You are here

Article Text

Article menu

Download PDFPDF

Original article
Use of e-cigarettes by individuals with mental health conditions
  1. Sharon E Cummins1,
  2. Shu-Hong Zhu1,
  3. Gary J Tedeschi2,
  4. Anthony C Gamst1,
  5. Mark G Myers3
  1. 1Department of Family and Preventive Medicine, University of California, San Diego, La Jolla, California, USA
  2. 2Cancer Center, University of California, San Diego, La Jolla, California, USA
  3. 3Psychology Service, Veteran Affairs San Diego Healthcare System, Department of Psychiatry, University of California, San Diego, La Jolla, California, USA
  1. Correspondence to Dr Sharon Cummins, Department of Family and Preventive Medicine, University of California, San Diego, 9500 Gilman Drive, MC 0905, La Jolla, CA 92093-0905, USA; scummins{at}ucsd.edu

Abstract

Background Individuals with mental health conditions (MHC) have disproportionately high tobacco-related morbidity and mortality due to high smoking prevalence rates. As high consumers of cigarettes, smokers with MHC may consider using e-cigarettes as an alternative form of nicotine delivery.

Objective Examination of the susceptibility to use e-cigarettes by individuals with MHC.

Methods A US population survey with a national probability sample (n=10 041) was used to assess ever use and current use of regular cigarettes, e-cigarettes, and US Food and Drug Administration-approved pharmacotherapy for smoking cessation. Survey respondents provided information about whether they had been diagnosed with an anxiety disorder, depression, or other MHC.

Results Individuals with MHC were more likely to have tried e-cigarettes (14.8%) and to be current users of e-cigarettes (3.1%) than those without MHC (6.6% and 1.1%, respectively; p<0.01). Ever smokers with MHC were also more likely to have tried approved pharmacotherapy (52.2% vs 31.1%, p<0.01) and to be currently using these products (9.9% vs 3.5%, p<0.01) than those without MHC. Additionally, current smokers with MHC were more susceptible to future use of e-cigarettes than smokers without MHC (60.5% vs 45.3%, respectively, p<0.01).

Conclusions Smokers with MHC are differentially affected by the rise in popularity of e-cigarettes. Clinical interventions and policies for tobacco control on e-cigarettes should take into account the possible outcomes and their implications for this priority population.

  • Electronic nicotine delivery devices
  • Harm Reduction
  • Disparities
  • Priority/special populations
  • Nicotine

This is an Open Access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 3.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/3.0/

https://doi.org/10.1136/tobaccocontrol-2013-051511

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    The use of electronic cigarettes (commonly known as e-cigarettes) has increased dramatically in the last few years. As of 2012, an estimated 8.1% of adults in the USA had used e-cigarettes.1 This estimate is up considerably from the 3.3% and 6.2% found in 2010 and 2011, respectively.2 Likewise, use of e-cigarettes among high school and middle school students doubled from 2011 to 2012.3 There has been considerable speculation about what impact these products will have on tobacco control efforts.1 ,4 ,5 Research on e-cigarettes is in the early stages and, as with most tobacco research, has been focused on the general population.1 ,2 ,6 The current study compares individuals with mental health conditions (MHC) to those without, on their use and perceptions of e-cigarettes. We believe this study is important because mental health status is rarely assessed in tobacco research, yet those with MHC are arguably the group most burdened by the health consequences of tobacco use.7

    Smoking prevalence for individuals with MHC is estimated to be about 70% higher than for those without (36.1% vs 21.4%).8 ,9 Rates of smoking differ by diagnosis, ranging from approximately 38% for those with anxiety disorder, 45% for those with affective disorders, and 64% for those with substance use disorders.10 Smokers with mental health disorders typically smoke more heavily than smokers without mental illness, and have poorer cessation outcomes.8 ,9 Over time, that difficulty with quitting has resulted in higher rates of MHC among the remaining smokers. In fact, estimates suggest that individuals with psychiatric disorders consume between a third to a half of all cigarettes sold in the USA.8 ,9 ,11 Consequently, tobacco-related morbidity and mortality are disproportionately high among individuals with MHC.12 ,13

    Historically, faulty assumptions about smokers with MHC have kept researchers from including them in cessation studies, and clinicians from addressing cessation in their treatment plans. In other words, there has been a persistent belief that smokers with MHC are not interested in quitting, not capable of quitting, and that cessation may have a negative impact on their mental illness.14–16 However, attitudes have been changing. Recent publications in the research and clinical literature have highlighted the need to address tobacco use among individuals with MHC.14–20 There is greater recognition that such smokers worry about the health effects of their smoking and want to quit.14 Mental health communities have called for healthcare professionals to address the issue of smoking with their patients.21 ,22 Also, many mental health treatment facilities have instituted smoke-free campuses and have begun to offer cessation treatment concurrent with treatment for mental health or substance disorders.22 It is now recognised that smokers with MHC may require a different treatment approach to quitting than other smokers, one characterised by an extended number of counselling sessions, extended use of pharmacotherapy, or both.16 ,23 ,24 Williams et al7 make a strong case that smokers with mental health and substance abuse conditions should be designated as a tobacco-disparity group, to focus national attention and resources on this priority population.

    We believe that in the midst of the lively debate about e-cigarettes it is important to consider whether e-cigarettes have a differential impact on individuals with MHC. E-cigarettes are particularly relevant to smokers with MHC because they have provoked debate about their potential to reduce the harm associated with smoking regular cigarettes and their usefulness as a cessation aid.25–27 Despite a shortage of evidence to support these claims, e-cigarettes are commonly promoted for smoking cessation28 and often are used for this purpose.1 ,29 Of particular concern is that some e-cigarette manufacturers are marketing their products as a means for managing psychiatric symptoms.30 ,31 These companies are directly targeting individuals with MHC, whether they are current smokers or not.

    This study examines the use of e-cigarettes, regular cigarettes, and approved quitting aids, comparing individuals with MHC to those without. If those with MHC are a priority for tobacco control, then it is necessary to examine whether they are being differentially affected by the rise in popularity of e-cigarettes. Specifically, this study examines consumption rates, and whether the reasons for using e-cigarettes differ by mental health status.

    Methods

    Data source

    A survey was commissioned by the University of California, San Diego and administered in February 2012 by GfK, a market research institute. GfK recruits a probability sample representative of the US population (KnowledgePanel) using an address-based sampling methodology.32 ,33 Surveys are performed online, and panel participants are provided with a netbook computer and network access, as needed. The KnowledgePanel methodology results in a probability sample of the US population with representativeness equivalent to most other well-known population surveys.34–36

    The present study came out of a cross-sectional survey that was designed to gather information on smoking history and cigarette use, perceptions about different tobacco products (including e-cigarettes) and quitting aids, attitudes toward tobacco control efforts, and beliefs about the process of quitting smoking.1 The response rate for the survey was 66.5% (n=10 041) of which 3111 were current smokers, 3676 were former smokers and 3254 were never smokers. The survey was conducted between 24 February and 8 March 2012.

    Measurement

    Upon enrolment, each KnowledgePanel participant is asked to complete a self-reported personal health profile that includes smoking status and a list of ailments and diagnosed medical conditions such as attention deficit disorder, diabetes, heart disease, high blood pressure and cancer. Three MHC are listed, including anxiety disorder, depression, and other ‘mental health condition.’ For the purposes of this study, anyone who reported any of the three mental health items was coded as having a MHC.

    Cigarette smoking was assessed using multiple questions with smokers defined as those who had smoked at least 100 cigarettes in their lifetime. Smokers were classified as: current, if they said they smoked ‘every day’ or ‘some days’ (when asked ‘Do you currently smoke cigarettes every day, some days, or not at all?’), recent former if they smoked within the past year, and long-term former if they last smoked over 1 year ago. Never smokers were defined as those who had not smoked 100 cigarettes in their lifetime.

    Likewise, the use of e-cigarettes was assessed using multiple questions. Since the product was relatively new, the survey described e-cigarettes as ‘electronic devices that deliver nicotine in a vapour and look like cigarettes, but contain no tobacco.’ Respondents who had ever used e-cigarettes were coded as current users if they had used them in the previous 30 days. Respondents were also asked about their ever use and current use (within the past 30 days) of the following approved quitting aids: nicotine patches, nicotine gum, nicotine lozenges, nicotine spray/inhaler, Zyban (Wellbutrin, bupropion), and Chantix (varenicline).

    Ever users of e-cigarettes were asked why they used e-cigarettes, and indicated yes or no for each of the following options: ‘safer than cigarettes’, ‘cheaper than cigarettes,’ ‘easy to use when I can't smoke,’ ‘to try to quit smoking cigarettes,’ or ‘just because.’ These options were presented in randomised order for each respondent to minimise the order effect in response. To assess susceptibility to using e-cigarettes in the future, respondents who had never used e-cigarettes were asked ‘How likely are you to try e-cigarettes in the future?’ Those who responded that they were very likely or somewhat likely to use e-cigarettes were considered to be susceptible, as were those who had tried e-cigarettes but were not currently using them.1 ,37

    Analysis

    The 2010 US Current Population Survey was used to calculate weighted percentages.36 A poststratification adjustment was used to account for non-response, for non-coverage, and undersampling and oversampling resulting from the KnowledgePanel survey-specific sampling design. An adjustment for the oversampling of smokers produced an overall smoking prevalence for the USA of 19.1%, equivalent to a national estimate of 19.3% based on the 2010 National Health Interview Survey.38 Results were analysed by demographic categories (gender, age, educational level, ethnic background) and by smoking status. Demographic characteristics were examined as potential confounders of mental health status and e-cigarette outcomes. Gender emerged as the only possible confounder, and only for the outcome of ever use of e-cigarettes. A logistic regression with gender, mental health status, and the interaction term confirmed that gender and the interaction were not significantly related to the outcome with mental health status in the model, and that the findings for mental health status were robust. SEs were calculated and 95% CIs were computed based on the sampling distribution of the corresponding summary statistic. CIs for binomial proportions were computed using the method of Agresti and Coull.39 ,40 P values were determined using logistic regression, with a Holm's adjustment to protect against familywise error from the large number of comparisons.41 All calculations were done using R 2.12.1.42

    Results

    Table 1 shows the demographic breakdown for those who endorsed at least one of the MHC: depression, anxiety disorder, or ‘other mental health condition.’ Those with MHC were more likely to be female (p<0.01). Other differences in ethnicity and age were not significant using a Holm-adjusted p value.

    View this table:
    Table 1

    Demographics by mental health condition

    Table 2 provides the breakdown of mental health condition by smoking status. Overall, the adjusted prevalence of MHC was 17.2%; 13.7% reported being diagnosed with depression, 8.6% with anxiety disorder, and 3.6% with some other mental health condition. Among those with any mental health condition, depression was the most commonly reported (79.4%), followed by anxiety disorder (49.7%); the least common was some other MHC which was reported by 21.0% of those with any MHC. Seventy percent of respondents who reported any MHC reported having a single condition, 30% reported having two, and 10% reported having all three (data not shown). Current smokers and recent former smokers had significantly higher rates of self-reported MHC (27.8% and 26.0%, respectively) than did non-smokers (13.4%) or long-term former smokers (16.4%). Never were marginally different from long-term former smokers (p=0.01, unadjusted value, p=0.025 with the Holm adjustment).

    View this table:
    Table 2

    Mental health condition by smoking status

    Table 3 presents data on the ever use and current use of a range of US Food and Drug Administration-approved quitting aids for those who had ever smoked cigarettes comparing across mental health status. Ever smokers with MHC were significantly more likely to have ever used a quitting aid (52.2%) than those without (31.1%, p<0.01), as well as to currently be using a quitting aid (9.9% and 3.5%, respectively, p<0.01). Each product was significantly more likely to be used by individuals with MHC than by those without, except for the nicotine inhaler/spray where the rates were comparable across mental health status. Use of bupropion was three times as high (for ever use) and six times as high (for current use) among those with MHC compared to those without. Bupropion is an antidepressant (under the name Wellbutrin), but is also prescribed as a cessation aid under the name Zyban. In this study, participants were not asked whether their use of bupropion was for depression, cessation, or both. However, removing bupropion from the analysis of ever use and current use of a quitting aid did not change the conclusions that those with MHC were more likely to have ever used approved cessation aids compared to those without MHC (28.6%, 95% CI 25.1 to 32.1 vs 23.2%, 95% CI 21.6 to 24.8, respectively, p<0.01) although the current rates of use by ever smokers were not statistically different by mental health status (5.5%, 95% CI 3.7 to 7.3 vs 3.1%, 95% CI 2.5 to 3.7 for MHC and no MHC, respectively, p>0.05). Additionally, smokers with MHC who had used any quitting aid were more likely to have tried multiple quitting aids. Removing bupropion from the analysis (since it is an antidepressant that is also used as a cessation aid), those with MHC had used an average of 1.90 different quitting aids compared to an average of 1.68 for those without MHC (p<0.001).

    View this table:
    Table 3

    Use of quitting aids by mental health condition among ever smokers

    Across smoking status, the rate of ever using e-cigarettes was 8.1% (table 4). Rates were significantly higher for current smokers and recent former smokers (32.2% and 26.8%, respectively) than for long-term former smokers or for non-smokers of conventional cigarettes (2.4% and 1.0%, respectively). Overall, participants with MHC were twice as likely to have tried e-cigarettes (14.8%) as those without (6.6%, p<0.01). Over 40% of current smokers with MHC had tried e-cigarettes compared to 28.7% of current smokers without such conditions, a statistically significant difference (p<0.01).

    View this table:
    Table 4

    Ever and current use of e-cigarettes by mental health condition and smoking status

    Current use of e-cigarettes was much lower than ever use (1.4% vs 8.1%). However, the pattern for current use was similar to the pattern for ever having used e-cigarettes (second row of table 4). Rates were significantly higher for current and recent former smokers (6.3% and 6.1%) than for long-term former smokers or non-smokers (0.2% and 0.04%). Although there was a trend toward more current smokers with MHC being dual users (ie, use conventional cigarettes and e-cigarettes) than those without MHC, the difference was not significant (8.6% vs 5.4%, p=0.056).

    The most common reason survey participants gave for trying e-cigarettes was ‘just because’ (68.9%), followed by ‘to try and quit smoking cigarettes’ (55.2%), ‘safer than cigarettes’ (51.2%), and ‘easy to use when I can't smoke’ (46.7%). The least common reason was that they were ‘cheaper than cigarettes’ (31.3%). Although participants with MHC were more likely to endorse all the reasons than those without, none of the differences reached significance (table 5).

    View this table:
    Table 5

    Reasons for use of e-cigarettes by mental health status*

    Table 6 shows that the susceptibility to using e-cigarettes in the future was significantly higher among current smokers than among recent former smokers (49.5% vs 24.9%). Recent former smokers were, in turn, more susceptible than long-term former smokers (3.3%) or never smokers (2.6%). Long-term former smokers were not significantly more susceptible than those who have never smoked. The table also indicates that only among current smokers does the difference in susceptibility between those with MHC and those without reach significance. Over 60% of current smokers with MHC are susceptible to using e-cigarettes compared to 45% of smokers without MHC (p<0.01).

    View this table:
    Table 6

    Susceptibility to using e-cigarettes in the future*

    Discussion

    This is the first study to demonstrate that individuals with MHC are more likely to try e-cigarettes, to be current users of e-cigarettes, and to be susceptible to future use of e-cigarettes. This study also showed that individuals with MHC were more likely to smoke cigarettes than those without MHC, which is consistent with earlier research in the field.8 ,9 ,11 Smokers with MHC meet all the criteria for classification as a tobacco disparity group, including higher tobacco prevalence, higher levels of nicotine dependence, lower cessation rates, disproportionate health burden from tobacco use, and so on.7 The finding that smokers with MHC are differentially susceptible to e-cigarette use is an important step in the surveillance and treatment of this priority population.

    Although smokers with MHC are more susceptible to e-cigarette use than those without MHC, it should be noted that they appear to use e-cigarettes for the same reasons. Some are experimenting with the product, but many indicate they are using e-cigarettes to quit smoking regular cigarettes or to reduce the harm associated with regular cigarettes. We found that e-cigarettes appeal most to smokers and, to a lesser degree, recent quitters, and that this is true among individuals with MHC and among those without. Never smokers and former smokers who quit smoking over a year appear to have very low susceptibility to using e-cigarettes, regardless of MHC. At present, e-cigarettes do not appear to be attractive to those who do not smoke, even though e-cigarettes have been marketed as a way to cope with depression and anxiety.

    What explains the greater use and susceptibility to using e-cigarettes seen among individuals with MHC? One possible explanation is that e-cigarette use is an extension of the tendency to use approved quitting aids. Although e-cigarettes are not approved as cessation aids (in fact, it is illegal to make the claim that they are effective aids to quitting), many smokers view them as such, regardless of mental health status. In this study, smokers with MHC were more likely to have used approved cessation aids than their non-MHC counterparts; they seem to find the use of cessation aids, such as nicotine replacement therapy and other pharmacotherapies, more acceptable. By extension, they may find e-cigarettes more acceptable as well.

    The greater use of approved cessation aids may also reflect an acceptance by individuals with MHC that quitting may be particularly difficult for them. Although research has shown that smokers with mental health disorders benefit from cessation treatment, they typically have poorer cessation outcomes.9 ,14 ,43 The current study found that smokers with MHC were more likely than those without MHC to have used approved aids, and were more likely to have tried two, three, or four of them. Even when bupropion, which is an antidepressant also used as a cessation aid, was removed from the analysis, those with MHC were more likely to have tried two or three of the approved aids. Perhaps, having failed to quit using other strategies, they are looking to try something new and may be open to using e-cigarettes or one of the newer nicotine delivery devices.44

    This study has some limitations, mostly related to the use of self-report for MHC, e-cigarette use, and reasons for using e-cigarettes. At the time of enrolment into the KnowledgePanel survey, participants were asked about three MHC as part of their health profile: depression, anxiety disorder, or ‘mental health condition.’ No guidance was given as to the definition of each except to indicate whether they had ever been diagnosed by a healthcare provider with any of the listed conditions. As a result, there could have been some overendorsement, particularly of depression, since that term is broadly used. The 13.7% rate of depression reported here is slightly lower than the 16% lifetime prevalence of depression reported in the National Comorbidity Survey Replication (NCS-R) study.45 Overendorsement is unlikely to have affected our findings. The use of e-cigarettes is also self-report, although there is little reason to think that survey participants would not remember having used the products or that they would misrepresent their use of them. Respondents selected their reasons for using e-cigarettes from options provided, which did not include other possibilities, such as a way to self-medicate or cope with psychological issues.

    Conclusions

    Limitations notwithstanding, data from this study clearly indicate that smokers with MHC are more susceptible to the use of e-cigarettes than are smokers without these conditions. They use e-cigarettes for the same reasons as other smokers—as quitting aids, and because they perceive them to be less harmful than smoking. Equally clear is the fact that e-cigarettes appeal to smokers, but not to those who have never smoked or who quit smoking for more than a year. In the near future, regulations, social norms and guidelines on the use of e-cigarettes in treatment settings are likely to change rapidly. We urge ongoing assessment of mental health status in order to be able to determine if new regulations on pricing, access to e-cigarettes, bans related to e-cigarettes, and the like affect individuals with MHC differentially, and whether differences in use and susceptibility serve to increase or decrease health disparities of this priority population.

    What this paper adds

    • This is the first study to demonstrate that individuals with mental health conditions are more likely to try e-cigarettes, to be current users of e-cigarettes, and to be susceptible to future use of e-cigarettes.

    • Individuals with mental health conditions use e-cigarettes for the same reasons that other smokers do—primarily to quit smoking regular cigarettes or to reduce the harm associated with regular cigarettes.

    • Use of e-cigarettes by smokers with mental health conditions may be an extension of their tendency to use approved quitting aids.

    Acknowledgments

    The authors would like to thank Lu Yin for her assistance in analysing the data, and Jessica Sun for her help with preparing the manuscript for submission.

    References

    1. ↵
      1. Zhu S,
      2. Gamst A,
      3. Lee M,
      4. et al
      . The use and perception of electronic cigarettes and snus among the U.S. population. PLoS ONE 2013;8:e79332.
      OpenUrlCrossRefPubMed
    2. ↵
      1. King BA,
      2. Alam S,
      3. Promoff G,
      4. et al
      . Awareness and ever-use of electronic cigarettes among U.S. adults, 2010–2011. Nicotine Tob Res 2013;15:1623–7.
      OpenUrlAbstract/FREE Full Text
    3. ↵
      Centers for Disease Control and Prevention (CDC). Tobacco product use among middle and high school students—United States, 2011 and 2012. MMWR 2013;62:893–7.
      OpenUrlPubMed
    4. ↵
      1. Benowitz NL,
      2. Goniewicz ML
      . The regulatory challenge of electronic cigarettes. JAMA 2013;310:685–6.
      OpenUrlCrossRefPubMedWeb of Science
    5. ↵
      1. Van Zyl-Smit RN
      . Electronic cigarettes: the potential risks outweigh the benefits. S Afr Med J 2013;103:7435. http://www.samj.org.za/index.php/samj/article/view/7435
      OpenUrl
    6. ↵
      1. Etter JF,
      2. Bullen C,
      3. Flouris AD,
      4. et al
      . Electronic nicotine delivery systems: a research agenda. Tob Control 2011;20:243–8.
      OpenUrlAbstract/FREE Full Text
    7. ↵
      1. Williams JM,
      2. Steinberg ML,
      3. Griffiths KG,
      4. et al
      . Smokers with behavioral health comorbidity should be designated a tobacco use disparity group. Am J Public Health 2013;103:1549–55.
      OpenUrlCrossRefPubMedWeb of Science
    8. ↵
      1. Lasser K,
      2. Boyd JW,
      3. Woolhandler S,
      4. et al
      . Smoking and mental illness: a population-based prevalence study. JAMA 2000;284:2606–10.
      OpenUrlCrossRefPubMedWeb of Science
    9. ↵
      Centers for Disease Control and Prevention (CDC). Vital signs: Current cigarette smoking among adults aged >/=18 years with mental illness—United States, 2009–2011. MMWR 2013;62:81–7.
      OpenUrlPubMed
    10. ↵
      1. Lawrence D,
      2. Mitrou F,
      3. Zubrick SR
      . Smoking and mental illness: results from population surveys in Australia and the United States. BMC Public Health 2009;9:285.
      OpenUrlCrossRefPubMed
    11. ↵
      1. Grant BF,
      2. Hasin DS,
      3. Chou SP,
      4. et al
      . Nicotine dependence and psychiatric disorders in the United States: results from the national epidemiologic survey on alcohol and related conditions. Arch Gen Psychiatry 2004;61:1107–15.
      OpenUrlCrossRefPubMedWeb of Science
    12. ↵
      1. Colton CW,
      2. Manderscheid RW
      . Congruencies in increased mortality rates, years of potential life lost, and causes of death among public mental health clients in eight states. Prev Chronic Dis 2006;3:A42.
      OpenUrlPubMed
    13. ↵
      1. Parks J,
      2. Svendsen D,
      3. Singer P,
      4. et al
      . Morbidity and mortality in people with serious mental illness. National Association of State Mental Health Program Directors (NASMHPD) Medical Directors Council. 2006;vol 13.
    14. ↵
      1. Prochaska JJ
      . Smoking and mental illness—breaking the link. N Engl J Med 2011;365:196–8.
      OpenUrlCrossRefPubMedWeb of Science
    15. ↵
      1. Morris CD,
      2. Waxmonsky JA,
      3. May MG,
      4. et al
      . What do persons with mental illnesses need to quit smoking? Mental health consumer and provider perspectives. Psychiatr Rehabil J 2009;32:276–84.
      OpenUrlCrossRefPubMedWeb of Science
    16. ↵
      1. Schroeder SA,
      2. Morris CD
      . Confronting a neglected epidemic: tobacco cessation for persons with mental illnesses and substance abuse problems. Annu Rev Public Health 2010;31:297–314.
      OpenUrlCrossRefPubMedWeb of Science
    17. ↵
      1. Ziedonis DM,
      2. Kolodziej ME
      . Innovative interventions for treating tobacco addiction among persons with co-occurring mental illness and addiction: new approaches to improve outcomes. J Dual Diagn 2012;8:85–8.
      OpenUrlCrossRefWeb of Science
    18. ↵
      1. Baker A,
      2. Ivers RG,
      3. Bowman J,
      4. et al
      . Where there's smoke, there's fire: high prevalence of smoking among some sub-populations and recommendations for intervention. Drug Alcohol Rev 2006;25:85–96.
      OpenUrlCrossRefPubMedWeb of Science
    19. ↵
      1. Prochaska JJ
      . Failure to treat tobacco use in mental health and addiction treatment settings: a form of harm reduction? Drug Alcohol Depend 2010;110:177–82.
      OpenUrlCrossRefPubMedWeb of Science
    20. ↵
      1. Williams JM,
      2. Ziedonis D
      . Addressing tobacco among individuals with a mental illness or an addiction. Addict Behav 2004;29:1067–83.
      OpenUrlCrossRefPubMedWeb of Science
    21. ↵
      1. Williams JM
      . Eliminating tobacco use in mental health facilities: patients’ rights, public health, and policy issues. JAMA 2008;299:571–3.
      OpenUrlCrossRefPubMedWeb of Science
    22. ↵
      National Association of State Mental Health Program Directors. Tobacco-free living in psychiatric settings: a best-practices toolkit promoting wellness and recovery. National Association of State Mental Health Program Directors (NASMHPD) Medical Directors Council. 2010.
    23. ↵
      1. Hitsman B,
      2. Moss TG,
      3. Montoya ID,
      4. et al
      . Treatment of tobacco dependence in mental health and addictive disorders. Can J Psychiatry 2009;54:368–78.
      OpenUrlPubMed
    24. ↵
      1. Fiore M,
      2. Jaén C,
      3. Baker T,
      4. et al
      . Treating tobacco use and dependence: 2008 update. U.S. Public Health Service Clinical Practice Guideline. 2008:1–257.
    25. ↵
      1. Cahn Z,
      2. Siegel M
      . Electronic cigarettes as a harm reduction strategy for tobacco control: a step forward or a repeat of past mistakes? J Public Health Policy 2011;32:16–31.
      OpenUrlCrossRefPubMedWeb of Science
    26. ↵
      1. Wagener TL,
      2. Siegel M,
      3. Borrelli B
      . Electronic cigarettes: achieving a balanced perspective. Addiction 2012;107:1545–8.
      OpenUrlCrossRefPubMedWeb of Science
    27. ↵
      1. Bullen C,
      2. Howe C,
      3. Laugesen M,
      4. et al
      . Electronic cigarettes for smoking cessation: a randomised controlled trial. Lancet 2013;382:1629–37.
      OpenUrlCrossRefPubMedWeb of Science
    28. ↵
      1. Ayers JW,
      2. Ribisl KM,
      3. Brownstein JS
      . Tracking the rise in popularity of electronic nicotine delivery systems (electronic cigarettes) using search query surveillance. Am J Prev Med 2011;40:448–53.
      OpenUrlCrossRefPubMedWeb of Science
    29. ↵
      1. Etter JF,
      2. Bullen C
      . Electronic cigarette: users profile, utilization, satisfaction and perceived efficacy. Addiction 2011;106:2017–28.
      OpenUrlCrossRefPubMedWeb of Science
    30. ↵
      Quit In Time—Smoking Cessation. Quit smoking: impact on anxiety disorders. http://www.quitintime.com/quit-smoking-anxiety-disorders/. Updated 2011. (accessed 11 Mar 2014)
    31. ↵
      Smokeless Delite. Electronic cigarette may reduce depression. http://www.smokelessdelite.com/electronic-cigarettes-reduce-depression/. Updated 2012. (accessed 11 Mar 2014).
    32. ↵
      1. DiSogra C
      . Update: address-based sampling nets success for KnowledgePanel® recruitment and sample representation. http://www.knowledgenetworks.com/accuracy/spring2010/pdf/disogra-spring10.pdf. Updated 2010. (accessed 11 Mar 2014).
    33. ↵
      Knowledge Networks. KnowledgePanel® design summary. http://www.knowledgenetworks.com/knpanel/docs/knowledgePanel(R)-design-summary-description.pdf. Updated 2013. (accessed 11 Mar 2014).
    34. ↵
      1. Yeager DS,
      2. Krosnick JA,
      3. Chang L,
      4. et al
      . Comparing the accuracy of RDD telephone surveys and internet surveys conducted with probability and non-probability samples. Public Opinion Q 2011;75:709–47.
      OpenUrlAbstract/FREE Full Text
    35. ↵
      1. Chang L,
      2. Krosnick JA
      . National surveys via RDD telephone interviewing versus the internet: comparing sample representativeness and response quality. Public Opinion Q 2009;73:641–78.
      OpenUrlAbstract/FREE Full Text
    36. ↵
      Knowledge Networks. KnowledgePanel demographic profile February 2012. http://www.knowledgenetworks.com/knpanel/docs/GfK-KnowledgePanel(R)-Demographic-Profile.pdf. Updated 2012. (accessed 11 Mar 2014).
    37. ↵
      1. Pierce JP,
      2. Choi WS,
      3. Gilpin EA,
      4. et al
      . Validation of susceptibility as a predictor of which adolescents take up smoking in the United States. Health Psychol 1996;15:355–61.
      OpenUrlCrossRefPubMedWeb of Science
    38. ↵
      Centers for Disease Control and Prevention. Vital signs: current cigarette smoking among adults aged ≥18 years—United States, 2005–2010. MMWR 2011;60:1207–12.
      OpenUrlPubMed
    39. ↵
      1. Brown LD,
      2. Cai TT,
      3. DasGupta A
      . Interval estimation for a binomial proportion. Stat Sci 2001;16:101–17.
      OpenUrlCrossRefWeb of Science
    40. ↵
      1. Agresti A,
      2. Coull BA
      . Approximate is better than “exact” for interval estimation of binomial proportions.  Am Statistician 1998;52:119–26.
      OpenUrlCrossRefWeb of Science
    41. ↵
      1. Holm S
      . A simple sequentially rejective multiple test procedure. Scand J Stat 1979:6:65–70.
      OpenUrlWeb of Science
    42. ↵
      The R Project for Statistical Computing. An introduction to R. http://cran.r-project.org/doc/manuals/R-intro.html. Updated 2014. (accessed 11 Mar 2014).
    43. ↵
      1. Hebert KK,
      2. Cummins SE,
      3. Hernandez S,
      4. et al
      . Current major depression among smokers using a state quitline. Am J Prev Med 2011;40:47–53.
      OpenUrlCrossRefPubMedWeb of Science
    44. ↵
      1. Shahab L,
      2. Brose LS,
      3. West R
      . Novel delivery systems for nicotine replacement therapy as an aid to smoking cessation and for harm reduction: rationale, and evidence for advantages over existing systems. CNS Drugs 2013;27:1007–19.
      OpenUrlCrossRefPubMedWeb of Science
    45. ↵
      1. Kessler RC,
      2. Berglund P,
      3. Demler O,
      4. et al
      . The epidemiology of major depressive disorder: results from the national comorbidity survey replication (NCS-R). JAMA 2003;289:3095–105.
      OpenUrlCrossRefPubMedWeb of Science

    Footnotes

    • Contributors SEC, S-HZ, ACG and GJT participated in conceptualising and designing the study and in collecting and analysing the data. SEC wrote the first draft, and all authors contributed to the writing and review of the paper.

    • Funding This study was supported by a grant from the National Cancer Institute, U01 CA154280.

    • Competing interests None.

    • Ethics approval This research was performed in accordance with a human subject protocol approved by the University of California, San Diego's Institutional Review Board (IRB# 111664).

    • Provenance and peer review Not commissioned; externally peer reviewed.