Cigarettes, alcohol, and depression ... - Semantic Scholar

Nicotine & Tobacco Research Volume 7, Number 2 (April 2005) 233–241

Cigarettes, alcohol, and depression: Characterizing head and neck cancer survivors in two systems of care Michael T. Lambert, Jeffrey E. Terrell, Laurel A. Copeland, David L. Ronis, Sonia A. Duffy [Received 11 May 2004; accepted 19 October 2004]

Tobacco exposure is a key risk factor for head and neck cancer, and continued smoking after diagnosis negatively affects outcomes. The present study examined tobacco smoking, nicotine dependence, alcohol use, and depression in survivors of head and neck cancer. Subjects at least 6 months post–initial diagnosis of head and neck cancer (N5694) drawn from three VA otolaryngology clinics (n5309, VA patients) and a university-based otolaryngology clinic (n5385, non-VA patients) were administered questionnaires and standardized rating instruments for nicotine and alcohol dependence and for depression. Additional clinical information was extracted from chart reviews. Despite high rates of prior smoking, less than one-quarter of all subjects continued to smoke. After controlling for significant confounding variables, we found that VA patients were more likely to be current smokers (OR51.9, 95% CI51.3–3.0), but current VA smokers did not differ significantly from non-VA smokers on the Fagerstro¨m Test for Nicotine Dependence criterion (p5.69). The VA patients were more likely to screen positive for problem drinking on the Alcohol Use Disorder Identification Test (OR52.1, 95% CI51.3–3.7). After adjusting for other variables, we found no statistical difference between the groups in depression scores on the Geriatric Depression Scale–Short Form. The study provides data on smoking, alcohol use, and depression in head and neck cancer survivors indicating that VA patients are at increased risk for continued smoking and problem drinking relative to non-VA patients. Head and neck cancer patients benefit from aggressive smoking cessation efforts by the VA, but many patients need specialized services that integrate smoking interventions with treatment of comorbid alcoholism.

Introduction Tobacco smoking, especially when combined with heavy alcohol consumption, is the paramount risk factor for otolaryngological cancer (head and neck cancer), including esophageal, oropharyngeal, and laryngeal squamous cell carcinomas (Chow, Michael T. Lambert, M.D., North Texas Veterans Health Care System and Department of Psychiatry, University of Texas Southwestern Medical School, Dallas; Jeffrey E. Terrell, M.D., Department of Otolaryngology, University of Michigan, Ann Arbor; Laurel A. Copeland, Ph.D., VA Ann Arbor Healthcare System, Ann Arbor, Michigan; David L. Ronis, Ph.D., VA Ann Arbor Healthcare System and University of Michigan School of Nursing, Ann Arbor; Sonia A. Duffy, Ph.D., R.N., VA Ann Arbor Healthcare System and Center for Practice Management and Outcomes Research, Ann Arbor, MI. Correspondence: Michael T. Lambert, M.D., Medical Director, Fort Worth VA Mental Health Clinic, 6000 Western Place, Suite 300, Fort Worth, TX 76107-4607, USA. Tel: +1 (214)-232-3882; Fax: +1 (817)570-2231; E-mail: [email protected]

McLaughlin, Hrubec, Nam, & Blot, 1993; Maier, Dietz, Gewelke, Heller, & Weidauer, 1992). Previous research indicates that many patients diagnosed with head and neck cancer continue to have significant preventable health-risk behaviors, including tobacco smoking and heavy alcohol consumption (Duffy et al., 2002; Mashberg, Boffetta, Winkelman, & Garfinkel, 1993). Continued tobacco smoking after cancer diagnosis is associated with poorer outcomes, increased risk of cancer recurrence, and higher rates of alcohol abuse (Gritz et al., 1993). Alcohol abuse persisting after the diagnosis of head and neck cancer is an independent predictor of survival (Deleyiannis, Thomas, Vaughan, & Davis, 1996). Multiple studies demonstrate the interrelations among smoking, alcohol use, and depression. Patients with mood disorders have higher rates of smoking and greater difficulty quitting (Glassman,

ISSN 1462-2203 print/ISSN 1469-994X online # 2005 Society for Research on Nicotine and Tobacco DOI: 10.1080/14622200500055418

234

CIGARETTES, ALCOHOL, AND DEPRESSION

Covey, Stetner, & Rivelli, 2001; Glassman, Helzer, & Covey, 2000; Laje, Berman, & Glassman, 2001). Smokers are more likely to consume alcohol, and those with alcohol problems experience greater difficulty quitting smoking (Hays, Schroeder, & Offord, 1999). Both smoking and alcohol abuse are associated with depression, and all three factors are associated with increased mortality (Amodei et al., 1994; Regier et al., 1990; Stommel, van den Akker, Milligan, & Goldman, 2002). Although patients with head and neck cancers often have histories of tobacco use, alcohol abuse, and depression, these disorders often go undetected or untreated (Terrell et al., 1997), despite evidence that treatment improves medical outcomes and quality of life and reduces costs (Cassano & Fava, 2002; Creed et al., 2002; Crown et al., 2002; Katon & Ciechanowski, 2002; Lave, Frank, Schulberg, & Kamlet, 1998; Terrell, Nanavati, Esclamado, Bradford, & Wolf, 1999; Weymuller et al., 2000). Nicotine and alcohol use disorders, as well as depression, are encountered more frequently in patients treated by the Department of Veterans Affairs (VA) than in the general population (Enstrom, 1999; Lambert & Fowler, 1997; Lavretsky et al., 2002; McKinney, McIntire, Carmody, & Joseph, 1997). A lifetime history of ever smoking tobacco is reported in 74.2% of veterans, compared with 48.4% of the general U.S. population (Klevens et al., 1995). It is estimated that over 50% of cancers occurring in veterans are causally associated with tobacco smoking (McLaughlin, Hrubec, Blot, & Fraumeni, 1995). The Veterans Health Study showed the prevalence of depression among veterans to be two to five times higher than in the general U.S. population (Hankin, Spiro, Miller, & Kazis, 1999). In 2002, 12% of the 4.5 million patients treated by the VA carried a diagnosis of depression (Department of Veterans Affairs, 2003b). Medical consequences of cigarette smoking and alcohol abuse are particularly common among VA psychiatric patients and contribute to elevated mortality from cancer, cardiovascular disease, and hepatic disorders (Lambert, Schmidt, & LePage, 2003). In summary, tobacco smoking, excessive alcohol use, and depression are common in the general population, more common among veterans, and greatly over-represented among patients receiving treatment for head and neck cancers. Given that continued tobacco smoking, alcohol use, and depression negatively affect clinical outcomes, information about this constellation of disorders in head and neck cancer patients is needed so that cessation programs can be optimally developed and tested. The present study examined tobacco smoking, nicotine dependence, alcohol use, and depression in head and neck

cancer survivors and compared clinical and demographic differences between patients receiving care in VA outpatient clinics and a university-based clinic.

Method Data sources and design Data presented are from subjects who agreed to be screened for enrollment in a multisite, longitudinal VA Health Services Research and Development– funded project evaluating nurse-administered cognitive-behavioral interventions for tobacco smoking, alcohol use, and depression among head and neck cancer patients. Across four sites, 71% of potential subjects approached to take the baseline survey over 3 years (2000–2002) completed the survey and signed informed consent. The sample screened for the intervention study consisted of 1,004 head and neck cancer patients. With institutional review board approval for administration of the study at all sites, patients completed a self-administered health survey including questions on health behavior, health status, and demographic information. A research nurse administered assessment instruments and conducted chart reviews to provide cancer diagnostic, staging, and treatment data. The present study reports on these baseline assessments and is cross-sectional in design. Subject selection criteria and final sample In our experience, some newly diagnosed head and neck cancer patients abruptly but only temporarily stop cigarette smoking or alcohol use, and reactive depressions are short lived. Patients with highly advanced incurable disease may not be motivated or encouraged to stop smoking and drinking. To prevent such effects from biasing the results, we included in the present study only patients surviving at least 6 months after the initial diagnosis of head and neck cancer. Additional exclusion criteria were pregnancy, age under 18 years, non-English speaking, or psychiatric instability. A total of 310 potential subjects were excluded based solely on the 6-month exclusion criterion, resulting in the final sample of 694 patients recruited from VA Ann Arbor Healthcare System (14%, n5100); VA North Texas Health Care System Dallas (15%, n5106); North Florida/South Georgia Veterans Health System (15%, n5103), and the University of Michigan Medical Center (55%, n5385). Measures Demographic information consisted of age, gender, race, education, and marital status. Female gender

NICOTINE & TOBACCO RESEARCH

was essentially collinear with non-VA status because only two VA patients were female. Race was collected as White, African American, Hispanic, American Indian, and other and recoded as White (1) versus non-White (0). Education was recoded into two categories: Some college or more (1) versus high school or less (0). Marital status was classified into two categories: Married (1) versus widowed/ divorced/separated/never married (0). Clinical data extracted from subjects’ medical records included time since diagnosis, tumor site, tumor stage, primary site surgery, secondary primary cancer site surgery, neck dissection, radiation, chemotherapy, tracheotomy, and feeding tube status. Tumor site was coded into two groups: larynx (1) versus oropharynx/hypopharynx, oral cavity or other (0). Using the American Joint Committee on Cancer staging system (Greene, 2002), we dichotomized tumor stage based on severity into stages 0, I, and II (early; 0) versus stages III and IV (advanced; 1). The number of medical comorbidities was assessed by a self-report of seven selected diseases (lung, heart, other cancer, stroke, diabetes, arthritis, and other mental health problems). Validated instruments assessed tobacco smoking and alcohol use. Use status for tobacco smoking was categorized into never used, previously used (former user), and currently using (current smoker). Questionnaires collected data regarding previous smoking cessation attempts. Nicotine dependence was assessed by the Fagerstro¨m Test for Nicotine Dependence (FTND; Fagerstro¨m, Heatherton, & Kozlowski, 1990), a 10-point scale with a score of 6 or greater indicating nicotine addiction. The FTND score was dichotomized into nicotine dependent (1) versus not (0) using the standardized cutoff score of 6. For patients identifying themselves as current smokers, the number of tobacco cigarettes consumed daily was documented. Alcohol dependence was assessed by the Alcohol Use Disorder Identification Test (AUDIT; Saunders, Ashland, Babor, de la Fuente, & Grant, 1993); a score of 8 or more is indicative of problematic drinking. Current alcohol use status was identified in a manner similar to that used to identify tobacco smoking status. For current drinkers, quantitative consumption data on number of drinks per week or month was collected. Depression was assessed by the Geriatric Depression Scale–Short Form (GDS-SF). A score of 4 or more on the GDS-SF indicates probable depression (Lewinsohn et al., 1997). Data analysis To determine whether significant differences in smoking, alcohol use, depressive symptoms, clinical characteristics, or demographics existed between the VA and non-VA patient cohorts, we initially applied

235

bivariate models. Chi-square tests were used to analyze categorical dependent variables, and Student’s t tests were used to assess interval-level variables. The Mantel-Haenszel chi-square test was used for the analyses of ordinal variables. Significance of bivariate results determined predictors applied in subsequent multivariate model analysis, with the exception of chemotherapy, included based on clinical relevance. We used logistic regression to examine dichotomous outcomes by VA patient status, controlling for age, education (some college vs. less education), marital status (married vs. other), time elapsed since initial diagnosis (in months), tumor site, tumor stage, comorbidities, and chemotherapy and radiation therapy. The small number of responses on some outcomes constrained the number of predictors. In those cases, more parsimonious models were applied omitting predictors that were nonsignificant in the full model. Poisson regression was used to analyze skewed outcomes such as depression and FTND scores in the multivariate models; a variant of this approach for zeroinflated data, negative binomial regression, was used to analyze data from the AUDIT, on which many subjects scored a 0. These models assessed the effect of VA care status, controlling for the same demographic and clinical predictors as the logistic regression models. For all analyses, a criterion alpha level of .05, established a priori, was applied.

Results Demographic and clinical findings Table 1 presents demographic information. On average, VA head and neck cancer patients were slightly older and predominantly male, compared with nonVA patients. Patients obtaining VA care were significantly less likely than non-VA patients to be White, have higher education, or be married. Few clinical characteristics differed significantly between the two cohorts (Table 1). No significant differences by VA care status were detected on neck dissection, primary site surgical intervention, radiation, use of a feeding tube, or tracheotomy. Overall, more time had passed since the initial diagnosis of cancer for VA patients compared with non-VA patients (60 months vs. 47 months, p,.01), and VA patients were more likely to have tumors in earlier stages 0, I, or II rather than more advanced stages III and IV, compared with non-VA patients (41% vs. 30%, p,.01). Patients receiving care in the university setting were significantly more likely than VA patients to have a second primary cancer surgery (23% vs. 13%, p,.01). Tumor site varied significantly; VA patients were more likely than non-VA patients to have cancer of the larynx (47% vs. 28%, p,.001). Compared with university

236


Table 1. Clinical and demographic characteristics of VA and non-VA head and neck cancer survivors Measure Demographic variablea Mean age, years (n5694) Gender, percent male (n5694) Race/ethnicity (n5684) White African American Other Education (n5686) High school or less Some college College degree Marital status (n5688) Married Widowed Divorced/separated/never Clinical variablea Months since diagnosis (n5694) Cancer site (n5694) Larynx Hypopharynx/oropharynx Oral/other sites Cancer stage (n5667) 0, I, II III or IV Primary site surgery (n5693) Second primary site surgery (n5693) Neck dissection (n5694) Chemotherapy (n5694) Radiation (n5694) Tracheotomy (n5694) Feeding tube (n5694) Comorbidities (n5680) None One Two or more

VA patients (n5309)a, Percent or mean (SD)

Non-VA patients (n5385)a, Percent or mean (SD)

63.6 (9.6) 99%

60.3 (10.8) 73%

80% 15% 5%

93% 4% 3%

59% 33% 8%

49% 29% 22%

51% 11% 38%

68% 10% 22%

60.0 (62.2)

47.3 (53.9)

47% 29% 25%

28% 36% 36%

41% 59% 60% 13% 53% 19% 83% 4% 12%

30% 70% 65% 23% 55% 26% 82% 4% 14%

22% 26% 51%

46% 29% 25%

p value ,.001d ,.001b ,.001b

,.001c

,.001b

.005d ,.001b

.002b .170b ,.001b .543b .022b .642b .817b .638b ,.001c

a

Incomplete responses lower totals for some variables. Pearson chi-square. c Mantel-Haenszel chi-square test for trend in ordinal data. d Student’s t test. b

patients, VA patients were less likely to have undergone chemotherapy (p,.05) but more likely to have additional medical comorbidities (p,.001). Tobacco smoking and nicotine dependence findings A history of previous successful smoking cessation was found in 66% of all subjects (former smokers), similar between the two groups; and a lifetime history of cigarette smoking was found in 89% of all subjects (Table 2). Of former smokers, 81.1% had quit more than 1 year prior to the study and 18.9% had quit in the past year. Of current smokers responding to the question ‘‘Have you tried to stop smoking in the past?’’ and, if yes, ‘‘How many times have you tried to stop in the past?’’, 90% of non-VA versus 94% of VA head and neck cancer patients (p5.29) reported having ever attempted cessation. The Wilcoxon two-sample test was applied to the number of cessation attempts because the data were ordinal but not normally distributed. No significant difference was found in reported attempts, which averaged 4.5 for VA patients and 6.9 for non-VA

patients (n5115, p5.96). Bivariate analysis initially revealed a significantly higher percentage of VA subjects with a lifetime history of smoking, compared with non-VA subjects (94% vs. 84%, p,.01), but among current smokers no significant differences were found in the number of cigarettes smoked daily. According to bivariate analysis, VA and non-VA smokers did not differ statistically on mean FTND scores. Significant differences existed between the two groups in terms of current smoking, with current smoking higher among VA head and neck cancer patients (p,.01) than among non-VA patients. In the multivariate model, this difference persisted (p,.01), and the control variables of older age, college educated, married, and higher tumor stage significantly decreased the risk of being a current smoker (Table 4). A multivariate analysis of having a FTND score positive for nicotine dependence confirmed that subjects receiving VA care were no more likely than university-care patients to meet this criterion, although several control variables were significant. Older age, college education, and higher tumor stage


237

Table 2. Smoking status, nicotine dependence scores, and cigarette consumption in VA and non-VA head and neck cancer survivors Variablea Smoking status (n5691) Current smoker Former smoker Never smoker (lifetime history) Current smokers as a percentage of ever-smokers (n5613) Mean FTND score for current smokers (n5151) Cigarettes per day for current smokers (n5152) (10 11–20 21–30 .30



29% 65% 6% 31.0% 4.7 (2.1)

17% 67% 16% 20.7% 4.5 (2.6)

28% 33% 33% 6%

30% 36% 22% 12%

p value ,.001b

.004b .454d .997c

Note. FTND, Fagerstro¨m Test for Nicotine Dependence. a Incomplete responses lower totals for some variables. b Pearson chi-square. c Mantel-Haenszel chi-square test for trend in ordinal data. d Wilcoxon two-sample test.

were protective against the risk of scoring positive for nicotine dependence, whereas more comorbid conditions increased this risk. Because the power of this analysis was low, a more parsimonious multivariate model omitting predictors with p values greater than .10 was applied. The adequately powered model confirmed the previous findings and added the result that chemotherapy increased the odds of scoring positive for nicotine dependence. Alcohol use and AUDIT findings Alcohol use (lifetime history) was highly prevalent in both subcohorts, approximately 90%. In the initial bivariate analyses, significant differences existed

between VA and non-VA head and neck cancer patients in terms of current, former, and never drinking (Table 3). Current use of alcohol was significantly less common among VA subjects than among non-VA subjects, although previous use was significantly higher among those obtaining VA care. Although VA patients were less likely than non-VA patients to be current drinkers, significantly more patients in the VA setting scored positive for problem drinking on the AUDIT, and VA patients who did drink reported higher frequency and heavier use of alcohol compared with non-VA patients. Scoring positive on the AUDIT for alcohol abuse was multivariately modeled with logistic regression (Table 4). Significant differences were found by care

Table 3. Alcohol use status, AUDIT scores, and alcohol consumption in VA and non-VA head and neck cancer survivors

a

Variable

Drinking status (n5682) Current drinker Former drinker Never drank Possible problem drinking (AUDIT>8; n5667) Mean AUDIT score for current drinkers (n5284) How often drink for current drinkers (n5279) Less than monthly 2–4 times/month 2–3 times/week >4 times/week Drinks per day for current drinkers (n5278) 1 or 2 3 or 4 5 or 6 7 to 9 >10 Note. AUDIT, Alcohol Use Disorder Identification Test. a Incomplete responses lower totals for some variables. b Pearson chi-square. c Mantel-Haenszel chi-square test for trend in ordinal data. d Wilcoxon two-sample test.



p value .002b

37% 57% 6% 16% 7.8 (6.7)

45% 44% 11% 9% 4.7 (4.4)

11% 21% 26% 42%

19% 29% 25% 26%

46% 29% 17% 7% 2%

67% 22% 6% 3% 1%

.004b ,.001d .002c

,.001c

238


Table 4. Multivariate logistic regression results for VA and non-VA head and neck cancer survivors (odds ratios with 95% confidence intervals) Current smoker VA patient status Age (29–86 years) College education Married vs. other White vs. other Months since diagnosis High tumor stage Tumor site Chemotherapy Radiation therapy Comorbid condition

1.94 0.97 0.55 0.43 0.98 1.00 0.60 0.90 1.24 0.96 1.06

(1.28–2.96)* (0.95–0.99)* (0.37–0.83)* (0.28–0.63)* (0.56–1.70) (0.99–1.00) (0.38–0.94)* (0.59–1.38) (0.77–2.02) (0.56–1.66) (0.90–1.24)

Positive FTND 1.16 0.94 0.27 0.61 1.37 1.00 0.45 0.90 1.86 1.06 1.43

(0.66–2.05) (0.91–0.96)* (0.15–0.48)* (0.36–1.03) (0.62–3.04) (0.99–1.00) (0.24–0.81)* (0.51–1.61) (0.99–3.52) (0.52–2.14) (1.17–1.75)*

Positive AUDIT 2.14 0.98 0.99 0.54 0.77 1.00 0.93 0.89 0.74 0.79 0.91

(1.25–3.67)* (0.95–1.00) (0.61–1.63) (0.33–0.89)* (0.40–1.48) (0.99–1.00) (0.52–1.66) (0.52–1.53) (0.39–1.38) (0.39–1.62) (0.73–1.12)

Positive GDS-SF 0.84 0.96 0.64 0.62 0.63 1.00 1.84 1.44 0.78 1.00 1.51

(0.58–1.23) (0.94–0.98)* (0.46–0.91)* (0.44–0.89)* (0.37–1.05) (1.00–1.01) (1.23–2.75)* (0.99–2.10) (0.51–1.19) (0.62–1.62) (1.30–1.75)*

Note. AUDIT, Alcohol Use Disorder Identification Test; FTND, Fagerstro¨m Test for Nicotine Dependence; GDS-SF, Geriatric Depression Scale–Short Form. *p,.05.

setting status using logistic regression, with VA patients more likely than non-VA patients to meet criteria for alcohol abuse (OR52.1, 95% CI51.3– 3.7). Marriage was a significant protective factor. Again, because of the small number of persons scoring positive, the power of the analysis was limited and a more parsimonious multivariate model was sought. The only other predictor with a p value less than .10 was age, which proved significantly associated with decreasing risk of alcohol abuse in the better-powered model. In the negative binomial regression model of the AUDIT score, the difference between VA and non-VA patients persisted (p5.04). Older age also acted as a significant protective factor in this model. Depression findings The proportion of head and neck cancer patients scoring at or above the cutoff considered positive for probable depression (GDS-SF score >4) was not significantly higher among VA subjects than non-VA subjects (48% vs. 41%, p5.08). However, using an unadjusted analysis, we found that VA head and neck cancer patients had significantly higher mean scores on the GDS-SF than did non-VA patients (4.2 vs. 3.5, p,.01), and their mean GDS-SF score of 4.2 was slightly above the threshold of 4.0 established for probable clinical depression. Using logistic regression, we found that scoring positive for depression was predicted by VA care status controlling for the same demographic and clinical predictors used in the previous analyses (Table 4). Older age, college education, and being married were again significantly protective factors, whereas higher tumor stage and more comorbidities increased the risk of depression. This model was adequately powered; therefore, marginal effects were not explored further. A Poisson regression of the depression score confirmed the lack of statistically significant difference by VA care status (p5.28).

Discussion To our knowledge, the present study is the first investigation to describe and compare characteristics of head and neck cancer survivors. Controlling for significant confounding variables, we found that patients receiving care from the VA were more likely to be current smokers, but, surprisingly, current VA smokers did not differ significantly from non-VA smokers on the FTND. Veteran care status was associated with an increased likelihood of screening positive for problem drinking. No statistical difference based on care setting status was found in depression scores after adjusting for other variables. The 29% of VA head and neck cancer patients still smoking exceeds rates among the general U.S. population (23.1%), among U.S. males aged 45–64 years (26.4%), and among U.S. males aged 65 years or more (10.2%; U.S. Census Bureau, 2002). For comparison, the overall smoking rate for all patients seeking VA care is estimated to be 33% (Department of Veterans Affairs, 2003a). Although cancer may not always be sufficient motivation for some patients to quit smoking, a large percentage of former smokers with head and neck cancer are able to achieve and maintain complete cessation (Gritz et al., 1993; Pancholi, 2000; Poirier et al., 2002). Credit is due to the VA for instituting widespread tobacco use screening and for establishing effective and readily accessible smoking cessation programs. One study showed that 42% of VA smoking program patients achieved complete cessation at 3 months (Sherman, Wang, & Nguyen, 1996), and 35% maintained smoking cessation at 6 months in another VA study (Simon, Carmody, Hudes, Snyder, & Murray, 2003). The VA Normative Aging Study found that smoking relapse rates fell to less than 1% per year after 10 years of abstinence (Krall, Garvey, & Garcia, 2002). The Agency for Health Policy and Research smoking cessation guidelines are demonstrated to be effective in the VA health care system (Fiore et al., 2000; Ward et al., 2003).


Additional recommendations exist to increase the effectiveness of smoking cessation programs in populations with comorbid psychiatric disorders (Ziedonis & Williams, 2003). This information should hearten clinicians providing smoking cessation treatment to these populations. Although VA head and neck cancer patients were no more likely to screen positive for depressive symptoms than were non-VA patients, rates of depressive symptoms were consistent with reported rates in other serious physical disorders (Katon & Schulberg, 1992). Treatment of depression improves outcomes in a variety of medical conditions and should be expected to benefit head and neck cancer patients as well. Treatment of depression also may enhance successful tobacco smoking and drinking cessation (Smith et al., 2003). Compared with non-VA patients, the VA head and neck cancer patients were older; predominantly male; more likely to have lower educational levels; and more commonly divorced, separated, or never married. Patients with these demographic characteristics often lack sufficient social support (Kessler, 2001). Moreover, being older and unmarried results in social isolation and decreased interpersonal support. Social support is demonstrated to improve outcomes and survival in selected cancer populations (Fawzy, Fawzy, Arndt, & Pasnau, 1995; Spiegel, Bloom, Kraemer, & Gottheil, 1989). Whereas VA patients had more medical comorbidities than did the non-VA university cohort, patients attending VA clinics were significantly more likely than non-VA subjects to have tumors in less advanced stages. This finding may be the result of patients with more advanced disease being frequently referred to tertiary care facilities such as the University of Michigan, or it may speak to the success of early cancer screening and detection in the VA system. Overall, clinical and treatment characteristics of both cohorts of head and neck cancer patients were similar, suggesting that for this disorder the VA provides treatment comparable with that from a large university-based care facility. Other more generalized studies recently profiled the overall high quality of VA medical care (Ashton et al., 2003; Kaboli, Barnett, Fuehrer, & Rosenthal, 2001). Tumors of the larynx (as opposed to other sites) were more common among patients in the VA setting. Because laryngeal carcinoma is related almost exclusively to tobacco smoking, the higher rates of smoking among VA patients may account for the greater number of cancers of the larynx. Patients receiving VA care were less likely than non-VA patients to have undergone chemotherapy, which may be because earlier-stage laryngeal cancers, more common in veterans, are treated primarily with surgery or radiation as opposed to the

239

chemoradiation protocols frequently used for more advanced cancers.

Study limitations The present findings may not be fully representative of head and neck cancer patients in other treatment settings or completely generalizable to other types of cancer. Although the subjects were a convenience sample of patients, the facilities were located in three distinct geographic regions of the country. Because some veterans may have sought university care, the present study is best viewed as a comparison of patients obtaining care at these treatment settings. The study assessed patients who had survived the initial rigors of cancer diagnosis and treatment and therefore may not be generalizable to newly diagnosed patients. Although laboratory confirmation of cessation status (e.g., carbon monoxide monitoring, breathalyzer testing, salivary or urinary cotinine) was not performed, another study found more than 90% concurrence between self-reported cessation and biochemical validation (Glasgow, Mullooly, & Vogt, 1993). Finally, because we collected only cigarette consumption data, the present study did not assess use of nonsmokable forms of tobacco or delineate the proportion of smokers who were pipe or cigar users.

Conclusion Regardless of treatment setting, histories of tobacco smoking, alcohol use, and depression are common in head and neck cancer survivors. Although current smoking and problematic drinking were more common in subjects receiving VA care than in non-VA subjects, the majority of head and neck cancer patients with past smoking histories were no longer currently smoking. Based on these findings, special programs for VA head and neck cancer patients that integrate smoking cessation with alcohol treatment and social support interventions deserve further investigation.

Acknowledgments This research was sponsored by VA grant IIR 98-500 and an unrestricted grant from Glaxo SmithKline Beecham. Continued investigation in this line of research is sponsored by the National Institutes of Health through the University of Michigan’s Head and Neck SPORE grant (CA97248). The authors thank Larry Myers and Carol Bishop for their work as site investigators at the Dallas and Gainesville VA hospitals, respectively. The authors thank the University of Michigan Head and Neck Cancer Team as well as project manager Karen Fowler and research assistant Cindy Wampler for their meticulous data collection. Tara Saia Lewis provided important suggestions for the manuscript. Shirley Campbell, VA North Texas librarian, provided valuable assistance. Finally, the authors thank the patients who contributed their time participating in the project.

240


References Amodei, N., Elkin, B. B., Burge, S. K., Rodriguez-Andrew, S., Lane, P., & Seale, J. P. (1994). Psychiatric problems experienced by primary care patients who misuse alcohol. International Journal of Addiction, 29, 609–626. Ashton, C. M., Souchek, J., Petersen, N. J., Menke, T. J., Collins, T. C., Kizer, K. W., Wright, S. M., & Wray, N. P. (2003). Hospital use and survival among Veterans Affairs beneficiaries. The New England Journal of Medicine, 23, 1637–1646. Cassano, P., & Fava, M. (2002). Depression and public health. Journal of Psychosomatic Research, 53, 849–857. Chow, W. H., McLaughlin, J. K., Hrubec, Z., Nam, J. M., & Blot, W. J. (1993). Tobacco use and nasopharyngeal carcinoma in a cohort of U.S. veterans. International Journal of Cancer, 55, 538–540. Creed, F., Morgan, R., Fiddler, M., Marshal, S., Guthrie, E., & House, A. (2002). Depression and anxiety impair health-related quality of life and are associated with increased costs in general medical inpatients. Psychosomatics, 43, 302–309. Crown, W. H., Finkelstein, S., Berndt, E. R., Ling, D., Poret, A. W., Rush, A. J., & Russell, J. M. (2002). The impact of treatmentresistant depression on health care utilization and costs. Journal of Clinical Psychiatry, 63, 963–971. Deleyiannis, W. B., Thomas, D. V., Vaughan, T. L., & Davis, S. (1996). Alcoholism: Independent predictor of survival in patients with head and neck cancer. Journal of the National Cancer Institute, 88, 542–549. Department of Veterans Affairs. (2003a). National smoking and tobacco use cessation program. (VHA Directive 2203-042). Washington, DC: Author. Department of Veterans Affairs. (2003b). VA fact sheet. Retrieved October 29, 2004, from http://www1.va.gov/opa/fact/docs/ vafacts.htm Duffy, S. A., Terrell, J. E., Valenstein, M., Ronis, D. L., Copeland, L. A., & Conners, M. (2002). Effect of smoking, alcohol, and depression on the quality of life of head and neck cancer patients. General Hospital Psychiatry, 2, 140–147. Enstrom, J. E. (1999). Smoking cessation and mortality among two United States populations. Journal of Clinical Epidemiology, 2, 813–825. Fagerstro¨m, K. O., Heatherton, T. F., & Kozlowski, L. T. (1990). Nicotine addiction and its assessment. Ear, Nose, & Throat Journal, 69, 763–765. Fawzy, F. I., Fawzy, N. W., Arndt, L. A., & Pasnau, R. O. (1995). Critical review of psychosocial interventions in cancer care. Archives of General Psychiatry, 52, 100–113. Fiore, M. C., Bailey, W. C., Cohen, S. J., Dorfman, S. F., Gritz, E. R., Heyman, R. B., Holbrook, J., Jaen, C. R., Kottke, T. E., Lando, H. A., Mecklenbur, R., Mullen, P. D., Nett, L. M., Robinson, L., Stitzer, M., Tommasello, A. C., Villejo, L., & Wewers, M. E. (2000). Treating tobacco use and dependence. Clinical practice guideline. Rockville, MD: U.S. Department of Health and Human Services. Public Health Service. Glasgow, R. E., Mullooly, J. P., & Vogt, T. M. (1993). Biochemical validation of smoking status: Pros, cons, and data from four lowintensity intervention trials. Addictive Behavior, 18, 511–527. Glassman, A. H., Covey, L. S., Stetner, F., & Rivelli, S. (2001). Smoking cessation and the course of major depression: A follow-up study. Lancet, 357, 1929–1932. Glassman, A. H., Helzer, J. E., & Covey, L. S. (2000). Smoking, smoking cessation, and major depression. The Journal of the American Medical Association, 264, 1546–1549. Green, F. L. (Ed.) (2002). AJCC cancer staging manual (6th ed.). New York: Springer. Gritz, E. R., Carr, C. R., Rapkin, D., Abemayor, E., Chang, L. J., Wong, W. K., Belin, T. R., Calcaterra, T., Robbins, K. T., & Chonkich, G. (1993). Predictors of long-term smoking cessation in head and neck cancer patients. Cancer Epidemiology, Biomarkers and Prevention, 2, 261–270. Hankin, S. H., Spiro, A. S., Miller, D. R., & Kazis, L. (1999). Mental disorders and mental health treatment among U.S. Department of Veterans Affairs outpatients: The Veterans Health Study. American Journal of Psychiatry, 156, 1924–1930. Hays, J. T., Schroeder, D. R., & Offord, K. P. (1999). Response to nicotine dependence treatment in smokers with current and past alcohol problems. Annals of Behavioral Medicine, 21, 244–250.

Kaboli, P. J., Barnett, M. J., Fuehrer, S. M., & Rosenthal, G. E. (2001). Length of stay as a source of bias in comparing performance in VA and private sector facilities: Lessons learned from a regional evaluation of intensive care outcomes. Medical Care, 39, 1014–1024. Katon, W., & Ciechanowski, P. (2002). Impact of major depression on chronic medical illness. Journal of Psychosomatic Research, 53, 859–863. Katon, W., & Schulberg, H. (1992). Epidemiology of depression in primary care. General Hospital Psychiatry, 14, 237–247. Kessler, R. C. (2001). The effects of chronic medical conditions on work loss and cutback. Journal of Occupational and Environmental Medicine, 43, 218–225. Klevens, R. M., Giovino, G. A., Peddicord, J. P., Nelson, D. E., Mowery, P., & Grumer-Strawn, L. (1995). The association between veteran status and cigarette-smoking behaviors. American Journal of Preventive Medicine, 11, 245–250. Krall, E. A., Garvey, A. J., & Garcia, R. I. (2002). Smoking relapse after 2 years of abstinence: Findings from the VA Normative Aging Study. Nicotine & Tobacco Research, 3, 95–100. Laje, R. P., Berman, J. A., & Glassman, A. H. (2001). Depression and nicotine: Preclinical and clinical evidence for common mechanisms. Current Psychiatry Reports, 3, 470–476. Lambert, M. T., & Fowler, D. R. (1997). Suicide risk factors among veterans: Risk management in the changing culture of the VA. The Journal of the Mental Health Administration, 24, 350–358. Lambert, M. T., Schmidt, A., & LePage, J. (2003). Five-year outcomes following psychiatric consultation in a tertiary care emergency room. American Journal of Psychiatry, 160, 1350–1353. Lave, J. R., Frank, R. G., Schulberg, H. C., & Kamlet, M. S. (1998). Cost-effectiveness of treatments for major depression in primary care practice. Archives of General Psychiatry, 55, 645–651. Lavretsky, H., Bastani, R., Gould, R., Huang, D., Llorente, M., Maxwell, A., & Jarvik, L. (2002). Predictors of two-year mortality in a prospective ‘‘UPBEAT’’ study of elderly veterans with comorbid medical and psychiatric symptoms. American Journal of Geriatric Psychiatry, 10, 458–468. Lewisohn, P. M., Seeley, J. R., Roberts, R. E., & Allen, N. B. (1997). Center for Epidemiologic Studies Depression Scale (CES-D) as a screening instrument for depression among community-residing older adults. Psychology and Aging, 12, 277–287. Maier, H., Dietz, A., Gewelke, U., Heller, W. D., & Weidauer, H. (1992). Tobacco and alcohol and the risk of head and neck cancer. Journal of Clinical Investigation, 70, 320–327. Mashberg, A., Boffetta, P., Winkelman, R., & Garfinkel, L. (1993). Tobacco smoking, alcohol drinking, and cancer of the oral cavity and oropharynx among US veterans. Cancer, 62, 1369–1375. McKinney, W. P., McIntire, D. D., Carmody, T. J., & Joseph, A. (1997). Comparing the smoking behavior of veterans and nonveterans. Public Health Reports, 112, 212–217. McLaughlin, J. K., Hrubec, Z., Blot, W. J., & Fraumeni, J. F. (1995). Smoking and cancer mortality among U.S. veterans: A 26-year follow-up. International Journal of Cancer, 60, 190–193. Pancholi, M. (2000). Agency for Healthcare Research and Quality Medical Expenditure Survey. Retrieved October 29, 2004, from www.meps.ahrq.gov/papers/st7/stat07.htm. Regier, J. A., Farmer, J., Rae, S. D., Locke, B. Z., Keith, S. J., Judd, L. L., & Goodwin, F. K. (1990). Comorbidity of mental disorders with alcohol and other drug abuse: Results from the Epidemiologic Catchment Area (ECA) Study. Alcohol, 264, 2511–2518. Saunders, J. B., Ashland, O. G., Babor, T. F., de la Fuente, J. R., & Grant, M. (1993). Development of the Alcohol Use Disorders Identification Test (AUDIT): WHO collaborative project on early detection of persons with harmful alcohol consumption-II. Addiction, 88, 791–804. Sherman, S. E., Wang, M. M., & Nguyen, B. (1996). Predictors of success in a smoking cessation clinic. Journal of General Internal Medicine, 11, 702–704. Simon, J. A., Carmody, T. P., Hudes, E. S., Snyder, E., & Murray, J. (2003). Intensive smoking cessation versus minimal counseling among hospitalized smokers treated with transdermal nicotine replacement: A randomized trial. American Journal of Medicine, 114, 555–562. Smith, S. S., Jorenby, D. E., Leischow, S. J., Nides, M. A., Rennard, S. I., Johnston, A., Jamerson, B., Fiore, M. C., & Baker, T. B. (2003). Targeting smokers at increased risk for relapse: Treating

NICOTINE & TOBACCO RESEARCH women and those with a history of depression. Nicotine & Tobacco Research, 5, 99–109. Spiegel, D., Bloom, J. R., Kraemer, H. C., & Gottheil, E. (1989). Effect of psychosocial treatment on survival of patients with metastatic breast cancer. Lancet, 2, 888–891. Stommel, M., van den Akker, O. B. A., Milligan, D. W., & Goldman, J. M. (2002). Depression and functional status as predictors of death among cancer patients. Cancer, 94, 2719–2727. Terrell, J., Nanavati, K., Esclamado, R., Bishop, J., Bradford, C., & Wolf, G. (1997). Head and neck cancer-specific quality of life: Instrument validation. Archives of Otolaryngology—Head & Neck Surgery, 123, 1125–1132. Terrell, J. E., Nanavati, K., Esclamado, R. M., Bradford, C. R., & Wolf, G. T. (1999). Health impact of head and neck cancer. Otolaryngology—Head & Neck Surgery, 120, 852–859.

241

U.S. Census Bureau. (2002). Statistical abstract of the United States, Year 2000, No. 184. Washington, DC: Author. Ward, M. M., Doebbling, B. N., Vaughn, T. E., Uden-Holman, T., Clarke, W. R., Woolson, R. F., Letuchy, E., Branch, L. G., & Perlin, J. (2003). Effectiveness of a nationally implemented smoking cessation guideline on provider and patient practice. Preventive Medicine, 36, 265–271. Weymuller, E. A., Yeuh, B., Deleyiannis, F. W., Kuntz, A. L., Alsarraf, R., & Cotrera, M. D. (2000). Quality of life in patients with head and neck cancer: Lessons learned from 549 prospectively evaluated patients. Archives of Otolaryngology—Head & Neck Surgery, 126, 329–335. Ziedonis, D. M., & Williams, J. M. (2003). Management of smoking in people with psychiatric disorders. Current Opinion in Psychiatry, 16, 305–315.