Bupropion Sustained Release Added to Group Support for Smoking Cessation in Schizophrenia: A New Randomized Trial and a Meta-Analysis

Bupropion Sustained Release Added to Group Support for Smoking Cessation in Schizophrenia: A New Randomized Trial and a Meta-Analysis

Objective: To clarify the efficacy and tolerability of bupropion sustained release (SR) for the treatment of cigarette smoking in people with schizophrenia.

Method: The first study is a double-blind, placebo-controlled clinical trial with 32 outpatients from the Maryland Psychiatric Research Center. From May 2003 to July 2007, clinically stable people with a DSM-IV diagnosis of schizophrenia or schizoaffective disorder who smoked at least 10 cigarettes per day and who were interested in quitting smoking or cutting down were recruited for participation. All participated in a 9-week support group and were randomly assigned to receive 12 weeks of bupropion SR or placebo. The primary outcome measure was 4 weeks’ sustained abstinence over the last 4 study weeks. Secondary outcome measures included decrease in smoking behavior and change in symptoms, neuropsychological performance, and side effects. In the second study, we performed an electronic literature search of MEDLINE in September 2008. Articles in English published between 2003 and 2008 were searched for the terms schizophrenia, bupropion SR, and smoking. Bibliographies of studies identified through the MEDLINE search were also examined. Case reports, open-label studies, crossover studies, and studies using nonstandard dosing of bupropion SR were excluded. In this way, 4 studies similar in methodology to the currently presented clinical trial were identified and the individual data combined in a meta-analysis. A random effects meta-analysis using Comprehensive Meta-Analysis software was used to obtain a pooled estimate of the odds ratio for 4-week smoking abstinence between bupropion SR and placebo.

Results: There were no significant results on the primary or secondary smoking measures for the clinical trial, although a numeric advantage favored the bupropion SR group. There were no significant findings for secondary symptom or side effect measures and no significant change in neuropsychological performance. For the meta-analysis totaling 226 subjects, there were significant findings in favor of bupropion SR. The pooled estimate of the odds ratio for 4-week abstinence was 2.7 (95% CI, 1.3 to 5.7; P = .009), and clinically significant greater smoking reduction in the bupropion SR group, with pooled difference estimates increasing over time between groups, became statistically significant by week 5 of study medication (P < .02).

Conclusions: New clinical trial data and a meta-analysis strongly support the tolerability and efficacy of bupropion SR for the treatment of cigarette smoking in people with schizophrenia

Trial Registration: clinicaltrials.gov Identifier: NCT00176449

J Clin Psychiatry 2012;73(1):95-102

Submitted: March 28, 2010; accepted August 3, 2010.

Online ahead of print: April 5, 2011 (doi:10.4088/JCP.10m06143gre).

Corresponding author: Elaine Weiner, MD, Maryland Psychiatric Research Center, PO Box 21247, Baltimore, MD 21228 (eweiner@mprc.umaryland.edu).

The rate of tobacco smoking among people with schizophrenia is up to 3 times greater than in the general population.1 Yet there has been a lack of aggressive treatment for tobacco dependence, even at a time when the general population was being bombarded with media messages and regulatory pressure to quit smoking. Today, physicians2 and other treatment providers of the chronically mentally ill3 are still less likely to refer their patients for smoking cessation treatment than are those who provide care for patients in the general population. The reasons for this lack of aggressiveness are very likely multifaceted. Many providers view smoking as a lower priority problem compared to psychosis, and many believe that smokers with schizophrenia cannot tolerate cessation attempts from a psychiatric standpoint, seeing smoking as an attempt at self-medication, with cessation leading to worsening of symptoms or side effects.4,5 Even when patients have tolerated smoking bans on inpatient units, mental health workers have not been reassured.6 Unfortunately, without support for abstinence on discharge, most patients resumed smoking on discharge.7 In the early 1990s, the research literature began to provide evidence that people with schizophrenia can stop smoking8-10 and, most importantly, that cessation attempts were not associated with worsening in psychiatric status.11 Subsequently, a number of studies have examined various combinations of bupropion sustained release (SR), nicotine replacement therapy, and support groups to treat nicotine dependence in this population.12-15 More recent studies have found that bupropion SR helps smokers achieve short-term cessation as compared to those randomized to placebo.16-19

The present report includes 2 studies of the efficacy of bupropion SR for smoking cessation in people with schizophrenia. The first is a double-blind, placebo-controlled follow-up study of a previously reported open-label study13 and the second a meta-analysis merging these new data with 4 previously reported placebo-controlled, parallel-group studies. Particularly as new agents become available for smoking cessation, it is critically important to have a better sense of the efficacy and tolerability of bupropion in people with schizophrenia, so that a more informed decision can be made when choosing from available smoking cessation treatments.

Study I (Clinical Trial): Method


Clinically stable outpatients from the Maryland Psychiatric Research Center, Baltimore, Maryland, volunteered for participation and were screened for study eligibility. Subjects had a Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) diagnosis of schizophrenia or schizoaffective disorder made through a best-estimate diagnostic approach utilizing information from the Structured Clinical Interview for DSM-IV,20 direct assessment, family informants, and past records. Subjects were interested in decreasing their smoking, smoked at least 10 cigarettes per day, and had a total score of at least 4 (moderate) out of a possible 10 (severe) on the Fagerstrom Test for Nicotine Dependency (FTND).21 Participants were excluded if they had a neurologic diagnosis, current major depressive episode, substance dependence (other than nicotine) in the last 6 months or substance abuse (excluding nicotine) in the last 3 months, if they were not medically stable, or if they were currently taking bupropion.

The University of Maryland School of Medicine Institutional Review Board approved the study protocol. Written informed consent was obtained after study procedures were explained, and the informed consent process was documented. Participants were told at consent that the goal was to stop smoking, but they were encouraged to continue participation even if they could not achieve complete cessation. The trial was registered at clinicaltrials.gov (NCT00176449).

Study Design

Random assignments made by the statistician were stratified by sex and use of first- versus second-generation antipsychotic medication, as these factors have been shown to influence the likelihood of successfully quitting.22-26 A power analysis based on our previous open-label study, which had detected a 44% reduction in end-expired carbon monoxide (CO) levels with bupropion treatment, suggested that 10 subjects in each group would be needed to detect the same effect. Because open-label studies tend to inflate observed effects, the recruitment goal was for 20 subjects in each group. The study design included an evaluation phase to ensure clinical stability, including standard laboratory measures, clinical behavioral and symptom assessments, and smoking measures. Subsequently, all participants entered the 14-week treatment phase beginning with the 9-week Group Support Program led by staff trained in the educational model of the American Cancer Society Fresh Start Program modified for people with schizophrenia.8 Each session was structured and incorporated relaxation exercises with practice "homework." The first group sessions were designed to increase awareness of specific smoking habits and to develop a "Quit Plan." A Quit Day Ceremony was held at the fifth group session, 2 weeks after the initiation of study medication. Subsequent sessions focused on reworking the Quit Plan. Later groups focused on strategies for subjects’ minimizing weight gain, managing "high risk" situations, and imagining themselves as nonsmokers. Subjects started study medication (bupropion SR 150 mg or placebo) the evening after the third group session (treatment phase week 2), with dosing once daily for 3 days and then twice daily for the remainder of the study. There was flexibility in dosing so that, if needed, study drug could be decreased to once per day. Nicotine gum was offered to all participants on Quit Day, but none of the participants chose to use the gum. Following the conclusion of the Group Support Program, participants remained on study medication therapy for 6 weeks, until the end of study. Medication compliance was assessed by weekly pill count review, with 75% compliance judged acceptable.

Study Assessments

Smoking measures. The primary efficacy outcome measure was sustained abstinence, defined as end-expired CO levels less than 10 ppm at the last 4 study visits of the treatment phrase.27 Secondary efficacy measures included point prevalence abstinence, defined by CO levels less than 10 ppm at a specific time point; change in CO levels between baseline and each time point; reduction in urine levels of the nicotine metabolite cotinine28; and change in FTND total score from baseline to end of study. End-expired CO levels were collected weekly, and cotinine and FTND measures were collected at evaluation phase and treatment phase weeks 2, 4, 8, and 14. End-expired CO levels were measured by having participants exhale forcefully and completely into a Smokerlyzer breath CO monitor (Bedfont Scientific Ltd, Maidstone, United Kingdom) at the same time each week to ensure a steady-state lung CO level. End-expired CO levels measured this way have been shown to be a valid measure of smoking status in individuals who have achieved steady state.27,28

Safety measures. Clinical measures. To study whether bupropion or withdrawal from nicotine might affect psychiatric symptoms, assessments were obtained at evaluation phase and treatment phase weeks 2, 4, 8, and 14. The Brief Psychiatric Rating Scale (BPRS)29 was performed to assess positive symptoms, depression, and anxiety, and the Scale for the Assessment of Negative Symptoms (SANS)30 was used to assess negative symptoms.

Neuropsychological measures. A neuropsychological battery was designed to assess whether decreased smoking might impair cognitive function, since nicotine may improve cognition in people with schizophrenia.31,32 The battery, administered in the evaluation phase and treatment phase week 14, consisted of the following: the Rey Auditory Verbal Learning Test,33 the Brief Visual-Spatial Memory Test34 to measure verbal and visual memory; the Wechsler Adult Intelligence Scale, Third Edition,35 Letter Number Sequencing test to measure verbal working memory and Digit Symbol-Coding task to measure processing speed; the Gordon Diagnostic System Continuous Performance Task36 to measure attention; and the Grooved Peg Board37 to measure manual dexterity.

Adverse effect measures. The Simpson-Angus Scale (SAS)38 was used to assess for the emergence of motor side effects possibly occurring as a result of an increase in antipsychotic blood levels caused by decreased cigarette consumption. These levels were obtained at evaluation phase and end of study. Other common side effects were assessed using the Side Effect Checklist (SEC), an unpublished 22-item list (developed by R.W.B.) in which subjects rate their complaints on a 1 (none) to 4 (severe) scale. Assessments were obtained during evaluation phase and treatment phase weeks 2, 4, 8, and 14. (Copies of the SEC are available from the corresponding author upon request.)

Statistical Analyses

Smoking measures. Abstinence at each weekly visit was defined by an end-expired CO level measurement < 10 ppm. For analysis of sustained abstinence, participants who dropped out early or had missing data were imputed as not abstaining. Fisher exact test measured differences in the percentage of participants who achieved sustained abstinence at the end of the double-blind treatment phase. For the secondary analyses, all observations collected during double-blind treatment were used in an intention-to-treat analyses; imputation was not used for missed visits following dropouts. In these analyses, the baseline visit was week 2 (the last visit prior to the start of study medication), and a mixed model for analysis of covariance (ANCOVA) for unbalanced repeated measures was used. Post hoc t tests based on estimates and standard errors from the general mixed-model ANCOVAs were performed to determine whether there were significant differences between groups at each study milestone. Point prevalence of smoking cessation was analyzed using the generalized estimating equation method for repeated-measures logistic regression.39

Safety measures. Mixed-model ANCOVAs were used to assess group differences on symptom measures.40 For each neuropsychological test, participant scores were converted to z scores (z = [score—baseline mean]/baseline SD). An overall composite z score was computed from the mean of the individual test z scores.41

To test for group differences in extrapyramidal symptoms (EPS), a rank correlation score was calculated between SAS score and visit. Then the distribution of these scores was compared using a rank test selective for sensitivity to the presence of differences between groups. Rank correlations were also used to examine the relationship between end-expired CO levels and antipsychotic blood levels. For the SEC, Fisher exact test was used to compare groups on the percentage of patients who had any new incidence or worsening of a side effect during double-blind treatment.


Forty-six participants were randomized, and 32 completed the 14-week treatment phase (see Consolidated Standards of Reporting Trials flowchart, Figure 1). While the target completion number was 40, there was insufficient study drug available to meet this goal. Of the 32 who completed the study, 5 left before beginning double-blind treatment and are not included in the analyses. The 30% dropout rate is within the range seen in similar studies. There were no group differences on any demographic characteristics. There were no significant differences on severity of positive or negative symptoms, degree of dependency (as measured by the FTND or end-expired CO level), or the percentage of participants taking first- or second-generation antipsychotic medications or clozapine (Table 1). Participants all achieved the goal of 75% compliance.

Figure 1

Click figure to enlarge

Table 1

Click figure to enlarge

Smoking Measures

The rates of sustained abstinence were 4 of 22 (18%) for bupropion participants and 2 of 19 (11%) for placebo participants (Fisher exact test, P = .67). Because of the low rates of achievement on this measure, no further analysis was done.

With regard to the secondary smoking measures, weekly point prevalence cessation rates tended to favor numerically the bupropion group at most visits during the double-blind treatment phase (Figure 2). However, the repeated-measures logistic regression analyses suggested no overall treatment effect (χ21 = 1.10, P = .29), no overall time effect (χ211 = 14.55, P = .20), and no evidence that time trends varied between the groups (χ211 = 7.41, P = .77). Similarly, scores of the other secondary measures showed a numerical advantage for the bupropion group over placebo, especially after Quit Day, but no statistically significant group differences were observed (Table 2 and Figure 2). Specifically, when the change in end-expired CO levels between the start of the double-blind treatment phase and each treatment phase week was compared between groups, mixed-model ANCOVA revealed no significant overall treatment effect of bupropion (F1,37 = 0.38, P = .54),) and no evidence that time trends varied between the groups (F11,310 = 1.52, P = .12). There was a significant time effect for both groups (F11,310 = 4.16, P = .001), showing a decrease in end-expired CO levels for both groups over time. For the FTND total score, the mixed-model ANCOVA revealed no significant group differences in mean nicotine dependency (F1,30 = 2.08, P = .16), but there was a significant time effect (F2,52 = 7.39, P < .001). The treatment-by-time effect was not significant (F2,52 = 2.33, P = .11). For urine cotinine, there were no significant group differences in mean urine cotinine levels (F1,24.2 = 2.51, P = .13), but there was a significant main effect of time (F2,23.4 = 3.61, P = .043). The treatment-by-time effect was not significant (F2,23.4 = 0.35, P = .71).

Figure 2

Click figure to enlarge

Table 2

Click figure to enlarge

Table 2r

Safety Measures

Clinical measures. There were no significant between-group differences in the BPRS positive symptom item (conceptual disorganization, suspiciousness, unusual thought content, and hallucinatory behavior) (F1,25.9 = 1.19; P = .29), BPRS anxiety/depression factor (F1,30.6 = 0.22; P = .64), or SANS total scores (F1,29.9 = 1.11; P = .30).

Neuropsychological measures. The change in the overall composite z score was not significantly different between groups (Wilcoxon test P = .34), and the heterogeneity of effect analysis was also not significant (χ24 = 4.66, P = .32), which suggests that there were no differential effects of bupropion on any of the individual measures. Therefore, no follow-up analyses were done for the individual test scores.

Adverse events/side effects. Five of the 6 participants who dropped out from the bupropion group complained of adverse events or side effects. Of those 5, 2 dropped-out at week 1 with a complaint of restlessness/anxiety, and another who was quite ill at baseline and often had mild exacerbations complained of worsening of psychosis. One participant was terminated from the study after 2 weeks of study medication due to the development of a rash that resolved after study medication was discontinued. Another participant, with a history of polydipsia, had a seizure at week 13 and was found to be hyponatremic. The remaining participant discontinued his participation at week 11 due to reported loss of motivation to stop smoking. Of the 3 participants randomized to placebo who discontinued, 2 participants dropped out at week 4: 1 secondary to loss of motivation and 1 due to worsening anxiety and restlessness. The third participant dropped out at week 5, with nonspecific complaints of sedation and malaise.

When the side effect data were analyzed, there were no group differences on any of the SEC measures, including the common bupropion side effects of restlessness, insomnia, dry mouth, and sedation. There was a numeric increase in dry mouth in the bupropion group. With respect to motor side effects, SAS total score means were low at baseline in both groups and remained low throughout the study, most likely due to the high percentage of subjects taking novel antipsychotics. Among 8 subjects (4 taking bupropion, 4 taking placebo) treated with olanzapine who had drug levels at both evaluation phase and end of study, the Spearman rank correlation between change in end-expired CO levels and change in olanzapine level from evaluation phase to end of study was R = −0.78, P = .023. Of these subjects, 6 had no change in reported level of sedation. One subject with a small reduction in olanzapine level reported a 1-point increase in sedation; a second, whose olanzapine level dropped by 24, had a 1-point reduction in sedation. Notably, none of the subjects whose olanzapine levels were increased at EOS reported a change in level of sedation.


The present study shows a numeric but not a statistically significant benefit of bupropion for smoking cessation in schizophrenia. A small but consistent benefit of bupropion is seen across the secondary smoking measures, but none of the group differences reached statistical significance, possibly due to lack of power. Viewed graphically (Figure 2), it appears that the bupropion group response was generally better than the placebo group. Importantly, we replicated previous studies that showed bupropion can be used safely in this population. Despite individual cases of bupropion-related side effects, there were no significant group differences, including those that might have emerged from an elevation in antipsychotic blood levels, as subjects decreased their smoking. There was no statistically significant worsening on any of the clinical measures, although 1 subject of the 6 randomized to bupropion withdrew due to worsening of psychiatric symptoms. There was no worsening of cognitive function, despite a decrease in cigarette consumption. Finally, neither a beneficial nor a detrimental effect was observed for negative symptoms. The relatively low level of negative symptoms in this study cohort may have precluded our being able to replicate previous reports of a beneficial effect of bupropion for these symptoms.13

There are several possible explanations for our inability to detect a statistical advantage for the bupropion group. Most obvious is the problem of attempting to measure a low-probability outcome (smoking cessation) in a relatively small number of study participants. This confound was further complicated by the large variability in the outcomes measured. Group differences were further obscured by the fact that both groups responded to the psychosocial treatment.

Study II (Meta-Analysis): Method

Using our data, we performed a meta-analysis to obtain a more definitive estimate of bupropion’s therapeutic effect. In September 2008, we performed a MEDLINE search for articles in English published between 2003 and 2008 that included the terms schizophrenia, bupropion SR, and smoking, excluding case reports, open-label studies, crossover studies, and studies using nonstandard bupropion dosing. Bibliographies of studies identified through the MEDLINE search were also examined. We identified 4 studies similar to our own presented here.16-19 All were double-blind, placebo-controlled; used standard dosing; offered group support; and used end-expired CO levels as the smoking outcome measure. The duration of administration of study medications varied, with 12 weeks in the Evins et al17,18 and Weiner et al (present) studies, and 9 weeks in the George et al16,19 studies. The Evins et al (2007)18 and George et al (2008)19 studies also made use of nicotine replacement therapy, whereas the Evins et al (2005),17 George et al (2002),16 and the present study did not. Together, these studies included a total of 119 participants randomized to bupropion and 121 randomized to placebo. Participant retention rates varied among studies from 73% to 81%.

Statistical Analyses

Individual participant data, provided by George and A.E.E., were reanalyzed together with the data from the present study in a meta-analysis to examine the percentage of participants achieving sustained abstinence and to examine week-by-week mean differences in end-expired CO levels between treatment groups. Performing this secondary analysis allowed us to utilize all of the participant data, not just data from those who completed a study.

Comprehensive Meta-Analysis software,42 version 1 (Biostat, Englewood, New Jersey), was used to obtain a pooled estimate of the odds ratio for sustained abstinence between bupropion SR and placebo and to estimate week-by-week mean differences in end-expired CO levels between groups. Dropouts and subjects with missed visits were classified as smokers.

A Rosenthal’s estimate was calculated to estimate the number of negative studies it would take to cancel a statistically significant result.43


Study-specific odds of successful 4-week abstinence at the end of treatment with bupropion versus placebo ranged from 1.7 to 11.9. The pooled estimate of the odds ratio for 4-week abstinence was 2.7 (95% CI, 1.3-5.7; P = .009), indicating that bupropion SR treatment is associated with a higher probability than placebo of successful abstinence from smoking. The Rosenthal’s formula43 estimated that it would require 8.9 studies whose mean odds ratio was 1.0 to raise the P value for the combined odds ratio to a nonsignificant level.

In a meta-analysis of ANCOVA-adjusted week-by-week treatment differences in end-expired CO levels (Figure 3), the pooled difference estimates increased over time between participants taking bupropion SR versus placebo, with differences that were statistically significant (P < .02, unadjusted for multiple comparisons) by week 5 of double-blind treatment.

Figure 3

Click figure to enlarge


The meta-analysis results provide strong evidence for the efficacy of bupropion for the treatment of cigarette smoking in people with schizophrenia, and they provide a much more precise estimate of the effect of bupropion SR when combined with group support for smoking cessation in schizophrenia.

All of these clinical trials, including our study, have been hampered by the difficulty of studying a low-probability outcome in relatively low-powered studies. Although the other studies were able to show a significant advantage for the bupropion group, performing the meta-analysis and thereby incorporating our current study data helps to provide greater clarity as to the magnitude of the benefit. In addition, incorporating our data with the other studies helps to address the "file drawer" problem of ignoring nonsignificant (usually not published) studies, a notorious obstacle to meta-analysis and accurate estimation of the true efficacy and effectiveness of interventions. Thus, this meta-analysis provides a more reliable estimate of the odds ratio for effectiveness than any of the individual studies, revealing that bupropion may help those with schizophrenia to be over 21/2 times more likely to achieve sustained smoking cessation than those taking placebo.

Although the actual number of people with schizophrenia who are able to successfully quit smoking with bupropion is numerically smaller than that seen in the general population, the relative magnitude of the bupropion effect is possibly greater than the benefit observed in the general population.44,45 In addition, the meta-analysis demonstrated clearly that participants taking bupropion had a decrease in their end-expired CO levels over time, implying that, even if participants were unable to quit, they were able to reduce their smoking to a greater extent than those subjects randomized to placebo. These findings provide clear support for using bupropion for treating smoking addiction in people with schizophrenia. Although we were not able to examine the possibility of psychiatric worsening due to either nicotine withdrawal or as a side effect of bupropion in the meta-analysis, the studies used in the meta-analysis are in agreement that, although there may be individual cases of psychiatric worsening, there is no evidence of a statistical difference in the likelihood of psychiatric worsening in subjects randomized to bupropion compared to those randomized to placebo. If anything, there is suggestive evidence from our open-label study,13 Evins et al,16 and George et al17 that bupropion may improve negative symptoms. Similarly, with respect to side effects, although there may be individual cases of the common bupropion side effects, significant differences do not appear between the bupropion and placebo groups in double-blind studies.

The overall impression from both the current study and our pilot study13 is that group support has an important effect on the ability of participants to reduce and quit smoking, although we did not examine this question directly. In both the pilot study and the clinical trial presented above, the study design included 2 weeks of group intervention before initiating study medications and 3 weeks of study medication following the termination of the group intervention. In both studies, participants decreased their cigarette consumption even before study medications began, and the gains made were lessened after the group intervention ended. These observations suggest that support is a meaningful component for people with schizophrenia in their fight against smoking addiction. A meta-analysis reviewing group support in the general population literature demonstrates that group support, compared to self-help or no group, approximately doubles the odds of smoking cessation.46 Unfortunately, group-oriented smoking cessation treatment is not easily available to people with schizophrenia living in the community. In a study with psychotic patients, there was evidence of a dose-dependent relationship between the amount of psychosocial intervention and the likelihood of successful cessation.47 Further research is necessary to clarify which components of the support are essential and whether group support provides any additional benefit compared to individual support, which is more easily incorporated into various treatment settings. In this area, the general population meta-analysis provides little direction, as there are neither data to support the use of any particular components of group intervention nor sufficient data with which to compare group intervention to individual counseling of similar intensity.46

In conclusion, there are compelling scientific data to support the use of bupropion SR for the treatment of nicotine addiction in schizophrenia, and evidence-based practice guidelines are supporting its use.48 Recently, there have been a number of studies supporting the use of pharmacologic combination therapies to increase the odds of quitting in medically ill smokers in primary care settings,49,50 with one study providing evidence for the use of triple combination therapy.51 There have been no similar studies in psychiatric patients, despite calls for more work to address the needs of this under treated, high risk population.52,53 Therefore, if one assumes that data from the general population literature are applicable to patients with schizophrenia, although a less robust response might be seen in the latter group, the evidence to date would suggest a multipronged approach to aggressively treat dependence on tobacco-derived nicotine or tobacco dependence in people with schizophrenia. Further studies should focus on a more aggressive nicotine-dependence treatment approach for smokers with schizophrenia.

Drug names: bupropion (Zyban), clozapine (Clozaril, FazaClo, and others), olanzapine (Zyprexa).

Author affiliations: The Department of Psychiatry, Maryland Psychiatric Research Center, University of Maryland School of Medicine, Baltimore (Drs Weiner, Gold, McMahon, and Buchanan and Mss Ball and Buchholz); and the Addiction Research Program, Massachusetts General Hospital, Harvard Medical School, Boston (Dr Evins).

Potential conflicts of interest: Ms Ball has served as a consultant to ePharmaSolutions and Pfizer; Dr Gold has served as a consultant to Merck, AstraZeneca, Solvay, Pfizer, and GlaxoSmithKline. Dr Evins has served as a consultant to Pfizer, Boehringer, and Schering Plough and has received grant/research support from GlaxoSmthKline and Pfizer. Dr Buchanan has served as a consultant to Abbott and GlaxoSmithKline; has received grant/research support from Novartis and Janssen; has served on advisory boards for AstraZeneca, Wyeth, Schering Plough, Solvay, and Pfizer; and has received other material or financial support from Bristol-Myers Squibb, Otsuka, Pfizer, and Cephalon. Drs Weiner and McMahon and Ms Buchholz report no financial or other relationship relevant to the subject of this article.

Funding/support: Funded in part by the Veterans Affairs Capitol Network (VISN 5) Mental Illness Research, Education, and Clinical Center and National Institute of Mental Health Grant MH068580-01, Advance Center for Intervention Services Research.

Previous presentation: Presented as a poster at the 11th International Congress of Schizophrenia Research; March 26-April 1, 2007; Colorado Springs, Colorado.


1. de Leon J, Diaz FJ. A meta-analysis of worldwide studies demonstrates an association between schizophrenia and tobacco smoking behaviors. Schizophr Res. 2005;76(2-3):135-157. PubMed doi:10.1016/j.schres.2005.02.010

2. Druss BG, Bradford WD, Rosenheck RA, et al. Quality of medical care and excess mortality in older patients with mental disorders. Arch Gen Psychiatry. 2001;58(6):565-572. PubMed doi:10.1001/archpsyc.58.6.565

3. Himelhoch S, Daumit G. To whom do psychiatrists offer smoking-cessation counseling? Am J Psychiatry. 2003;160(12):2228-2230. PubMed doi:10.1176/appi.ajp.160.12.2228

4. Lohr JB, Flynn K. Smoking and schizophrenia. Schizophr Res. 1992;8(2):93-102. PubMed doi:10.1016/0920-9964(92)90024-Y

5. Dalack GW, Meador-Woodruff JH. Smoking, smoking withdrawal and schizophrenia: case reports and a review of the literature. Schizophr Res. 1996;22(2):133-141. PubMed doi:10.1016/S0920-9964(96)80441-5

6. Smith CM, Pristach CA, Cartagena M. Obligatory cessation of smoking by psychiatric inpatients. Psychiatr Serv. 1999;50(1):91-94. PubMed

7. El-Guebaly N, Cathcart J, Currie S, et al. Public health and therapeutic aspects of smoking bans in mental health and addiction settings. Psychiatr Serv. 2002;53(12):1617-1622. PubMed doi:10.1176/appi.ps.53.12.1617

8. Addington J. Group treatment for smoking cessation among persons with schizophrenia. Psychiatr Serv. 1998;49(7):925-928. PubMed

9. Addington J, el-Guebaly N, Campbell W, et al. Smoking cessation treatment for patients with schizophrenia. Am J Psychiatry. 1998;155(7):974-976. PubMed

10. Hughes JR, Frances RJ. How to help psychiatric patients stop smoking. Psychiatr Serv. 1995;46(5):435-436. PubMed

11. Ziedonis DM, George TP. Schizophrenia and nicotine use: report of a pilot smoking cessation program and review of neurobiological and clinical issues. Schizophr Bull. 1997;23(2):247-254. PubMed

12. Ziedonis DM, Williams JM, Smelson D. Serious mental illness and tobacco addiction: a model program to address this common but neglected issue. Am J Med Sci. 2003;326(4):223-230. PubMed doi:10.1097/00000441-200310000-00014

13. Weiner E, Ball MP, Summerfelt A, et al. Effects of sustained-release bupropion and supportive group therapy on cigarette consumption in patients with schizophrenia. Am J Psychiatry. 2001;158(4):635-637. PubMed doi:10.1176/appi.ajp.158.4.635

14. Evins AE, Mays VK, Rigotti NA, et al. A pilot trial of bupropion added to cognitive behavioral therapy for smoking cessation in schizophrenia. Nicotine Tob Res. 2001;3(4):397-403. PubMed doi:10.1080/14622200110073920

15. George TP, Ziedonis DM, Feingold A, et al. Nicotine transdermal patch and atypical antipsychotic medications for smoking cessation in schizophrenia. Am J Psychiatry. 2000;157(11):1835-1842. PubMed doi:10.1176/appi.ajp.157.11.1835

16. George TP, Vessicchio JC, Termine A, et al. A placebo controlled trial of bupropion for smoking cessation in schizophrenia. Biol Psychiatry. 2002;52(1):53-61. PubMed doi:10.1016/S0006-3223(02)01339-2

17. Evins AE, Cather C, Deckersbach T, et al. A double-blind placebo-controlled trial of bupropion sustained-release for smoking cessation in schizophrenia. J Clin Psychopharmacol. 2005;25(3):218-225. PubMed doi:10.1097/01.jcp.0000162802.54076.18

18. Evins AE, Cather C, Culhane MA, et al. A 12-week double-blind, placebo-controlled study of bupropion sr added to high-dose dual nicotine replacement therapy for smoking cessation or reduction in schizophrenia. J Clin Psychopharmacol. 2007;27(4):380-386. PubMed doi:10.1097/01.jcp.0b013e3180ca86fa

19. George TP, Vessicchio JC, Sacco KA, et al. A placebo-controlled trial of bupropion combined with nicotine patch for smoking cessation in schizophrenia. Biol Psychiatry. 2008;63(11):1092-1096. PubMed doi:10.1016/j.biopsych.2007.11.002

20. First MB, Spitzer RL, Gibbon M, et al. Structured Clinical Interview for DSM-IV Axis I Disorders, Clinician Version (SCID-CV). Washington, DC: American Psychiatric Press; 1996.

21. Heatherton TF, Kozlowski LT, Frecker RC, et al. The Fagerström Test for Nicotine Dependence: a revision of the Fagerström Tolerance Questionnaire. Br J Addict. 1991;86(9):1119-1127. PubMed doi:10.1111/j.1360-0443.1991.tb01879.x

22. Shiffman S, Paton SM. Individual differences in smoking: gender and nicotine addiction. Nicotine Tob Res. 1999;1(suppl 2):S153-S157, discussion S165-S166. PubMed doi:10.1080/14622299050011991

23. Piper ME, Fox BJ, Welsch SK, et al. Gender and racial/ethnic differences in tobacco-dependence treatment: a commentary and research recommendations. Nicotine Tob Res. 2001;3(4):291-297. PubMed doi:10.1080/14622200110050448

24. Procyshyn RM, Ihsan N, Thompson D. A comparison of smoking behaviours between patients treated with clozapine and depot neuroleptics. Int Clin Psychopharmacol. 2001;16(5):291-294. PubMed doi:10.1097/00004850-200109000-00007

25. McEvoy J, Freudenreich O, McGee M, et al. Clozapine decreases smoking in patients with chronic schizophrenia. Biol Psychiatry. 1995;37(8):550-552. PubMed doi:10.1016/0006-3223(94)00365-A

26. George TP, Sernyak MJ, Ziedonis DM, et al. Effects of clozapine on smoking in chronic schizophrenic outpatients. J Clin Psychiatry. 1995;56(8):344-346. PubMed

27. Henningfield JE, Stitzer ML, Griffiths RR. Expired air carbon monoxide accumulation and elimination as a function of number of cigarettes smoked. Addict Behav. 1980;5(3):265-272. PubMed doi:10.1016/0306-4603(80)90049-0

28. Benowitz NL; SRNT Subcommittee on Biochemical Verification. Biochemical verification of tobacco use and cessation. Nicotine Tob Res. 2002;4(2):149-159. PubMed doi:10.1080/14622200210123581

29. Overall JE, Gorham DR. The Brief Psychiatric Rating Scale. Psychol Rep. 1962;10(3):799-812. doi:10.2466/PR0.10.3.799-812

30. Andreasen NC. Negative symptoms in schizophrenia: definition and reliability. Arch Gen Psychiatry. 1982;39(7):784-788. PubMed

31. Domier CP, Monterosso JR, Brody AL, et al. Effects of cigarette smoking and abstinence on Stroop task performance. Psychopharmacology (Berl). 2007;195(1):1-9. PubMed doi:10.1007/s00213-007-0869-x

32. Barr RS, Culhane MA, Jubelt LE, et al. The effects of transdermal nicotine on cognition in nonsmokers with schizophrenia and nonpsychiatric controls. Neuropsychopharmacology. 2008;33(3):480-490. PubMed doi:10.1038/sj.npp.1301423

33. Spreen O, Strauss E. A Compendium of Neuropsychological Tests. New York, NY: Oxford University Press; 1998.

34. Benedict R. Brief Visuospatial Memory Test—Revised. Odessa, FL: Psychological Assessment Resources, Inc.; 1997.

35. Wechsler D. Wechsler Adult Intelligence Scale. 3rd ed. San Antonio, TX: The Psychological Corporation; 1997.

36. Gordon M, Mettelman B. Technical Guide to the Gordon Diagnostic System. Dewitt, NY: Gordon Systems; 1987.

37. Matthews C, Klove H. Instruction Manual for the Adult Neuropsychological Battery. Madison, WI: University of Wisconsin Medical School; 1964.

38. Simpson GM, Angus JWS. A rating scale for extrapyramidal side effects. Acta Psychiatr Scand suppl. 1970;212:11-19. PubMed doi:10.1111/j.1600-0447.1970.tb02066.x

39. Mallinckrodt CH, Kaiser CJ, Watkin JG, et al. The effect of correlation structure on treatment contrasts estimated from incomplete clinical trial data with likelihood-based repeated measures compared with last observation carried forward ANOVA. Clin Trials. 2004;1(6):477-489. PubMed doi:10.1191/1740774504cn049oa

40. Liang KY, Zeger SL. Longitudinal data analysis using generalized linear models. Biometrika. 1986;73(1):13-22. doi:10.1093/biomet/73.1.13

41. Pepe MS, Whitaker RC, Seidel K. Estimating and comparing univariate associations with application to the prediction of adult obesity. Stat Med. 1999;18(2):163-173. PubMed doi:10.1002/(SICI)1097-0258(19990130)18:2<163::AID-SIM11>3.0.CO;2-F

42. Borenstein M, Hedges LV, Higgins J, et al. Comprehensive Meta-Analysis. Englewood, NJ: Biostat; 2000.

43. Rosenthal R. The "file drawer problem" and tolerance for null hypothesis. Psychol Bull. 1979;86(3):638-641. doi:10.1037/0033-2909.86.3.638

44. Ranney L, Melvin C, Lux L, et al. Systematic review: smoking cessation intervention strategies for adults and adults in special populations. Ann Intern Med. 2006;145(11):845-856. PubMed

45. Aubin HJ. Tolerability and safety of sustained-release bupropion in the management of smoking cessation. Drugs. 2002;62(suppl 2):45-52. PubMed doi:10.2165/00003495-200262002-00005

46. Stead LF, Lancaster T. Group behaviour therapy programmes for smoking cessation. Cochrane Database Syst Rev. 2002;(3):CD001007. PubMed

47. Baker A, Richmond R, Haile M, et al. A randomized controlled trial of a smoking cessation intervention among people with a psychotic disorder. Am J Psychiatry. 2006;163(11):1934-1942. PubMed doi:10.1176/appi.ajp.163.11.1934

48. Buchanan RW, Kreyenbuhl J, Kelly DL, et al; Schizophrenia Patient Outcomes Research Team (PORT). The 2009 schizophrenia PORT psychopharmacological treatment recommendations and summary statements. Schizophr Bull. 2010;36(1):71-93. PubMed doi:10.1093/schbul/sbp116

49. Piper ME, Smith SS, Schlam TR, et al. A randomized placebo-controlled clinical trial of 5 smoking cessation pharmacotherapies. Arch Gen Psychiatry. 2009;66(11):1253-1262. PubMed doi:10.1001/archgenpsychiatry.2009.142

50. Smith SS, McCarthy DE, Japuntich SJ, et al. Comparative effectiveness of 5 smoking cessation pharmacotherapies in primary care clinics. Arch Intern Med. 2009;169(22):2148-2155. PubMed doi:10.1001/archinternmed.2009.426

51. Steinberg MB, Greenhaus S, Schmelzer AC, et al. Triple-combination pharmacotherapy for medically ill smokers: a randomized trial. Ann Intern Med. 2009;150(7):447-454. PubMed

52. Kelly DL, McMahon RP, Wehring HJ, et al. Cigarette smoking and mortality risk in people with schizophrenia. Schizophr Bull. 2009. doi:10.1093/schbul/sbp152 PubMed

53. Ziedonis D, Hitsman B, Beckham JC, et al. Tobacco use and cessation in psychiatric disorders: National Institute of Mental Health report. Nicotine Tob Res. 2008;10(12):1691-1715. PubMed doi:10.1080/14622200802443569

Related Articles

Volume: 72

Quick Links: Schizophrenia and Schizoaffective Disorders


Buy this Article as a PDF