If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Corresponding author. Group Health Research Institute, 1730 Minor Ave., Suite 1600, Seattle, WA 98101, USA. Tel.: +1 206 287 2737; fax: +1 206 287 2871.
Treatment outcomes were compared across smokers enrolled in the COMPASS cessation trial with (positive psychiatric history [PH+], n = 271) and without (PH−, n = 271) a diagnosis of PH based on medical record evidence of anxiety, depression, psychotic disorder, or bipolar disorder. Everyone received behavioral counseling plus varenicline and was followed for 6 months post quit date. PH+ smokers took varenicline for fewer days on average (59.4 vs. 68.5, p ≤ .01) but did not differ in their use of behavioral treatment. PH+ smokers were more likely to report anxiety and depression, but side-effect intensity ratings did not differ after adjusting for multiple comparisons. Overall, all side effects were rated as moderate intensity or less. Groups had similar 30-day abstinence rates at 6 months (31.5% PH+ vs. 35.4% PH−, p = .35). In sum, having a psychiatric diagnosis in this trial did not predict worse treatment outcome or worse treatment side effects.
U.S. Department of Health and Human Services, Substance Abise and Mental Health Services Administration, Center for Mental Health Services, National Institutes of Health, National Institute of Mental Health,
Rockville, MD1999
). Providing effective cessation treatment to these individuals is important, but there are limited data on the effectiveness of cessation treatments among persons with these conditions (
). This is particularly true for interventions involving varenicline.
Varenicline (aka, Chantix) is the latest smoking cessation medication approved by the U.S. Food and Drug Administration (FDA). It significantly increases abstinence rates compared to placebo, bupropion, and NRT (
Varenicline, an alpha4beta2 nicotinic acetylcholine receptor partial agonist, vs sustained-release bupropion and placebo for smoking cessation: A randomized controlled trial.
Journal of the American Medical Association.2006; 296: 47-55
Efficacy of varenicline, an alpha4beta2 nicotinic acetylcholine receptor partial agonist, vs placebo or sustained-release bupropion for smoking cessation: A randomized controlled trial.
Journal of the American Medical Association.2006; 296: 56-63
Smoking cessation with varenicline, a selective alpha4beta2 nicotinic receptor partial agonist: Results from a 7-week, randomized, placebo- and bupropion-controlled trial with 1-year follow-up.
Archives of Internal Medicine.2006; 166: 1561-1568
). Increased neuropsychiatric symptoms such as depressed mood, agitation, and suicidal ideation and behavior have been reported. As a result, the FDA added a black box warning in 2009 to alert physicians and patients to these risks. Persons with a psychiatric history (PH) might be particularly vulnerable to these side effects (
), but empirical evidence is limited. It is also unclear whether varenicline is equally effective among persons with and without PH. If this medication is associated with worse side effects among those with psychiatric histories, it may also be less effective. If not, cessation outcomes might not differ significantly.
This study examined the association between PH and smoking cessation in a randomized effectiveness trial of three behavioral intervention programs. Smokers received one of three forms of cognitive–behavioral intervention (telephone-based counseling, Web-based treatment, or a combination of telephone and Web-based treatment) plus a standard 12-week course of varenicline. Additional detail about each behavioral program is reported separately (
Swan, G. E., McClure, J. B., Zbikowski, S., McAfee, T., Jack, L., Catz, S., et al. (in press). Behavioral counseling and varenicline treatment for smoking cessation American Journal of Preventive Medicine.
Treating tobacco use and dependence: 2008 update. Clinical practice guideline. U.S. Department of Health and Human Services. Public Health Service,
Rockville, MD2008
). In a previous secondary data analysis, we reported that smokers with a likely history of depression based on a brief, single-item self-report screen had similar short-term treatment outcomes to smokers who did not screen positive for likely prior depression (
); however, depression screeners may not accurately reflect prior clinical depression and we were unable to comment on participants' other psychiatric conditions. The current article extends our prior report by examining long-term treatment outcome; relying on documented diagnoses of depression, anxiety, psychosis, or bipolar disorder based on medical chart review; and examining process variables that could account for cessation differences (e.g., frequency and intensity of treatment utilization). Given our prior findings and the lower lifetime quit rates among psychiatric populations, it was unclear whether smokers with a history of psychiatric illness would have lower abstinence rates after receiving a combination of behavioral intervention and varenicline. However, based on FDA warnings, we hypothesized these individuals would report worse treatment side effects. We also hypothesized they would be less adherent to treatment as a result of these side effects.
2. Methods
2.1 Setting
This research was conducted by Group Health (GH), SRI International (SRI), and Free & Clear, Inc. (F&C). GH is a large, nonprofit, consumer-governed health plan in the United States. SRI is a nonprofit independent research organization, and F&C is a for-profit vendor of tobacco quitline services. All study activities were approved by the appropriate institutional review boards. Study enrollment began in October 2006 and completed in October 2007.
2.2 Recruitment, intervention, and sample procedures
Adult smokers (n = 1,202) were recruited from GH to participate in a randomized effectiveness trial called the COMPASS study. Participants were randomized to one of three behavioral treatment programs (telephone-based counseling, Web-based intervention, or combined telephone and Web-based intervention), and all received a standard 12-week course of varenicline. Holding the medication constant across all three conditions allowed the independent effects of the behavioral treatment programs to be evaluated, which was the primary intent of COMPASS (
Swan, G. E., McClure, J. B., Zbikowski, S., McAfee, T., Jack, L., Catz, S., et al. (in press). Behavioral counseling and varenicline treatment for smoking cessation American Journal of Preventive Medicine.
). This article reports on a secondary analysis of the data examining differential outcomes by prerandomization PH.
Prior PH was determined from participants' medical records based on the International Classification of Diseases, Ninth Revision, Clinical Modification codes in the 5 years preceding the date of study enrollment. Conditions thought to be most relevant to either smoking cessation or varenicline adverse events were examined and included anxiety disorders (e.g., panic disorder, generalized anxiety disorder, phobic disorders), major depression (single or recurrent), bipolar disorder, and psychotic disorders (e.g., schizophrenia, schizoaffective disorder, paranoid psychosis). For ease of interpretation, diagnoses were collapsed into five categories: anxiety, depression, bipolar disorder, or psychotic disorder and a sixth category of any psychiatric diagnosis.
To compare participants with and without a documented PH, the sample was restricted to persons who had been enrolled in GH for at least 5 years prior to study enrollment (n = 542). The timeframe was chosen to allow adequate time for significant psychiatric diagnoses to be documented. Persons with less than 5 years of GH enrollment (n = 660) were excluded because persons enrolled for a shorter period may have had a prior psychiatric diagnosis, which was not documented in the medical chart. Therefore, including these individuals could bias the sample of smokers thought not to have a mental health history. Conditions occurring more than 5 years prior with no evidence of recent diagnosis or treatment were considered less relevant as prognostic indicators and therefore less biasing. The derivation of the final analytic samples is presented in Fig. 1. Coincidentally, the sample included an equal number of participants with (PH+, n = 271) and without (PH−, n = 271) a prior psychiatric diagnosis.
COMPASS participants were screened for eligibility by telephone. Individuals were eligible if they were 18 years or older, smoked 10 or more cigarettes per day over the past year and 5 or more cigarettes per day over the past week, were ready to quit smoking, could read and speak English, had telephone and Internet access, and were a member of GH. Individuals were excluded if they self-reported medical contraindications for varenicline use, poor health, severe heart disease, chronic obstructive pulmonary disease, current dialysis treatment or kidney disease, high-frequency alcohol intake or binge drinking, current use of other cessation treatment, and current use of investigational or recreational/street drugs or drugs thought to potentially interfere with renal clearance of varenicline. Study enrollment preceded FDA warnings about the use of varenicline in psychiatrically vulnerable populations. Persons who self-reported a diagnosis of or treatment for schizophrenia, bipolar disorder, or mania were excluded from the trial, but diagnostic interviews and medical record review were not conducted to screen out individuals with a PH. As a result, the final study sample included individuals with these psychiatric diagnoses who did not self-disclose their history at intake.
2.4 Assessment
Enrolled participants were interviewed by telephone at baseline and 21 days, 3 months, and 6 months post target quit date. Baseline assessment included demographics, smoking history, nicotine dependence assessed via the Fagerström Test of Nicotine Dependence (FTND;
). This measure evaluates depressive symptoms (disturbed sleep, low energy, blaming self, feeling blue, feeling everything is an effort, hopelessness about the future). Respondents rated how much they were bothered by each symptom in the past month on a 4-point scale ranging from not at all to extremely. Mean score was calculated. Scores on this symptom screen have been found to be inversely predictive of smoking cessation (
At follow-up, participants reported their current smoking status, varenicline use, and presence/intensity of common side effects associated with nicotine withdrawal and/or varenicline use. Use of the behavioral treatment programs was monitored using automated telephonic service delivery records and data captured during each visit to the online program. Nonsmoking was defined as a self-report of not smoking, even a puff, for the last 7 days (point prevalent abstinence [PPA]) and the last 30 days at each follow-up. For each potential treatment side effect, participants endorsed whether they had experienced it in the prior 30 days. If so, symptom severity was rated on a 5-point Likert scale ranging from very mild to very severe.
2.5 Data analyses
As would be expected clinically, PH groups were not mutually exclusive. Some participants had more than one diagnosis (see footnote in Table 1). In addition, only a small number of participants had diagnoses of bipolar disorder (n = 11) or a psychotic disorder (n = 4) because self-report with these disorders was an exclusion criteria in the COMPASS trial. Consequently, these individuals were included among those with any PH but not examined as separate diagnostic groups. In addition, diagnostic groups were not compared against one another due to the overlap in group membership. We considered comparing mutually exclusive subgroups (e.g., AH+ with no DH+ against DH+ with no AH+) against one another but ruled this out due to the added complexity of the presentation and interpretation of pairwise comparisons among subgroups and the minimal marginal gain from the clinical relevance of these findings. Rather, persons with a positive PH (PH+), depression history (DH+), or anxiety history (AH+) were compared against people without a prior psychiatric diagnosis (PH−).
Note. Each diagnostic group was compared against persons with no psychiatric diagnosis (PH−). AH+ and DH+ groups were not compared to one another due to participant overlap. Ninety-three people had concurrent diagnoses of depression and anxiety, 10 had concurrent diagnoses of depression and bipolar disorder, 7 had concurrent diagnoses of anxiety and bipolar disorder, 4 had concurrent diagnoses of depression and a psychotic disorder, 3 had concurrent diagnoses of anxiety and a psychotic disorder, and 1 had a concurrent diagnosis of bipolar disorder and a psychotic disorder.
The primary outcome of interest was abstinence at 6-month follow-up. Abstinence results are reported using an intent-to-treat (ITT) methodology where missing participants were conservatively counted as smokers. Secondary outcomes included days to relapse, treatment utilization (an indicator of adherence), and the presence and intensity of treatment side effects.
Based on previous findings, we were unsure whether PH groups would have differential abstinence rates but hypothesized that PH+ participants would have greater side effects and would be less adherent to behavioral and pharmaceutical treatment. Unadjusted group differences were examined using chi-square analyses for categorical variables and t tests for continuous variables. Logistic regression models were run for unadjusted and adjusted analyses of binary outcomes. Generalized linear models were run for adjusted analyses of Likert-scale variables (e.g., side-effect intensity scores). Adjusted analyses controlled for treatment group and baseline differences in marital status and gender. We did not adjust for baseline differences in depression score because depression is in the causal pathway between psychiatric diagnosis and treatment outcome. Finally, we applied the Benjamini–Hochberg (BH) procedure (
) to control for false discovery across multiple comparisons of treatment side-effect outcomes. The procedure was applied at the 5% False Discovery Rate level to comparisons within six sets of comparisons: (a) the proportion of AH+ versus PH− reporting symptoms, (b) the proportion of DH+ versus PH− reporting symptoms, (c) the proportion of PH+ versus PH− reporting symptoms, (d) average symptom intensity rating of AH+ versus PH−, (e) average symptom rating of DH+ versus PH−, and (f) the average symptom rating of PH+ versus PH−. The use of this procedure assures that in each of these six sets of comparisons, we expect that no more than 5% of the symptoms that are identified as statistically significant by the BH procedure are false-positives.
3. Results
3.1 Participants
Participant characteristics at baseline are presented in Table 1. Half (n = 271) were PH+, 23.6% were AH+, and 43.4% were DH+. PH+ smokers were more likely to be female, less likely to be married, and had higher baseline depression scores. Groups were otherwise similar.
An equal number were randomized to each behavioral treatment group.
3.2 Abstinence and time to relapse
ITT PPA rates at 6-month follow-up by PH group are indicated in Table 2. Abstinence rates did not differ between PH groups at 6-month follow-up (7-day PPA PH+ 32.5% vs. PH− 39.5%, odds ratio [OR] = 1.36, confidence interval [CI] = 0.95–1.93; 30-day PPA PH+ 31.0% vs. PH− 35.4%, OR = 1.22, CI = 0.85–1.75). Outcomes were not significantly changed in analyses adjusting for treatment group and baseline differences in gender and marital status (7-day PPA OR = 0.77, CI = 0.53–1.11; 30-day PPA OR = 1.18, CI = 0.81–1.72).
Table 2Abstinence at 6-month follow-up by baseline PH
Abstinence
Anxiety (AH+)
Depression (DH+)
Any psychiatric diagnosis (PH+)
No psychiatric diagnosis (PH−)
7-Day PPA (%)
33.6
32.8
32.5
39.5
30-Day PPA (%)
31.3
31.5
31.0
35.4
Note. Each diagnostic group was compared against persons with no psychiatric diagnosis. No comparisons were statistically significant in unadjusted or adjusted comparisons.
Groups also did not differ in their self-reported time to relapse (Table 3). PH+ smokers relapsed after 51.1 days on average, compared to 59.5 days for persons with no psychiatric diagnosis (p = .23). Results were unchanged after controlling for treatment group, gender, and marital status (p = .24).
Table 3Days to relapse and treatment utilization by PH
p ≤ .01 comparison to persons with no psychiatric diagnosis (PH−) in analyses that are unadjusted for covariates or multiple comparisons. Results of all comparisons were unchanged controlling for treatment group and baseline differences in gender and marital status.
p ≤ .01 comparison to persons with no psychiatric diagnosis (PH−) in analyses that are unadjusted for covariates or multiple comparisons. Results of all comparisons were unchanged controlling for treatment group and baseline differences in gender and marital status.
p ≤ .01 comparison to persons with no psychiatric diagnosis (PH−) in analyses that are unadjusted for covariates or multiple comparisons. Results of all comparisons were unchanged controlling for treatment group and baseline differences in gender and marital status.
Total number of proactive and reactive telephone calls completed. Maximum proactive calls (i.e., counselor initiated calls to participants) attempted to be completed was five and was limited to participants in telephone and telephone + Web treatment groups. Additional participant initiated (i.e., reactive) calls allowed among all participants.
Total number of log-ins to online treatment program. Applicable to participants in Web and telephone + Web groups.
4.7 (9.9)
3.7 (9.2)
3.9 (9.3)
5.6 (21.0)
Duration of telephone counseling (total minutes)
46.0 (4.1)
47.6 (3.0)
47.1 (2.8)
43.4 (2.4)
Duration Web treatment (total minutes)
94.3 (1.1)
46.7 (9.2)
65.6 (0.7)
81.8 (1.6)
a Total number of proactive and reactive telephone calls completed. Maximum proactive calls (i.e., counselor initiated calls to participants) attempted to be completed was five and was limited to participants in telephone and telephone + Web treatment groups. Additional participant initiated (i.e., reactive) calls allowed among all participants.
b Total number of log-ins to online treatment program. Applicable to participants in Web and telephone + Web groups.
p ≤ .01 comparison to persons with no psychiatric diagnosis (PH−) in analyses that are unadjusted for covariates or multiple comparisons. Results of all comparisons were unchanged controlling for treatment group and baseline differences in gender and marital status.
Psychiatric groups significantly differed in the mean duration of their varenicline use compared to persons with no psychiatric diagnosis (Table 3). Persons with no PH reportedly used varenicline the longest, an average of 68.5 days compared to 59.4 days for smokers with any of the psychiatric diagnoses of interest (p ≤ .01). Persons with anxiety or depression diagnoses reported using varenicline for significantly fewer days than smokers with no psychiatric diagnosis. All group comparisons were still statistically significant after controlling for treatment group, gender, and marital status (p ≤ .01).
On the contrary, there were no indications that persons with a psychiatric diagnosis differed in their use of behavioral counseling services. Compared to persons with no prior PH, each psychiatric group exhibited a similar total number of counseling calls completed, total Web log-ins, and similar overall duration of counseling by telephone (in minutes) or time spent logged into the Web program (Table 3). Results were unchanged in analyses adjusting for treatment group and baseline differences in gender and marital status.
3.4 Treatment side effects
Self-reported side effects were evaluated at 21-day follow-up (corresponding to approximately 1-month post varenicline initiation). An equal proportion of smokers with (n = 235) and without (n = 235) a PH endorsed experiencing side effects in the prior month. Smokers with a PH were more likely to report tension/agitation, depression, anxiety, confusion, nausea, retching, and difficulty concentrating in both unadjusted analyses and analyses controlling for treatment group and baseline differences in gender and marital status; however, after applying the BH procedure to control for multiple comparisons, none of the comparisons were statistically significant. Comparisons significant at the p ≤ .01 level in adjusted analyses are indicated in Table 4 for interested readers.
Table 4Proportion of participants reporting each symptom at 21-day follow-up
p ≤ .01 compared to persons with no psychiatric diagnosis after adjusting for treatment group, gender, and marital status. However, no comparisons were statistically significant after applying the BH procedure to control for multiple comparisons.
p ≤ .01 compared to persons with no psychiatric diagnosis after adjusting for treatment group, gender, and marital status. However, no comparisons were statistically significant after applying the BH procedure to control for multiple comparisons.
p ≤ .01 compared to persons with no psychiatric diagnosis after adjusting for treatment group, gender, and marital status. However, no comparisons were statistically significant after applying the BH procedure to control for multiple comparisons.
a Self-reported symptoms experienced in the past month, regardless of intensity.
b Symptom commonly associated with varenicline use.
c Symptom commonly associated with nicotine withdrawal.
p ≤ .01 compared to persons with no psychiatric diagnosis after adjusting for treatment group, gender, and marital status. However, no comparisons were statistically significant after applying the BH procedure to control for multiple comparisons.
Disparities in symptom intensity were significant for depression, anxiety, and constipation in adjusted analyses. However, after controlling for multiple comparisons, only the average symptom rating for constipation was significant between AH+ and PH− smokers. Mean ratings and items significant at the p ≤ .01 level are indicated in Table 5.
Table 5Average symptom rating among those reporting each symptom at 21-day follow-up
p ≤ .01 compared to persons with no psychiatric diagnosis after adjusting for treatment group, gender, and marital status. However, only constipation (AH+ vs. PH−) remained significant after applying the BH procedure to control for multiple comparisons.
p ≤ .01 compared to persons with no psychiatric diagnosis after adjusting for treatment group, gender, and marital status. However, only constipation (AH+ vs. PH−) remained significant after applying the BH procedure to control for multiple comparisons.
p ≤ .01 compared to persons with no psychiatric diagnosis after adjusting for treatment group, gender, and marital status. However, only constipation (AH+ vs. PH−) remained significant after applying the BH procedure to control for multiple comparisons.
p ≤ .01 compared to persons with no psychiatric diagnosis after adjusting for treatment group, gender, and marital status. However, only constipation (AH+ vs. PH−) remained significant after applying the BH procedure to control for multiple comparisons.
p ≤ .01 compared to persons with no psychiatric diagnosis after adjusting for treatment group, gender, and marital status. However, only constipation (AH+ vs. PH−) remained significant after applying the BH procedure to control for multiple comparisons.
Symptom commonly associated with nicotine withdrawal.
2.5 (1.0)
2.6 (0.8)
2.5 (0.9)
2.3 (0.8)
Confusionc
2.1 (0.8)
2.0 (0.8)
2.0 (0.8)
2.1 (0.9)
a Self-reported symptoms experienced in the past month. Symptoms rated on 5-point Likert scale from very mild to very severe.
b Symptom commonly associated with varenicline use.
c Symptom commonly associated with nicotine withdrawal.
p ≤ .01 compared to persons with no psychiatric diagnosis after adjusting for treatment group, gender, and marital status. However, only constipation (AH+ vs. PH−) remained significant after applying the BH procedure to control for multiple comparisons.
Having a prior psychiatric diagnosis was not associated with worse outcomes in the COMPASS trial. As a group, PH+ smokers were equally likely to be abstinent at 6-month follow-up as persons with no documented PH. This is despite PH+ participants' reports that they used less varenicline. The results were the same for persons with a diagnosis of anxiety (AH+), depression (DH+), or any of the monitored psychiatric conditions (PH+; anxiety, depression, bipolar disorder, or psychosis). It is also noteworthy that smokers with a preexisting psychiatric diagnosis were no more likely to report treatment side effects, and the only significant difference in side-effect intensity ratings was for constipation. Differences in affect-related symptoms were not significant after controlling for multiple comparisons.
In epidemiological trials, persons with PH have lower lifetime quit rates (
), suggesting they may have a more difficult time quitting smoking than persons with no PH, although this has not been borne out in some clinical cessation trials, at least for depression (
). The lack of a significant difference in cessation rates in the COMPASS trial might be due to the groups' use of behavioral services. AH+, DH+, and PH+ smokers each utilized the telephone and Web counseling services in an equivalent manner to smokers with no documented PH (PH−). One goal of the behavioral interventions was to encourage participants to remain focused on their abstinence goals in spite of medication side effects or nicotine withdrawal symptoms. These results underscore the Clinical Practice Guideline for Treating Tobacco Dependence recommendation that all smokers receive a combination of behavioral and pharmacological treatment, when medically appropriate (
Treating tobacco use and dependence: 2008 update. Clinical practice guideline. U.S. Department of Health and Human Services. Public Health Service,
Rockville, MD2008
). Finally, although there was no main effect for PH on cessation, it is possible that PH could interact with other individual characteristics such as age or gender to influence outcome. This possibility is worthy of future study.
This study's findings are significant for several reasons. Not only are they consistent with prior meta-analyses showing that a positive history of depression does not decrease one's odds of quitting smoking in clinical trials (
), but they also extend this finding to smokers with other psychiatric diagnoses. Almost no smoking cessation clinical trials have been conducted among smokers with anxiety disorders (
), so these data in particular provide valuable insight into the effectiveness of behavioral counseling plus varenicline in this population. However, readers are cautioned not to draw definitive conclusions about this treatment's effectiveness in populations with bipolar or psychotic disorders. Because we attempted to screen out individuals with these diagnoses from the COMPASS trial, the resulting subsamples were too small to analyze independently.
These data also add to the growing evidence base on the safety of varenicline. FDA warnings associate varenicline use with potential neuropsychiatric side effects such as depressed mood, agitation, and suicidal ideation and behavior (
). We did not systematically and proactively assess suicidal ideation, but we did monitor changes in mood and agitation during the active varenicline treatment phase. Although one might expect PH+ smokers to experience more and more intense side-effect symptoms, particularly affect-related symptoms, this was not the case. It is also worth noting that none of the symptoms were rated as more than moderately severe on average. These findings do not undermine the FDA concerns about varenicline but should provide some reassurance to providers who prescribe this drug to patients with preexisting depression or anxiety diagnoses. However, we cannot comment on differential treatment side effects among participants with psychosis or bipolar disorder versus those with no PH. Case reports have raised the possibility that varenicline may increase psychiatric symptoms among some of these patients (
), whereas other research has found no change in psychopathology among patients with psychosis taking varenicline when compared to persons with similar psychiatric diagnoses not taking this drug (
). Until more definitive, well-controlled clinical trials have been conducted to examine this specific issue, it is best to use caution when prescribing varenicline to anyone with psychosis or mania.
This study has a number of noteworthy strengths including the fact that treatment side effects were systematically assessed at follow-up during the active treatment phase, behavioral treatment utilization was based on automated service records, psychiatric diagnosis was based on participants' actual medical records and not brief screening tools or self-report, and the sample was restricted to persons with at least 5 years of medical records, thereby ensuring that significant psychiatric conditions warranting treatment were identified. At the same time, several limitations should be noted. First, this is a secondary analysis of data from a randomized clinical trial designed to compare the effectiveness of three behavioral treatments. As such, all participants received varenicline. We did not have a no-medication placebo control and therefore treatment side effects cannot be attributed to varenicline specifically. In addition, abstinence was not biochemically verified at follow-up and medication use was based on self-report.
In sum, these data provide sound evidence that having a psychiatric diagnosis, particularly anxiety or depression, does not necessarily portend worse outcomes for smokers treated with varenicline and behavioral intervention. Future research is needed to inform the safety and efficacy of varenicline in other psychiatric populations. In the meantime, clinicians should continue to carefully monitor all patients taking varenicline for adverse effects, but there is no indication that the combination of behavioral counseling and varenicline will necessarily be less effective in persons with a prior psychiatric diagnosis.
Acknowledgments
The authors wish to thank the study staff at SRI and GH for their assistance with data collection; the staff of F&C, for their assistance in recruiting and counseling participants; and Sallie Dacey, MD, and Abigail Halperin, MD, for their assistance with medical oversight. Special thanks to Chester Pabiniak, MS, for his programming assistance. The COMPASS trial was funded by the National Cancer Institute (NCI; RO1 071358, G. Swan PI). Pfizer, Inc., donated medication and provided nominal financial support to SRI to help support data collection at that site. However, neither NCI nor Pfizer was involved in the collection or analysis of these data, had any influence on the study design, or had any influence over the content of this report. GH provided in-kind pharmacy support of this research. This study is registered at Clinicaltrials.gov (NCT00301145).
Dr. McAfee, Dr. Zbikowski, and Ms. Deprey are employed by F&C. Dr. Swan received financial support from Pfizer to attend a 1-day advisory meeting in 2008. The authors have no other potential conflicts of interest to report.
Treating tobacco use and dependence: 2008 update. Clinical practice guideline. U.S. Department of Health and Human Services. Public Health Service,
Rockville, MD2008
Varenicline, an alpha4beta2 nicotinic acetylcholine receptor partial agonist, vs sustained-release bupropion and placebo for smoking cessation: A randomized controlled trial.
Journal of the American Medical Association.2006; 296: 47-55
Efficacy of varenicline, an alpha4beta2 nicotinic acetylcholine receptor partial agonist, vs placebo or sustained-release bupropion for smoking cessation: A randomized controlled trial.
Journal of the American Medical Association.2006; 296: 56-63
Smoking cessation with varenicline, a selective alpha4beta2 nicotinic receptor partial agonist: Results from a 7-week, randomized, placebo- and bupropion-controlled trial with 1-year follow-up.
Archives of Internal Medicine.2006; 166: 1561-1568
Swan, G. E., McClure, J. B., Zbikowski, S., McAfee, T., Jack, L., Catz, S., et al. (in press). Behavioral counseling and varenicline treatment for smoking cessation American Journal of Preventive Medicine.
U.S. Department of Health and Human Services, Substance Abise and Mental Health Services Administration, Center for Mental Health Services, National Institutes of Health, National Institute of Mental Health,
Rockville, MD1999
p < .05, not adjusted for multiple comparisons.
00068-1&id=gr1.jpg){kind=link}