who received radiation therapy only (n = 10), surgery only (n = 29), and both surgery and radiation therapy (n = 6), respectively (P = 0.042; Fig. 2).

Figure 1.

Self-reported tobacco abstinence rates 6 months after a consultation at the Mayo Clinic Nicotine Dependence Center (NDC) for patients with head and neck carcinoma according to length of time from initial cancer diagnosis to NDC consultation.

Figure 2.

Self-reported tobacco abstinence rates 6 months after a consultation at the Mayo Clinic Nicotine Dependence Center (NDC) for patients with head and neck carcinoma treated within 3 months of the initial cancer diagnosis according to initial treatment received.

DISCUSSION

As more reports detail the ill effects of smoking during cancer therapies,8, 10, 33 they emphasize that patients should receive a strong and consistent message from their entire oncology team encouraging them to quit smoking. However, just receiving this advice might not be enough for those who are severely addicted.18 Oncologists should be aware that some patients may need additional behavioral and pharmacologic interventions or may need to be referred to specialized centers at which patients with more severe nicotine dependence can be offered more intensive services. The current study is unique because we were able to match a group of patients with head and neck carcinoma by age, gender, type of NDC consultation, and date of NDC consultation to a group of control patients from the general population of NDC patients and compare their tobacco abstinence rates. We observed that patients with head and neck carcinoma who receive individualized nicotine dependence interventions experienced similar tobacco use outcomes as patients without head and neck carcinoma. Furthermore, these results demonstrate that patients who are treated shortly after their cancer diagnosis is received may be more receptive to changing their behavior. Finally, patients with a recent diagnosis of head and neck carcinoma who receive primary radiation therapy are more likely to have lower tobacco abstinence rates than those who undergo surgery as part of their initial treatment.

We observed that 33% of the patients with head and neck carcinoma treated at the NDC self-reported tobacco abstinence at 6 months. Our results for recently diagnosed patients with head and neck carcinoma, (i.e., a 48% tobacco abstinence rate at 6 months) approach those of Gritz et al.,4 who found that 65% of the smokers in their study were abstinent at 1 year. Other, more recent, observational studies reported that 22–40% of patients continue to smoke after a diagnosis of head and neck carcinoma.5, 34–37 The current study population is slightly different from the population of patients utilized by Gritz et al. and the observational studies. We included patients who were diagnosed with head and neck carcinoma and then treated at the NDC to help address their nicotine addiction. Approximately 50% were treated at the NDC within 3 months of their diagnosis and the remainder were treated > 1 year after their diagnosis. This latter group continued to smoke despite their diagnosis or had returned to smoking before being treated at the NDC and tended to have lower tobacco abstinence rates after their NDC consultation. This selection bias may result in a group of patients who are more likely to have higher levels of dependence on nicotine.

Furthermore, one of our hypotheses was that patients who are diagnosed with head and neck carcinoma and who are treated during a teachable moment attributable to a life-threatening illness will be more likely to stop using tobacco. A priori, we wanted to evaluate the group of patients who were referred to the NDC close to the time of their diagnosis to determine whether they were more likely to be abstinent than those who were referred months or years after their diagnosis. For those with a recent diagnosis of head and neck carcinoma, we did find a higher rate of tobacco abstinence, which supports the premise that an abstinence attempt close to the time of initial diagnosis is the most effective time. Our findings have been corroborated in patients with lung carcinoma.38 McBride et al.23 suggested a heuristic model of the teachable moment. They emphasize that it is not only the health event itself that motivates patients to change their behavior, but also their perceived risk and emotional response, and how their perceived social role might change due to the key health event. They also advise that the timing of the intervention should be optimized to make the intervention as effective as possible.

The findings regarding cpd, alcohol use, and PVD merit discussion. First, the patients with head and neck carcinoma did report utilizing more cpd in the 6 months before their consultation than the controls and were more likely to have a history of alcohol abuse or dependence. These two factors together would lead one to believe that the patients with head and neck carcinoma would have been less likely to stop using tobacco as alcohol dependence and higher smoking rates are considered markers for higher levels of dependence.39 However, although not statistically significant, the patients with head and neck carcinoma were found to have a higher frequency of tobacco abstinence (33% vs. 26%, P = 0.279). Furthermore, adjusting for cpd did not appear to change the results. The lower rate of PVD among the patients with head and neck carcinoma compared with controls was not expected and was somewhat difficult to interpret. It may be that alcohol was protective for the development of PVD among the patients with head and neck carcinoma, although the protective effect is typically observed with moderate alcohol use.40, 41 We do not believe that this difference in PVD should have an impact on tobacco use outcomes. Therefore, we did not adjust for this diagnosis in our analyses and we did not perform any additional analyses or chart reviews to identify possible explanations for this observed difference. This finding may warrant future study.

Finally, as radiation therapy combined with chemotherapy continues to become a viable alternative to surgery for patients with head and neck carcinoma,42 we may need to address the finding that the patients who receive primary radiation therapy appear to have lower tobacco abstinence rates. Abstinence rates may decrease among patients with head and neck carcinoma as more patients receive radiation therapy as their primary modality. We found a tobacco abstinence rate of 10% for patients who received radiation therapy compared with 50–55% for patients who underwent surgery with or without adjuvant radiation therapy. Similarly, others have found that patients who receive radiation therapy have lower rates of tobacco abstinence.4, 34 Gritz et al.4 found that among the baseline smokers, the patients who received radiation therapy as their primary treatment modality had lower rates of continuous tobacco abstinence than patients who received surgical management for their primary head and neck carcinoma. The marginal rates of continuous tobacco abstinence at 12 months were 36% and 69% for radiation therapy and surgical management (excluding laryngectomy patients), respectively. The lower cessation rate among the patients who receive primary radiation therapy may occur because these patients receive their treatments as outpatients and likely do not have the same physical barriers (from the surgery) nor do they have a forced inpatient tobacco abstinence period.4, 43 However, one study did not find the same results because it did not include patients who received primary radiation therapy.36 Patients who receive radiation therapy who continue to smoke after being diagnosed with head and neck carcinoma or who resume smoking shortly after completing treatment may need novel approaches.

We acknowledge that the current study has limitations. First, the patients with head and neck carcinoma examined in our study cannot be assumed to reflect the average patient with head and neck carcinoma because of potential referral bias for treatment of their disease at a large tertiary center. There is limited ethnic diversity in the current study population. Another limitation is the statistical power of the current study to detect a difference between abstinence outcomes for patients with head and neck carcinoma and controls. The sample size was limited to the number of patients with head and neck carcinoma who were referred to the NDC during the study period. This selection bias could result in patients with more severe nicotine dependence being treated at the NDC. However, we were not able to calculate the number of patients with head and neck carcinoma who continued to smoke and were not referred to the NDC during this time frame. Finally, we did not biochemically confirm patients' self-reported tobacco abstinence at 6 months because this is a retrospective study of a clinical database.

Despite these limitations, the study has a number of strengths. First, there is little information in the literature pertaining to the treatment of nicotine dependence in patients with head and neck carcinoma. Because patients with head and neck carcinoma who continue to use tobacco are at high risk for developing future tobacco-related morbidity and mortality, attention to the treatment of this group is needed. The NDC database is well suited to study patients with head and neck carcinoma because it has supported other analyses of nicotine dependence treatment outcomes in the general patient population as well as in special populations including the elderly, adolescents, and patients with lung carcinoma.31, 38, 44, 45

The results of the current study demonstrate that patients with head and neck carcinoma who had an NDC consultation close to the time of their diagnosis had better tobacco abstinence outcomes than patients who were treated at a later time, especially for those who underwent a surgical approach as their primary cancer treatment.

Acknowledgements

The authors thank Jessica Gardner for her assistance with the preparation of this article and Laura Trautman for her data entry skills.

Ancillary