Original investigation. Laurie A. Zawertailo PhD 1,2, Dolly Baliunas PhD 1, Anna Ivanova MPH 1, Peter L. Selby MBBS 1,3,4. Abstract.

Transcription

1 Nicotine & Tobacco Research, 2015, doi: /ntr/ntv013 Original investigation Original investigation Individualized Treatment for Tobacco Dependence in Addictions Treatment Settings: The Role of Current Depressive Symptoms on Outcomes at 3 and 6 Months Laurie A. Zawertailo PhD 1,2, Dolly Baliunas PhD 1, Anna Ivanova MPH 1, Peter L. Selby MBBS 1,3,4 1 Nicotine Dependence Service, Centre for Addiction and Mental Health, Toronto, ON, Canada; 2 Department of Pharmacology and Toxicology, University of Toronto, Toronto, ON, Canada; 3 Department Family and Community Medicine, University of Toronto, Toronto, ON, Canada; 4 Dalla Lana School of Public Health, University of Toronto, Toronto, ON, Canada Corresponding Author: Laurie Zawertailo, PhD, Nicotine Dependence Service, Addictions Program, Centre for Addiction and Mental Health, 175 College Street, Room 212, Toronto, ON M5T1P7, Canada. Telephone: x77422; Fax: ; laurie.zawertailo@camh.ca Abstract Introduction: Individuals with concurrent tobacco dependence and other addictions often report symptoms of low mood and depression and as such may have more difficulty quitting smoking. We hypothesized that current symptoms of depression would be a significant predictor of quit success among a group of smokers receiving individualized treatment for tobacco dependence within addiction treatment settings. Methods: Individuals in treatment for other addictions were enrolled in a smoking cessation program involving brief behavioral counseling and individualized dosing of nicotine replacement therapy. The baseline assessment included the Patient Health Questionnaire (PHQ9) for depression. Smoking cessation outcomes were measured at 3 and 6 months post-enrollment. Bivariate associations between cessation outcomes and PHQ9 score were analyzed. Results: Of the 1,196 subjects enrolled to date, 1,171 (98%) completed the PHQ9. Moderate to severe depression (score >9) was reported by 28% of the sample, and another 29% reported mild depression (score between 5 and 9). Contrary to the extant literature and other findings by our own group, there was no association between current depression and cessation outcome at either 3 months (n = 1,171) (17.0% in those with PHQ9 > 9 vs. 19.8% in those with PHQ9 < 5, p =.32) or 6 months (n = 834) (17.8% vs. 18.9%, p =.74). Conclusions: Contrary to our hypothesis, depression severity as measured by the PHQ9 did not predict cessation outcome in this clinical population. A possible explanation may be the individualized treatment and supportive environment of an addictions treatment setting. These data indicate that patients in an addictions treatment setting can successfully quit smoking regardless of current depressive symptoms. Introduction Smoking remains the leading preventable cause of death in Canada and the United States, accounting for one in every five deaths each year. 1 Furthermore, over 40% of the approximately 70,000 people who access the addiction treatment system in Ontario, Canada in a given year use tobacco, 2 compared to 18% of the general Ontario The Author Published by Oxford University Press on behalf of the Society for Research on Nicotine and Tobacco. All rights reserved. For permissions, please journals.permissions@oup.com. 937

2 938 Nicotine & Tobacco Research, 2015, Vol. 17, No. 8 population. Despite this significant prevalence, treatment for tobacco dependence has been conspicuously absent from the addiction treatment system. Of those clients who enter addiction treatment programs, more will die from tobacco-related diseases than from all other causes combined, 3 and clients in addiction treatment programs who refrain from smoking have better overall outcomes than those who continue to smoke 4 6 although there is some conflicting research in this area. 7 Despite high client demand and need, most addiction agencies have been reluctant to provide chronic disease prevention and management services, such as reducing tobacco use. Mood management may be particularly important in this population during smoking cessation because, there is an association between smoking, alcohol and drug dependence, and depression. 8,9 Depressive symptoms may be an important predictor of cessation outcome among clients of addictions treatment facilities. In addition to the relationship between tobacco dependence and depressive disorders, sub-clinical symptoms of depression, including negative affect, anhedonia, sleep disturbances and irritability, are common in nondepressed individuals during a quit attempt. 10,11 One of the most common mental health conditions among smokers in general is a current or past history of depression and approximately 60% of individuals with a history of depression are either current or past smokers. 12 It is not known whether smoking initiation occurs prior to, or following the onset of depressive symptoms; however, genetics research indicates a shared etiology. 13 Those with a history of depression are more likely to progress to daily smoking, 14 and a history of daily smoking has also been shown to significantly increase the risk of major depression. 15 As well, depression has been cited as the most common antecedent prior to relapse during smoking cessation. 16 Only two published studies have specifically examined whether there are differential effects of smoking cessation strategies by depression status (i.e., history positive vs. current depression). One study did not show any difference in outcome by depression status 17 and the other study showed that mood management information improved cessation rates in history positive smokers but not currently depressed smokers. 18 While there have been no studies published to date on the role of current depression on smoking cessation outcomes in addictions patients, there is one published study of predictors of smoking cessation outcome among individuals with substance use and/or psychiatric disorders. 19 While those with a dual diagnosis had poorer cessation outcomes compared to those with a single disorder, this was not statistically significant and in a regression model was not a significant predictor of cessation. However, other studies suggest that depressed mood and anhedonia adversely affect smoking cessation and lead to relapse, 20,21 and a recent meta-analysis suggests that a history of depression and recurrent depression is associated with poorer cessation outcomes. 22 To what extent clinical beliefs that smoking cessation will increase psychological symptoms among addictions patients or jeopardize recovery from substance use play into the current situation whereby cessation is not routinely addressed in these settings is unknown. However, some research indicates that this can be a major barrier for implementing smoking cessation programs in drug treatment settings. 23,24 Even so, there is a strong rationale for treating tobacco dependence in addictions settings including the fact the tobacco dependence is itself an addictive disorder and as such addiction professionals already have the needed skills to treat tobacco dependence, tobacco dependence is almost universally comorbid with other substance use disorders, smoking cessation has been shown to improve treatment outcomes for other substance use disorders. 25 Therefore, in response to the need to integrate tobaccodependence treatment into addictions treatment facilities in Ontario, the Ministry of Health has funded a program since 2012 involving both smoking cessation training for staff and cost-free tobacco dependence treatment for addiction agency clients consisting of nicotine replacement therapy (NRT) and counseling support. In order to assess the real-world effectiveness of this program, patient-level data were collected at baseline and 3 and 6 months post-enrollment. While previous research in addictions treatment populations have indicated less success with smoking cessation treatment compared to nonaddicted populations, the reasons for this are unclear. Since patients in addictions treatment often have significant depressive symptoms and it is thought that these mood disturbances may contribute negatively to treatment outcomes, we were particularly interested in assessing baseline depressive symptoms in this population and whether these symptoms would predict smoking cessation treatment outcomes in patients seeking treatment at specialty addictions agencies. Therefore, we hypothesized that current symptoms of depression would be negatively associated with quit success among a group of smokers receiving individualized treatment for tobacco dependence (tailored doses of NRT and behavioral counseling) within outpatient addictions treatment settings. Methods Study Design This is an analysis of longitudinal data obtained from participants in a smoking cessation program implemented in 14 Addictions Agencies across the province of Ontario, Canada between October 2012 and November Data were obtained at four types of time-points: baseline or initial assessment, 3 months post-enrollment, 6 months post-enrollment, and at each clinic visit to receive treatment. The study was approved by the Research Ethics Board at the Centre for Addiction and Mental Health. Participants and Procedure The participants were clients of the participating addiction agencies. Participants were smokers interested in quitting smoking using NRT and were recruited to the program via either addiction practitioner referral or self-referral. They were daily dependent smokers as determined by cigarettes per day and time to first cigarette. Informed consent was obtained from all participants. Practitioners at each site underwent specific training in cessation counseling through the University of Toronto accredited TEACH training program. 26 The training focuses primarily on motivational interviewing techniques and provides specific strategies for counseling patients with psychiatric or addiction comorbidities. Practitioners typically met with clients every 2 weeks for brief cessation counseling and NRT if needed, although visit frequency was set by client needs and practitioner availability. Practitioners were instructed not to exceed more than 4 weeks between visits, and that clients were eligible to receive up to 26 weeks of NRT in total over a maximum of 1 year duration of treatment. The types of NRT available were: Patch (7 mg, 14 mg, and 21 mg); Gum (2 mg); Lozenge (2 mg); Inhaler (10 mg; 4 mg of nicotine delivered). Types and doses of NRT were individualized depending on severity of tobacco dependence and could be titrated up or down at each visit depending on response. Combination of patch and short-acting NRT or double patching was allowed. Practitioners were also trained to provide clients with individualized (brief) or group counseling during these visits.

3 Nicotine & Tobacco Research, 2015, Vol. 17, No Measures Baseline information was collected at enrollment visit, via participant self-report. Information on participant demographics, tobacco use characteristics, current alcohol and drug use, self-report lifetime history of selected medical and mental illness, and current depression severity were collected. Participant demographics included age, sex, annual household income (dichotomized), education, and employment status. Baseline tobacco use characteristics were number of cigarettes smoked per day and time to first cigarette. Current substance use was defined as use within the past 30 days and was asked about use of alcohol, marijuana, or other drugs which included cocaine, sedatives, opiates, stimulants, and hallucinogens. Medical comorbidities were identified as self-reported lifetime diagnosis, and included one or more of the following: high blood pressure, high cholesterol, heart problems, diabetes, asthma, chronic bronchitis/ emphysema/copd, seizures, cancer, or chronic pain. Reported past diagnosis of mental illness included the following: depression, anxiety, schizophrenia, or bipolar disorder. Current depression severity was assessed with the Patient Health Questionnaire (PHQ9), a 9-item instrument for making DSM-IV criteria-based diagnoses of depression that was developed specifically for use in primary care. It has been validated in two separate studies involving a total of 6,000 patients 27,28 and has been shown to be valid both as a diagnostic instrument for current depressive disorders and as a measure of depression severity along a continuum. 29 Typically, there are five categories for the PHQ9 (none to minimal depression [0 4], mild depression [5 9], moderate depression [10 14], moderately severe depression [15 19] and severe depression [20 27]). For this study we collapsed the three highest categories into a moderate to severe category (10+). The primary study outcome was smoking cessation, as defined by self-reported 7-day point prevalence of abstinence from smoking. 30 This was assessed at 3 and 6 months after enrollment in the program via or telephone surveys (by trained research staff). The follow-up completion rates were 61.0% and 58.7% for 3- and 6-month follow-ups, respectively. Since 7-day point prevalence abstinence from smoking was also asked at each clinic visit (to help clinicians gauge treatment progress), eligible visit form data when available, was used to impute missing responses for the main outcome. If there was more than one clinic visit during the 30 day time frame of eligibility for follow-up completion, the visit closest to the start date of eligibility was used. Those participants who did not have either a response to the follow up survey, or clinic visit data, were treated as lost to follow-up and categorized as not quit. Data Analysis All statistical analyses were performed in SPSS Version 20 (IBM). Baseline characteristics were examined for the entire sample, and also compared by PHQ9 score category using one-way analysis of variances (ANOVAs) for continuous variables and chi-square tests for categorical variables, to examine differences between subgroups. Bivariate associations between follow-up quit outcomes at 3 and 6 months and baseline demographic and clinical characteristics were examined using ANOVAs or t tests for continuous variables and chi-square tests for categorical variables. PHQ9 score category as a predictor of quitting at 3- and 6-month follow-up was assessed via binary logistic regressions. Multivariable models were adjusted for effects of all variables that were found to be significantly associated with both PHQ9 scores and quitting in bivariate analyses, as they were likely to be confounders. Variables that were significantly associated with only quitting were treated as potential confounders, and retained in the final model if their inclusion resulted in pronounced change of the main association between PHQ9 score category and quitting. The remaining demographic, substance use, and smoking characteristics collected at baseline (including discarded confounder candidate variables) were tested for effect modification of the main association. Moderators were tested by addition of their interaction term with PHQ9 score into the adjusted regression model one at a time, and observing any changes in the main association. Due to the low response rate to follow-up surveys, study attrition was examined via bivariate associations between baseline demographic variables and follow-up completion status, and also via Kaplan Meier survival analysis and a log rank test comparing treatment retention by baseline depression severity. All recorded visits were compiled for each participant and instances where participants were still eligible for treatment but stopped attending the clinic (i.e., dropped out), were classified as events. Instances where participants stopped attending the clinic but were outside of their window of eligibility were classified as censored. Kaplan-Meier estimates of the enrollment duration function were then calculated and plotted by baseline PHQ9 scores, and these distributions were then compared by depression score for significant differences. Results Sample Characteristics The study sample included 1,196 participants recruited between October 2012 and November PHQ9 scores were available for 97.9% of the study sample. Participants who did not answer three or more items on the questionnaire were classified as having missing scores. Nearly half of the study participants had none to minimal depression 41.1% (n = 491) (PHQ9 score 0 4), 28.8% (n = 345) had mild depression (score 5 9), 13.9% (n = 166) had moderate depression, 8.4% (n = 100) had moderately severe depression, and 5.8% (n = 69) had severe depression. Due to the small numbers of participants in each of the three highest depression score groups, participants with PHQ9 scores of 10 or greater were grouped into a category titled moderate to severe depression, which included 335 participants or 28.0% of the study sample. Demographic characteristics of the overall sample are presented in Table 1. Nearly two thirds of the participants were male, with a mean age of 46 years old. The majority of participants (80.3%) had an annual household income of $40,000 or less; more than half had some or completed post-secondary education (55.5%); more than one third of the sample was on disability (37.2%); one quarter of participants were full- or part-time employed (26.4%). Study participants smoked on average 21 cigarettes per day, 70.1% consumed alcohol but 24.2% were not currently using alcohol or any drugs. One third of the study sample did not have any medical comorbidities (32.0%), but one quarter had reported three or more health comorbidities (24.1%). Self-reported history of mental illness diagnosis was reported by 69.8% of the sample. Baseline participant characteristics were also compared by PHQ9 depression score categories (Table 1). Participants with higher depression scores were younger, had lower levels of household income, were more likely to be on disability, consumed more cigarettes per day, had a shorter time to first cigarette in the morning after waking up, used marijuana and other drugs, consumed more drinks per occasion, had a greater number of medical comorbidities, and were more likely to report a lifetime history of a mental illness.

4 940 Nicotine & Tobacco Research, 2015, Vol. 17, No. 8 Table 1. Baseline Participant Demographic and Clinical Characteristics Total sample, (n = 1,196) PHQ9 0 4, (n = 491) PHQ9 5 9, (n = 345) PHQ , (n = 335) p value a Demographic characteristics Age (years), mean (SD) (13.12) (13.19) (13.17) (12.56).001 Sex (male), n (%) 751 (62.8%) 318 (64.9%) 214 (62.4%) 203 (60.8%).47 Annual household income, n (%) <.001 $40, (80.3%) 292 (72.6%) 234 (81.5%) 259 (89.0%) >$40, (19.7%) 110 (27.4%) 53 (18.5%) 32 (11.0%) Education, n (%).23 Less than secondary 299 (25.6%) 116 (23.9%) 79 (23.3%) 99 (30.4%) High school diploma 221 (18.9%) 94 (19.4%) 64 (18.9%) 58 (17.8%) Some/complete post-secondary 649 (55.5%) 275 (56.7%) 196 (57.8%) 169 (51.8%) Employment status, n (%) <.001 Employed 304 (26.4%) 163 (34.5%) 94 (28.4%) 45 (13.8%) Unemployed 306 (26.6%) 116 (24.5%) 102 (30.8%) 85 (26.2%) Student/retired 113 (9.8%) 65 (13.7%) 19 (5.7%) 25 (7.7%) Disability 429 (37.2%) 129 (27.3%) 116 (35.0%) 170 (52.3%) Smoking characteristics Cigarettes per day, mean (SD) (11.95) (10.76) (11.74) (13.62) Time to first cigarette, n (%) <0.001 <5 min 562 (47.0%) 213 (46.1%) 146 (45.6%) 192 (60.2%) 6 30 min 394 (32.9%) 169 (36.6%) 121 (37.8%) 96 (30.1%) min 91 (7.6%) 37 (8.0%) 33 (10.3%) 18 (5.6%) >60 min 78 (6.5%) 43 (9.3%) 20 (6.2%) 13 (4.1%) Alcohol and drug use Current substance use, n (%) None 288 (24.2%) 124 (25.4%) 75 (21.9%) 77 (23.1)%.48 Alcohol 833 (70.1%) 346 (70.9%) 247 (72.0%) 230 (69.3%).73 Marijuana 242 (20.2%) 69 (14.1%) 71 (20.6%) 97 (29.0%) <.001 Other drugs b 164 (13.7%) 40 (8.1%) 46 (13.3%) 75 (22.4%) <.001 Past year alcohol use, n (%).62 Never 355 (29.9%) 142 (29.1%) 96 (28.0%) 102 (30.7%) Less than once a month to once 380 (32.0%) 168 (34.4%) 107 (31.2%) 101 (30.4%) a week 2 3 times a week or more 453 (38.1%) 178 (36.5%) 140 (40.8%) 129 (38.9%) Drinks per occasion, n (%) (43.1%) 224 (46.6%) 147 (30.6%) 110 (22.9%) 6 or more 293 (26.0%) 101 (34.9%) 89 (30.8%) 99 (34.3 Does not drink 348 (30.9%) 138 (41.4%) 93 (27.9%) 102 (30.6%) Health comorbidities Medical comorbidities c, n (%) < (32.0%) 188 (38.4%) 116 (33.7%) 74 (22.2%) (27.3%) 143 (29.2%) 95 (27.6%) 82 (24.6%) (16.7%) 70 (14.3%) 59 (17.2%) 64 (19.2%) 3 or more 286 (24.1%) 89 (18.2%) 74 (21.5%) 113 (33.9%) Self-reported past diagnosis of mental illness d, n (%) 829 (69.8%) 258 (53.0%) 263 (76.5%) 291 (87.1%) <.001 Note. PHQ9 = Patient Health Questionnaire. Sample sizes vary due to missing data (pairwise deletion). a p values for analysis of variance or chi-square tests of association. b Other drug use includes one or more of cocaine, sedatives, opiates, stimulants, hallucinogens. c Medical comorbidities include high blood pressure, high cholesterol, heart problems, diabetes, asthma, chronic bronchitis/emphysema/chronic obstructive pulmonary disease (COPD), seizures, cancer, chronic pain. d Past diagnosis of mental illness includes one or more of depression, anxiety, schizophrenia, bipolar disorder. Non-Response and Study Retention The 3-month follow-up response rate was 61.0% with only followup surveys included, and increased to 67.7% when 7-day point prevalence of abstinence data from eligible visit forms were added. The 6-month follow-up response rate was 58.7% with only follow-up surveys included, and increased to 61.8% when eligible visit form cessation outcomes were added. Checks into response attrition bias (not shown) found that participants with missing 3-month survey and visit form data were more likely to be younger (p <.001), male (p <.001), have lower income (p =.03), have high school education (p =.002), be unemployed (p <.001), use any substance (p <.001), use alcohol (p <.001), use other substances besides alcohol, marijuana and tobacco (p <.001), and have fewer comorbidities (p <.001), but did not differ from responders on any of the smoking characteristics, marijuana use, or history of mental illness. At 6-month follow-up, participants with missing survey and visit form

5 Nicotine & Tobacco Research, 2015, Vol. 17, No data were more likely to be younger (p <.001), male (p =.009), have only high school education (p =.002), be unemployed (p <.001), use any substances (p =.005), use alcohol (p <.001), and have fewer comorbidities (p <.001), but did not differ from responders on household income, any smoking characteristics, history of marijuana use, history of other substance use, or history of mental illness. To determine whether depression severity was associated with treatment retention, survival curves for treatment retention were plotted and compared by baseline depression scores via log rank test (Figure 1). There was no significant difference in attrition between PHQ9 depression score groups (p =.33). All groups saw a large drop in retention following the first visit. Retention continued to decrease until the 26-week mark (end of current study follow-up period). Overall, at 26 weeks after study enrolment, participants had a cumulative probability of remaining in the study. Depression and Quit Outcomes at 3 and 6 Months Bivariate associations between PHQ9 scores, other baseline characteristics, and quit outcomes at 3 and 6 months are presented in Table 2. PHQ9 scores were not significantly associated with quitting at either the 3- or 6-month follow-up (both p >.05). The other bivariate associations between baseline participant characteristics and follow-up quit outcomes are outlined in Table 2. Characteristics significantly associated with quitting at both 3 and 6 months post-enrollment included age, employment status, frequency of and amount of alcohol use, and number of medical comorbidities (all p <.05). Annual household income and education were significantly associated with quitting only at 3 months but not at 6 months (both p <.05). Table 3 presents the odds of quitting for those with mild or moderate depression and for those with severe depression, compared to those with none or minimal depression. There was no significant association between depression score category and quitting at either 3- or 6-month outcomes in the unadjusted analyses. There was also no significant association between depression score category and Figure 1. Treatment retention survival curves by baseline depression scores. Survival curves for treatment retention by baseline depression scores via log rank test. There was no significant difference found in attrition by baseline Patient Health Questionnaire (PHQ9) depression scores (p =.33) and therefore baseline depression severity was not significantly associated with treatment retention. quitting in a model adjusted for age, income, employment status, consumption of six or more drinks per occasion, and number of medical comorbidities for 3-month quit outcome. There was also no significant association between depression and quitting in a model adjusted for age, employment status, consumption of five or more drinks per occasion, and number of medical comorbidities for 6-month quit outcome. Finally, we tested whether the main association between depression scores and quit success was modifiable by any of the baseline demographic, smoking, or clinical characteristics. We found that none of the baseline characteristic interaction terms were significant predictors in the adjusted regression model, or affected the main association. We further stratified the adjusted regression analyses by the levels of each categorical moderator variable to examine the main association within specific study sample sub-groups, and also failed to find any significant effect of depression on quitting at either 3 or 6 months in any of the sub-groups. Discussion The findings of this current study indicate that high depression ratings at baseline (as measured by a validated self-complete questionnaire, the PHQ9), had no effect on smoking cessation outcomes at 3 and 6 months after initiating cessation treatment involving individualized NRT and behavioral counseling within addictions settings in unadjusted and adjusted analyses. Lack of a found association between depression severity and cessation outcome was likely not due to an inadequate sample size. Detecting a significant difference at p <.05 and 80% power with our found quit rates would have required a sample size of 3,006 for 3-month follow-up outcomes and 19,437 for 6-month outcomes. When we compared the characteristics of study participants to those of individuals receiving substance abuse treatment in Ontario during the fiscal year, 2 there were similarities in the proportion of males and alcohol use. However, there were notable differences between our study sample and the Ontario substance abuse treatment sample in the levels of education (25.6% of our study sample vs. 45.2% treatment population had less than secondary education and 55.5% of our study sample vs. 29.6% of the treatment sample had some or completed post-secondary education). Our study sample also had a lower rate of unemployed (26.6% vs. 38.1%) and a greater proportion of participants on disability (37.2% vs. 14.6%). Our study sample also had a greater proportion of participants with a lifetime history of mental illness (69.8% vs. 42.2%). Finally, 45.6% of Ontario substance abuse treatment patients reported having used tobacco in the past 12 months, and 24.2% had tobacco use as the presenting problem. These data underscore the importance of implementing specific programs and practitioner training for smoking cessation within addictions treatment settings. While there have been no studies published to date on the role of current depression on smoking cessation outcomes in addictions patients, there is one published study of predictors of smoking cessation outcome among individuals with substance use and/or psychiatric disorders. 19 Those with a dual diagnosis had poorer cessation outcomes compared to those with a single disorder, but this was not statistically significant and in a regression model was not a significant predictor of cessation. In studies of individuals recruited from the general population and screened for other disorders, current depression did not adversely affect cessation outcomes. For example, in smokers with either current or past depression, successful

6 942 Nicotine & Tobacco Research, 2015, Vol. 17, No. 8 Table 2. Quit Outcomes at 3 and 6 Months by Baseline Depression, Demographic and Clinical Characteristics Quit outcome 3-month follow-up, N = 1,196 6-month follow-up, N = 846 N quit % quit p value a N quit % quit p value a Baseline depression ratings PHQ9 scores None to minimal depression % % Mild depression % % Moderate to severe depression % % PHQ9 Major Depressive Syndrome criteria satisfied Yes % % No % % Self-reported past diagnosis of depression Yes % % No % % Baseline demographic characteristics Age (mean years) <.001 <.001 Quit Not quit Gender Male % % Female % % Annual household income $40, % % >$40, % % Education Less than secondary % % High school diploma % % Some or completed post-secondary % % Employment status Employed % % Unemployed % % Student/retired % % Disability % % Baseline smoking characteristics Cigarettes per day (mean) Quit Not quit Time to first cigarette <5 min % % 6 30 min % % min % % >60 min % % Baseline alcohol and drug use Current marijuana use Yes % % No % % Current use of other drugs b Yes % % No % % Current alcohol use.015 <.001 Yes % % No % % Frequency of alcohol use (past year).018 <.001 Never % % Less than once a month to once a week % % 2 3 times a week or more % % Drinks per occasion.007 < % % 6 or more % % Does not drink % %

7 Nicotine & Tobacco Research, 2015, Vol. 17, No Table 2. Continued Quit outcome 3-month follow-up, N = 1,196 6-month follow-up, N = 846 N quit % quit p value a N quit % quit p value a Baseline health comorbidities Number of medical comorbidities c % % % % % % 3 or more % % Self-reported past diagnosis of mental illness d Yes % % No % % Note. Quit defined as having stopped smoking intentionally for at least 7 days. Imputed outcomes presented. a p values for t test or chi-square tests of association. b Other drug use includes one or more of cocaine, sedatives, opiates, stimulants, hallucinogens. c Comorbidities include high blood pressure, high cholesterol, heart problems, diabetes, asthma, chronic bronchitis/emphysema/copd, seizures, cancer, chronic pain. d Past diagnosis of mental illness includes one or more of depression, anxiety, schizophrenia, bipolar disorder. Table 3. Odds of Quitting at 3 and 6 Months, Predicted by Baseline Patient Health Questionnaire (PHQ9) Score quitting when treated with bupropion was unrelated to baseline level of depression. 31 In another study, smokers with concurrent depression were able to achieve long-term cessation with cessation rates in the intervention group almost double that of the control group. 32 However, other studies suggest that depressed mood and anhedonia adversely affect smoking cessation and lead to relapse, 20,21 and a recent meta-analysis suggests that a history of depression and recurrent depression is associated with poorer cessation outcomes. 23 A possible explanation for the lack of association between current depression and smoking cessation outcomes in the current study may be that with the individualized treatment given, both the tailored doses of NRT and the behavioral supportive counseling mitigated the role of depressed mood on cessation attempts. Treatment providers in addictions settings are well-trained in mood management techniques including cognitive behavioral therapy (CBT). While we did not evaluate this specifically in the 14 study sites, it may have contributed to our outcomes. Two clinical trials on smoking cessation have shown that smokers with recurrent major depressive disorder respond best to intensive, mood-oriented CBT. 10,33 3-month follow-up 6-month follow-up OR 95% CI p value OR 95% CI p value PHQ9 score (unadjusted) None to minimal depression Mild depression Moderate to severe depression PHQ9 score (adjusted a,b ) None to minimal depression Mild depression Moderate to severe depression Note. CI = confidence interval. a For 3-month follow-up multivariable models adjusted for age, household income, employment status, history of consumption of six drinks or more per drinking occasion, and number of medical comorbidities. b For 6-month follow-up multivariable models adjusted for age, employment status, history of consumption of 6 drinks or more per drinking occasion, and number of medical comorbidities. The overall cessation outcome of 18% in our study is comparable to those reported in the literature for this sub-population of smokers. In a meta-analysis of smoking cessation treatment studies in addictions settings, the pooled cessation rate in the treatment arms was 20.5% at end of treatment. 6 Other studies have reported end of treatment cessation rates of 36.4% 19 and 41.1% among completers of a tailored smoking cessation program in an addictions setting. 34 Limitations These data are from a smoking cessation program implemented in 14 addiction treatment facilities across the province of Ontario, Canada. As such, there is no control or arm and there was no randomization to treatment condition. Instead, this analysis examined predictors of smoking cessation among people seeking treatment in an addictions setting. Cessation outcomes in this study were not biochemically confirmed using expired carbon monoxide or urine or salivary cotinine. As well, there are currently no longitudinal data to examine longerterm abstinence and relapse in this population. However, 12-month follow-up surveys started in July 2014 so these data will be forthcoming in a future publication.

8 944 Nicotine & Tobacco Research, 2015, Vol. 17, No. 8 To conclude, contrary to our hypothesis, depression severity was not a significant predictor of cessation outcome at 3- and 6-month follow-up time points in this population of smokers. A probable explanation may be the individualized treatment and supportive environment of an addictions treatment setting. These data indicate that patients in outpatient addictions treatment can successfully quit smoking regardless of current depressive symptoms. Future studies should randomize smokers to standard treatment versus individualized treatment stratified by baseline depression severity in order to confirm the observational findings of this study. Funding This study was funded by the Health Promotion Division of the Ontario Ministry of Health and Long-term Care(HLTC5047FL). LW has received grant funding from the Canadian Institutes of Health Research (CIHR), the Ontario Lung Association (OLA), Cancer Care Ontario, the Canadian Tobacco Control Research Initiative, Pfizer Canada Inc. and the Global Research Awards for Nicotine Dependence (GRAND), and has received travel funds from Pfizer Canada Inc. PS has received grant funding from Health Canada, CIHR, Pfizer, OLA, ECHO, and NIDA. Declaration of Interests PS has received speaker s bureau/honoraria and consulting fees from Johnson and Johnson Consumer Healthcare Canada, Pfizer Canada Inc., Pfizer Global, and NABI Pharmaceuticals. The remaining authors report nothing to disclose. Acknowledgments The authors would like to acknowledge the 14 Addictions Agencies who participated and all of the staff who helped to implement the smoking cessation program. We would also like to thank the program participants. References 1. Fiore M, Jaen CR, Baker TB, et al. Treating Tobacco Use and Dependence: 2008 Update. Rockville, MD: U.S. Public Health Service; DATIS. Substance Abuse Statistical Tables: Drug and Alcohol Treatment Information System. Centre for Addiction and Mental Health, Toronto, Canada, Hurt RD, Offord KP, Croghan IT, et al. Mortality following inpatient addictions treatment. Role of tobacco use in a community-based cohort. JAMA. 1996;275: Bobo JK, McIlvain HE, Lando HA, Walker RD, Leed-Kelly A. Effect of smoking cessation counseling on recovery from alcoholism: findings from a randomized community intervention trial. Addiction. 1998;93: Frosch DL, Shoptaw S, Nahom D, Jarvik ME. Associations between tobacco smoking and illicit drug use among methadone-maintained opiate-dependent individuals. Exp Clin Psychopharmacol. 2000;8: Prochaska JJ, Delucchi K, Hall SM. A meta-analysis of smoking cessation interventions with individuals in substance abuse treatment or recovery. J Consult Clin Psychol. 2004;72: Joseph AM, Willenbring ML, Nugent SM, Nelson DB. A randomized trial of concurrent versus delayed smoking intervention for patients in alcohol dependence treatment. J Stud Alcohol. 2004;65: Ait-Daoud N, Lynch WJ, Penberthy JK, Breland AB, Marzani-Nissen GR, Johnson BA. Treating smoking dependence in depressed alcoholics. Alcohol Res Health. 2006;29: Martin RA, Rohsenow DJ, MacKinnon SV, Abrams DB, Monti PM. Correlates of motivation to quit smoking among alcohol dependent patients in residential treatment. Drug Alcohol Depend. 2006;83: Haas AL, Munoz RF, Humfleet GL, Reus VI, Hall SM. Influences of mood, depression history, and treatment modality on outcomes in smoking cessation. J Consult Clin Psychol. 2004;72: Kassel JD, Stroud LR, Paronis CA. Smoking, stress, and negative affect: correlation, causation, and context across stages of smoking. Psychol Bull. 2003;129: Lasser K, Boyd JW, Woolhandler S, Himmelstein DU, McCormick D, Bor DH. Smoking and mental illness: a population-based prevalence study. JAMA. 2000;284: Tsuang MT, Francis T, Minor K, Thomas A, Stone WS. Genetics of smoking and depression. Hum Genet. 2012;131: John U, Meyer C, Rumpf HJ, Hapke U. Smoking, nicotine dependence and psychiatric comorbidity a population-based study including smoking cessation after three years. Drug Alcohol Depend. 2004;76: Breslau N, Peterson EL, Schultz LR, Chilcoat HD, Andreski P. Major depression and stages of smoking. A longitudinal investigation. Arch Gen Psychiatry. 1998;55: Shiffman S. Relapse following smoking cessation: a situational analysis. J Consult Clin Psychol. 1982;50: Evins AE, Culhane MA, Alpert JE, et al. A controlled trial of bupropion added to nicotine patch and behavioral therapy for smoking cessation in adults with unipolar depressive disorders. J Clin Psychopharmacol. 2008;28: Munoz RF, Marin BV, Posner SF, Perez-Stable EJ. Mood management mail intervention increases abstinence rates for Spanish-speaking Latino smokers. Am J Community Psychol. 1997;25: Gershon Grand RB, Hwang S, Han J, George T, Brody AL. Short-term naturalistic treatment outcomes in cigarette smokers with substance abuse and/or mental illness. J Clin Psychiatry. 2007;68: ; quiz Cook J, Spring B, McChargue D, Doran N. Effects of anhedonia on days to relapse among smokers with a history of depression: a brief report. Nicotine Tob Res. 2010;12: Leventhal AM, Piper ME, Japuntich SJ, Baker TB, Cook JW. Anhedonia, depressed mood, and smoking cessation outcome. J Consult Clin Psychol. 2014;82: Hitsman B, Papandonatos GD, McChargue DE, et al. Past major depression and smoking cessation outcome: a systematic review and meta-analysis update. Addiction. 2013;108: Fuller BE, Guydish J, Tsoh J, et al. Attitudes toward the integration of smoking cessation treatment into drug abuse clinics. J Subst Abuse Treat. 2007;32: Knudsen HK, Studts JL, Boyd S, Roman PM. Structural and cultural barriers to the adoption of smoking cessation services in addiction treatment organizations. J Addict Dis. 2010;29: Perka EJ Jr. Culture change in addictions treatment: a targeted training and technical assistance initiative affects tobacco-related attitudes and beliefs in addiction treatment settings. Health Promot Pract. 2011;12(suppl 2):159S 165S. 26. Herie M, Connolly H, Voci S, Dragonetti R, Selby P. Changing practitioner behavior and building capacity in tobacco cessation treatment: the TEACH project. Patient Educ Couns. 2012;86: Spitzer RL, Kroenke K, Williams JB. Validation and utility of a self-report version of PRIME-MD: the PHQ primary care study. Primary Care Evaluation of Mental Disorders. Patient Health Questionnaire. JAMA. 1999;282: Spitzer RL, Williams JB, Kroenke K, Hornyak R, McMurray J. Validity and utility of the PRIME-MD patient health questionnaire in assessment of 3000 obstetric-gynecologic patients: the PRIME-MD Patient Health Questionnaire Obstetrics-Gynecology Study. Am J Obstet Gynecol. 2000;183:

9 Nicotine & Tobacco Research, 2015, Vol. 17, No Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16: Hughes JR, Keely JP, Niaura RS, Ossip-Klein DJ, Richmond RL, Swan GE. Measures of abstinence in clinical trials: issues and recommendations. Nicotine Tob Res. 2003;5: Evins AE, Culhane MA, Alpert JE, et al. A controlled trial of bupropion added to nicotine patch and behavioral therapy for smoking cessation in adults with unipolar depressive disorders. J Clin Psychopharmacol. 2008;28: Hall SM, Tsoh JY, Prochaska JJ, et al. Treatment for cigarette smoking among depressed mental health outpatients: a randomized clinical trial. Am J Public Health. 2006;96: Brown RA, Kahler CW, Niaura R, et al. Cognitive-behavioral treatment for depression in smoking cessation. J Consult Clin Psychol. 2001;69: Khara M, Okoli CT. The tobacco-dependence clinic: intensive tobaccodependence treatment in an addiction services outpatient setting. Am J Addict. 2011;20:45 55.