You are here

Article Text

Article menu

Download PDFPDF

XML
Addiction
Protocol
Real-time patterns of smoking and alcohol use: an observational study protocol of risky-drinking smokers
  1. Amy Cohn1,
  2. Thomas Brandon2,
  3. Stephen Armeli3,
  4. Sarah Ehlke4,
  5. Molly Bowers4
  1. 1Schroeder Institute, Washington DC, USA
  2. 2Moffitt Cancer Center, University of South Florida, Tampa, Florida, USA
  3. 3Department of Psychology, Fairleigh Dickinson University, Teaneck, New Jersey, USA
  4. 4Schroeder Institute for Tobacco Research and Policy Studies, Legacy, Washington DC, USA
  1. Correspondence to Dr Amy Cohn; Acohn{at}legacyforhealth.org

Abstract

Introduction Despite the strong relationship between smoking and health-related consequences, very few smokers quit. Heavy drinking is a significant risk factor for health consequences, and is implicated in persistent smoking and less success at quitting smoking. Self-efficacy (SE) to abstain from smoking is an important determinant of smoking outcomes and may link alcohol use to poor quit rates. Even though research has demonstrated a strong association between drinking and smoking, and the multiplicative effect of these substances on cancer-related, heavy-drinking smokers has been largely ignored in the literature. Further, research has not taken advantage of innovative methods, such as ecological momentary assessment, to capture the impact of daily factors on smoking cessation outcomes in this particular group. The proposed study identifies daily changing factors that impede or promote SE and future smoking cessation efforts in risky-drinking smokers.

Methods and analysis This is an observational study of 84 regular smokers (≥10 cigarettes per day) who drink at risky levels, report a desire to quit in the next 6 months, and show no evidence of psychiatric disturbance, severe history of alcohol withdrawal or drug dependence (excluding nicotine and caffeine). Participants report on their smoking, alcohol consumption and SE related to smoking twice a day for 28 days using interactive voice response (IVR) surveys. Multilevel regression and path models will examine within-person daily associations among drinking, smoking and SE, and how these variables predict the likelihood of future smoking behaviour at 1 and 6 months follow-up.

Ethics and dissemination This protocol was approved by an accredited Institutional Review Board. The findings will help us understand the factors that promote or impede smoking cessation among a high-risk group of smokers (heavy-drinking smokers) and will be disseminated through peer-reviewed journal articles and presentations at national conferences.

  • PUBLIC HEALTH
  • STATISTICS & RESEARCH METHODS

This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: http://creativecommons.org/licenses/by/4.0/

https://doi.org/10.1136/bmjopen-2014-007046

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

    Cigarette smoking is one of the leading causes of cancer-related disability in the USA.1–3 Problematic drinking is frequently associated with persistent smoking, less success at quitting smoking and smoking relapse,4–7 while the concurrence of smoking and drinking represents a significant risk factor for cancer-related illness and mortality.8–14

    Self-efficacy (SE) to quit smoking, defined as confidence in one's ability to abstain from smoking in ‘high risk’ situations, is an important determinant of smoking cessation.15 ,16 According to Bandura's influential model of behaviour change17 (figure 1), smoking and behaviour change from smoking are influenced by SE. However, Bandura's model does not address individual differences in the co-occurrence of daily drinking and smoking, and their impact on subsequent levels of SE, even though (1) alcohol and cigarette smoking are strongly associated with one another and (2) alcohol frequently serves as a contextual trigger for smoking.18–20 This study expands Bandura's model by focusing explicitly on constructs at the momentary or daily level of analysis. It is hypothesised that, over and above average levels of alcohol use and smoking, individuals with stronger co-occurrence of drinking and smoking on a given day will have weaker SE at subsequent time points (ie, later that day or on the next day), and individuals with weaker SE would have a reduced likelihood of smoking behaviour change (reduced cigarettes per day, cessation attempts or successful abstinence) at follow-up.

    Figure 1
    Figure 1

    The impact of daily factors (drinking, smoking, and self-efficacy to abstain from smoking) on smoking cessation outcomes. Adapted from Bandura's17 model of behavior change.

    The measurement of SE and the contextual mechanisms linking SE to smoking remain largely unspecified.16 ,21 One reason is because most research has relied on retrospective measures obtained at weekly, monthly or longer intervals,16 which do not capture the naturally occurring events that impact SE or SE's variations across situations, contexts and days.17 Second, most of the previous research had not included sufficient within-person details about acute momentary changes in SE, smoking or drinking that, as purported in the literature, occur on a day-to-day basis.22 ,23

    One strategy for precisely measuring SE in response to everyday events is through the use of electronic diary methods, which allow for the collection of behavioural phenomena in natural settings and in, or near, real-time. These methods can establish temporal sequences among key constructs;23–26 they are more ecologically valid than traditional paper-and-pencil formats; and they allow for assessment of within-person variability.27–29 In this project, we use interactive voice response (IVR) technology, in which participants answer a set of prerecorded audio-voice questions by pushing buttons on the keypad of their telephone.

    Project aims

    This study follows the daily smoking, drinking and SE patterns of a sample of 84 risky-drinking smokers. Aim 1 is to test the relationship of drinking and smoking (at the daily level) with smoking cessation outcomes at 1 and 6 months follow-up. We propose that more smoking will occur on days characterised by greater alcohol consumption (referred to as daily alcohol-smoking concordance) and that individuals with greater alcohol-smoking concordance will report worse smoking cessation outcomes.

    Aim 2 is to test the mediating effect of daily SE (within days and day-to-day) on the relationship of daily alcohol-smoking concordance with short-term and long-term smoking cessation outcomes.

    Methods and analysis

    Project design

    The current study consists of three phases. Phase 1 is the baseline (BL) assessment, during which participants sign informed consent and complete self-report questionnaires, and gave semistructured interviews about patterns and history of cigarette and alcohol use, and related behaviours. In phase 2, participants complete twice daily IVR surveys (morning and night) for 28 days. Finally, in phase 3, participants complete a follow-up at 1 and 6 months post-BL to assess smoking and drinking since the previous assessment, motivation and desire to change (smoking and drinking), SE to quit smoking, smoking cessation attempts and the barriers to quitting (where applicable). Figure 2 displays the study design.

    Figure 2
    Figure 2

    Study design and key areas of measurement at each phase of the study. IVR, interactive voice response.

    Inclusion and exclusion criteria

    Participants are 84 regular smokers (≥10 cigarettes per day) aged 18–65 who drink at risky levels. Risky drinking is defined according to guidelines set forth by the National Institute on Alcohol Abuse and Alcoholism: consuming >2 drinks per day for men or >1 drink per day for women; and >14 drinks per week for men or >7 drinks per week for women.30 The present study excludes those who report suicidal or homicidal ideation, intent or plan, or who report severe psychiatric disturbance, substance dependence with physiological dependence (excluding nicotine and caffeine), the current use of psychotropic medication for mental health, the potential for severe alcohol withdrawal (history of seizures, shakes or severe physical discomfort when absent from alcohol for at least 1 day), or who are pregnant or planning to become pregnant in the next 6 months.i

    Recruitment

    Individuals are recruited via online advertisements, flyers, print advertisements in local newspapers and respondent driven sampling. All recruitment materials direct participants to complete an online screening or to call the study phone number to determine eligibility. Advertisements ask for research participants who smoke and drink regularly.

    Procedures

    Data for this project are obtained through two sources: (1) self-report surveys and clinical interviews completed in person at BL and at 1 and 6 months post-BL, and (2) automated telephone surveys using IVR.

    After an initial screening, eligible individuals are invited for a BL assessment at the study site (lasting approximately 1.5–2 h) to confirm eligibility. Before attending the BL, participants are instructed not to consume alcohol that day. At the BL, participants provide an assessment of blood alcohol concentration by expiring breath into a breathalzyer to ensure they are not intoxicated at the time of the assessment (BAC of 0.00). The informed consent document is reviewed orally at the beginning of each BL, participants are provided with a hard copy to review and they are given time to ask any questions regarding participation. A Certificate of Confidentiality (CoC) has been obtained to protect against further disclosure of the use of illegal substances or underage consumption of alcohol. A copy of the COC is available to participants on request.

    Participants then complete self-report questionnaires and several interviewer-administered questionnaires about alcohol, substance use disorder and nicotine dependence symptoms as well as current alcohol, cigarette, tobacco and other drug use.

    At the conclusion of the BL, participants complete a 20 min IVR training to help them better understand the survey questions. Participants are taught to record drinking data (in standard drinks) and provided the questions and response options administered by the IVR system in a paper format. They also complete a practice survey while the interviewer is in the room so that they are able to familiarise themselves with the survey and ask any questions. Participants are given a wallet-sized ‘pocket reference’ card that contains their unique subject ID (needed when completing the surveys), phone number for the study site and the date range of their IVR monitoring period. IVR monitoring begins the next morning and is scheduled to coincide with the participant's sleep-wake cycle.

    BL assessment

    Measures with known psychometric properties were selected when possible. Most measures listed below are standard instruments commonly used in smoking and/or alcohol research studies.

    Demographics and health history form

    We collect information about age, race, ethnicity, income, employment and educational information, medical history (eg, diabetes, cancer, heart disease, etc), and health habits (eg, frequency of exercise and nutritional diet).

    Prior smoking cessation experiences

    Participants complete an assessment of prior smoking cessation experiences, including the number of previous quit attempts; longest (ever) quit attempt; number of quit attempts lasting at least 1 week, 1 month and 3 months; quit aids used (eg, nicotine gum, patch, etc) and reasons for relapse (eg, friends’ smoke, social pressure and life stressor).

    Use of other tobacco products

    Participants are asked about lifetime use of chewing tobacco, dip, pipe, snus, dissolvable products, hookah, little cigars/cigarillos/bidis, large cigars and e-cigarettes. Items ask about age of first use, frequency of use, period of use, and time since last use. This form was developed expressly for this study.

    Nicotine dependence

    The Fagerström Test for Nicotine Dependence (FTND31) is a 6-item questionnaire used to assess nicotine dependence. Respondents answer yes/no to questions about difficulty to refrain from smoking, frequency of smoking after awakening and smoking while ill. Participants also answer three categorical questions about how soon after awakening they smoke, which cigarette they would most hate to give up and how many cigarettes they smoke per day. A total score is calculated with scores ranging from 0 to 10. The FTND has demonstrated high reliability, strong validity and good internal consistency in samples of daily smokers.32

    Alcohol and substance use disorder diagnoses

    The Structured Clinical Interview for (Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition) DSM-IV (SCID33) is used to assess lifetime and current diagnoses for substance and alcohol use disorders. Sections have been modified to obtain information specific to the diagnosis of nicotine dependence.

    Motivation to change smoking

    The Contemplation Ladder (CL34 ,35) is a single item questionnaire designed to assess one's level of contemplation about quitting smoking on a 0 to 10 scale, where 0=no thoughts of quitting and 10=taking action to quit. The CL has shown good convergent validity with other measures of motivation to change and predicts longer term readiness to quit smoking.35 ,36

    The Readiness to Change Questionnaire (RTCQ37) is a 15-item measure that assesses readiness to change current smoking patterns based on stages of change model (precontemplation, contemplation and action). This measure is adapted from a similar measure that examines readiness to change alcohol use. Participants respond to each item using a five-point scale where −2=strongly disagree and 2=strongly agree. This measure has shown good internal and test–retest reliability in excessive drinkers.38

    Motivation to change drinking

    The Stages of Change and Treatment Readiness Scale (SOCRATES39) is a 19-item questionnaire that is used to assess readiness to change drinking and drug use habits. All responses are measured on a five-point Likert scale where 1=strongly disagree and 5=strongly agree. This measure is made up of three subscales: Take steps, measures of how much a person is already making changes to their alcohol or drug use; Recognition assesses whether a person acknowledges that they have a problem with alcohol or drugs; Ambivalence assesses if the person is contemplating whether or not they have an alcohol or drug problem. This measure demonstrates excellent test–retest reliability in drinkers and good convergent validity with measures of longer term drinking.39

    SE to quit smoking

    The Smoking Self-Efficacy Scale (SSE40) is a 9 item measure that assesses confidence in the ability to abstain from smoking in certain high-risk situations. Respondents use a five-point Likert scale to answer questions about how tempted they would be to smoke in certain scenarios where 1=not at all tempted and 5=extremely tempted. This measure is comprised of three subscales. The Positive/Social scale assesses how tempted a person may be to smoke in social settings. The Negative/Affective scale assesses how tempted a person may be to smoke when experiencing negative feelings. The Habit/Addictive scale assesses how tempted a person is to smoke in situations where many smokers use cigarettes. A total score may also be calculated to examine overall SE, with higher scores indicating lower SE. The SSE has demonstrated good reliability in the current smokers and drinkers seeking treatment.40

    Cigarette, alcohol and other drug use consumption patterns

    The Time-Line Follow-Back Interview (TLFB41) yields indices for quantity and frequency of alcohol, cigarette, drug and other tobacco use in the 90 days prior to the BL assessment. The TLFB has shown high test–retest reliability, and strong correlations between participant and collateral reports of drinking and smoking.42 ,43

    Alcohol use treatment experiences

    Prior treatment utilisation for alcohol and other mental health problems, and barriers to seeking treatment are examined, using a form that was developed for this study. Participants select from a list of 11 different types of providers (eg, alcoholics anonymous, family services or social service agency, inpatient ward) they may have seen related to their drinking or emotional problems, or both in the past 12 months. Participants are also asked to indicate from a list of options potential barriers for not seeking treatment for alcohol problems and mental health problems.

    Post-IVR Survey

    The Post-IVR Survey is an 11-item survey designed to assess behaviour and attitudinal change (reactivity) that may have occurred over the course of 28 days in response to IVR monitoring and was designed expressly for this study. Reactivity was not an intended goal of the study; however, it could be an important piece of measurement bias impacting outcomes. Items query about the extent to which participants may have become more aware of certain behaviours, the degree to which they may have purposefully started to make changes to their behaviour (and the specific behaviours associated with those changes), as well as feasibility and satisfaction of completing the daily surveys (length of time to complete each survey, burden of monitoring schedule, degree of difficulty completing the surveys, etc).

    IVR measurement and procedure

    For 28 days following the BL, participants report their daily alcohol use, cigarette smoking, SE to quit smoking, and cravings to smoke and drink in twice daily interviews that are happen in response to two random prompts per day (eg, calls to their telephone). Prompts are programmed to coincide with the participant's sleep-wake cycle, during each of the two 4 h time intervals, one in the morning and the evening. The IVR system is configured to call (prompt) the participant's telephone and is enabled such that participants may directly access the survey after they receive the call by pressing ‘1’ on their telephone keypad. Prompting lasts for 10 s, and the participant has 2 min to respond. If the prompt is missed, the IVR system cues another prompt 5 min later. If the participant does not respond after the third prompt (15 min after the initial prompt), the trial is recorded as missed. The IVR system is set up so that no random prompts are issued within 2 h of each other; separate morning and evening interviews are programmed to facilitate different questions at each time period. IVR interviews are last 7–10 min, are date and time-stamped, and recorded immediately.

    Daily factors are measured in three areas: (1) Drinking, including (a) frequency of beer, wine and liquor consumed in standard drink conversions and (b) quantity of the same; (2) Smoking, measured as number of cigarettes consumed since the prior assessment; and (3) SE to quit smoking, adapted from the SSE (0 ‘Strongly disagree’ to 4 ‘Strongly agree’). Factors known to correlate with drinking, smoking and SE (stressful events, mood and cravings) are also included in the daily assessments.

    Several system features are enabled to promote adherence, including clear prompts, minimal skip outs, ability to return to questions, time and date stamping, and reminder phone calls from study personnel if calls are recorded as missed for one full day.44

    To enhance IVR compliance, participants receive $15 per week for the IVR phase and they can earn additional bonus incentives of $2 per week for completing prompts for 6 of 7 days, or $5 per week for completing prompts for all 7 days of the week. Thus, participants may receive up to $108 if they complete all IVR interviews. Participants are given the option of receiving IVR payment in cash, cheque or gift card (to a local convenience or grocery store).

    Follow-up assessments

    1-month follow-up

    Participants are contacted at the end of the daily monitoring phase for a 10 min interview of smoking status, desire to change smoking and drinking, SE to quit smoking and potential reactivity (changes in behaviour) to the IVR assessment. Participants have the option of completing the interview by phone, in-person or via the web. Participants are paid $15 for the 1-month follow-up.

    6-month follow-up

    Participants are contacted for a final follow-up 6 months post-BL for an in-person assessment of smoking status (self-reported 7-day point prevalence abstinence) and other factors. When possible, self-reported abstinence is verified by carbon monoxide analysis of breath samples (10 ppm cut-off) for stated abstinence of 24 h–2 weeks. Detected values above the stated cut-off scores will be considered indicative of smoking.

    In addition to abstinence outcomes, we also assess separate smoking behaviour change outcomes, defined as an altering of smoking behaviour with the intent or mandate to reduce or abstain, based on reports of one (or more) of the following (dichotomous yes/no): (1) reduced smoking by 50%, (2) attempt at cessation lasting more than 1 day and less than 7 days, and (3) last month continuous abstinence from smoking (yes/no). We also include duration (number of days/weeks) of cessation attempts and type of intervention/treatment utilised (nicotine gum, patch and medication). We also reassess alcohol and cigarette consumption, drug use and other tobacco product use (in the past 30 days), diagnostic status (alcohol use disorder and nicotine dependence), desire to change alcohol and smoking, SE to quit smoking, nicotine dependence and treatment utilisation for alcohol problems in the past 6 months. Participants who do not engage in cessation attempts or who increase their smoking are asked to identify and rate the degree to which certain barriers may have influenced their behaviour.

    Participants who are difficult to reach may complete follow-up surveys via mail, web or over the telephone. We are not able to verify smoking status for these individuals unless they are able to come to the study site.

    Design considerations

    We considered several alternative designs for this study. We had considered using a 12-month follow-up to maximise the possibility of capturing behaviour change. However, once we mapped out participant flow for this design, it did not seem feasible within the 2-year time limit of the grant mechanism. Studies show that up to 60% of smokers will make significant changes to their cigarette consumption or desire to quit in as little as 6 months.44–46 We have selected a sample that is high on desire to change to increase the likelihood that we will detect some change.

    We considered adding a no-IVR group to control for the possibility that monitoring may cue people to reduce their drinking or smoking (ie, reactivity). However, research indicates that reactivity is highly unlikely, especially when multiple behaviours are being monitored, as is the case in this particular study.25 ,47 ,48 We will include recommended analytical ‘checks’ of reactivity48 and other questions on our IVR survey, aside from those about smoking and drinking to dilute the potential impact of daily recording.49 As one additional method to assess for reactivity, at the conclusion of the IVR phase, the participants are queried about IVR's potential role in stimulating behaviour change.

    Analyses

    Outcomes

    Primary outcome: The main outcome analyses will be based on abstinence for at least 7 days prior to the 6-month follow-up. Self-reported abstinence reported at 6 months will be verified in person by carbon monoxide analysis of breath samples (10 ppm cut-off), if available.

    Secondary outcomes: Secondary outcomes will be based on reports of smoking and drinking behaviour, SE to quit smoking, nicotine dependence, and motivation to change drinking and smoking collected at 1 and 6 months post-BL. We will also include separate smoking behaviour change outcomes (reduced cigarettes per day, cessation attempts and use of NRT) and barriers to change smoking as secondary outcomes.

    Sample size

    Sample size projections for the daily diary analysis were based on small to medium effect sizes from prior cross-sectional and longitudinal studies between the predictors of interest (alcohol use, SE) and smoking. We are well powered to detect adequate effect sizes for all within-person/daily associations, assuming an 85% IVR compliance rate (based on daily diary studies with similar methodologies and samples44 ,50–56). Power analysis of the 1 and 6-month follow-ups were based on results from naturalistic studies of smoking in ‘untreated’ samples, indicating that between 20% and 40% of smokers will reduce their daily cigarette consumption or quit completely, and another 20% will increase their motivation to change in as little as 6 months.46 ,57–61 We assume a 15% attrition rate at the 6-month follow-up (N=71), based on longitudinal studies with smokers using a similar follow-up period,31 and we expect that approximately 60% of our study sample (n=43) will engage in some form of change by 6 months. G*Power 3.2.1 revealed that 84 participants would provide 0.80 power (α=0.05) to detect clinically significant health gains (a medium effect size62) for BL and daily data derived slopes on the 1 and 6-month outcomes of interest.

    Primary analyses

    We will first assess patterns of missing data, attrition rates, distributional properties of dependent and other measures and correlations among all measures. Analysis of IVR data will use hierarchical linear modelling (HLM).63–65 HLM provides flexibility in handling missing data65 even when data are missing at random (MAR).66 ,67 Models for aim 1 will examine the within-person daily association between drinking and smoking (drinking-smoking concordance). The within-person slope capturing drinking-smoking concordance will be saved in SPSS, and a regression model (continuous or binary logistic depending on the outcome variable) will be conducted to predict the 1 and 6-month outcome of interest from the within-person slope. Models for aim 2 will examine effects of drinking, smoking and the drinking×smoking interaction at time t (eg, morning) predicting SE at time t+1 (eg, morning predicting evening). All models will control for levels of drinking and smoking occurring in the previous interval (t−1), day of the week, and other relevant BL and demographic factors. The drinking×smoking term captures the association between smoking and SE as a function of higher or lower levels of alcohol use. Estimates of within-person slopes will be saved and imported into SPSS. Next, regression models (continuous or binary logistic) will predict the 1 and 6-month outcomes of interest from the within-person slopes (all models will control for mean levels of drinking and smoking). Mediation will be reflected by a reduction in the association between drinking-smoking concordance and smoking outcomes after including the drinking×smoking interaction on daily SE slope. To reduce the set of covariates, variables with p<0.15 will be retained in final models. To address missing data, we will control for potential variables related to missing data, as well as examine multiple imputation methods for handling missing data (expectation maximisation algorithm).48 ,63–66

    Secondary/exploratory analyses

    Effects of drinking on smoking over the course of hours (morning to evening), days or weeks (weekends vs weekdays) will be assessed. We will also examine changes in SE over the course of hours or days (creating an average change score for each person) and within-person variability in SE by calculating the SD of SE for each person.

    Ethics and dissemination

    Explicit informed consent is obtained from each individual prior to participation in the study. All participants are informed that they may withdraw from the study at any time without penalty and are be reimbursed for the portion of the study that they have completed up to that point. All participant data will be kept confidential and will be identified only be a unique identifier.

    Even though this is a low risk observational study, because the sample is comprised of heavy drinkers, there are several important ethical and safety issues. First, to provide additional safeguards to data confidentiality, we have obtained a CoC from the National Institutes of Health to protect participant identities and disclosure of illegal activity (drinking) for those under the age of 21. Given the high comorbidity of smoking and alcohol use with other mental health disorders, participants are given referrals to nearby treatment facilities if requested or deemed appropriate. Study staff have on-hand a list of referral sources to provide participants if they are interested.

    Manuscript and conference submissions will assist with dissemination of results from this study and will provide the necessary preliminary evidence to enhance grant application success to fund a future treatment development study.

    Discussion

    Major improvements in helping smokers quit will be found in the development of specialised treatments that capitalise on altering the ‘momentary’ association among risk factors that maintain smoking. Thus, findings from this proposal should provide useful pilot data for developing a low cost, minimally invasive intervention that could be used to increase smoking related SE in risky-drinking smokers, and deliver this intervention ‘in real time’ and on a platform that is free and easily accessible: one's cell phone.

    Study status

    Study recruitment began in February 2014. The target sample size is N=84. As of the time of this submission (October 2014), 404 participants had been screened for the study and 63 had completed the BL assessment and been enrolled in the IVR. Recruitment is expected to be completed by February 2015.

    References

    1. ↵
      American Cancer Society. Cancer facts and figures—1999. American Cancer Society, 1999.
    2. ↵
      United States Departmetn of Health and Human Services. The health benefits of smoking cessation. A report of the surgeon general. U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control, Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health, 1990. DHHS Publication No. (CDC) 90-8416.
    3. ↵
      American Cancer Society. Cancer facts & figures 2011. Atlanta: American Cancer Society, 2011.
    4. ↵
      1. Falk DE,
      2. Yi H,
      3. Hiller-Sturmhöfel S
      . An epidemiologic analysis of co-occurring alcohol and tobacco use and disorders. Alcohol Res Health 2006;29:162–71.
      OpenUrlPubMedWeb of Science
    5. ↵
      1. Kessler RC,
      2. Chiu WT,
      3. Demler O, et al
      . Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiat 2005;62:617–27. doi:10.1001/archpsyc.62.6.617
      OpenUrlCrossRefPubMedWeb of Science
    6. ↵
      1. Hasin DS,
      2. Stinson FS,
      3. Ogburn E, et al
      . Prevalence, correlates, disability, and comorbidity of DSM-IV alcohol abuse and dependence in the United States: results from the National Epidemiologic Survey on alcohol and related conditions. Arch Gen Psychiat 2007;64:830–42. doi:10.1001/archpsyc.64.7.830
      OpenUrlCrossRefPubMedWeb of Science
    7. ↵
      1. Brandon TH,
      2. Tiffany ST,
      3. Obremski KM, et al
      . Postcessation cigarette use: the process of relapse. Addict Behav 1990;15:105–14. doi:10.1016/0306-4603(90)90013-N
      OpenUrlCrossRefPubMedWeb of Science
    8. ↵
      1. Mayne ST,
      2. Morse DE,
      3. Winn DM
      . Cancers of the oral cavity and pharynx. Cancer Epidemiol Prev 2006;1:674–97. doi:10.1093/acprof:oso/9780195149616.003.0035
      OpenUrl
    9. ↵
      1. Pelucchi C,
      2. Gallus S,
      3. Garavello W, et al
      . Cancer risk associated with alcohol and tobacco use: focus on upper aero-digestive tract and liver. Alcohol Res Health 2006;29:193–8.
      OpenUrlPubMedWeb of Science
    10. ↵
      1. Purdue MP,
      2. Hashibe M,
      3. Berthiller J, et al
      . Type of alcoholic beverage and risk of head and neck cancer—a pooled analysis within the INHANCE Consortium. Am J Epidemiol 2009;169:132–42. doi:10.1093/aje/kwn306
      OpenUrlAbstract/FREE Full Text
    11. ↵
      1. Lubin JH,
      2. Purdue M,
      3. Kelsey K, et al
      . Total exposure and exposure rate effects for alcohol and smoking and risk of head and neck cancer: a pooled analysis of case-control studies. Am J Epidemiol 2009;170:937–47. doi:10.1093/aje/kwp222
      OpenUrlAbstract/FREE Full Text
    12. ↵
      1. Blot WJ,
      2. McLaughlin JK,
      3. Winn DM, et al
      . Smoking and drinking in relation to oral and pharyngeal cancer. Cancer Res 1988;48:3282–7.
      OpenUrlAbstract/FREE Full Text
    13. ↵
      1. Mokdad AH,
      2. Marks JS,
      3. Stroup DF, et al
      . Actual causes of death in the United States, 2000. JAMA 2004;291:1238–45. doi:10.1001/jama.291.10.1238
      OpenUrlCrossRefPubMedWeb of Science
    14. ↵
      1. Hart CL,
      2. Smith GD,
      3. Gruer L, et al
      . The combined effect of smoking tobacco and drinking alcohol on cause-specific mortality: a 30year cohort study. BMC Public Health 2010;10:789–800. doi:10.1186/1471-2458-10-789
      OpenUrlCrossRefPubMed
    15. ↵
      1. Brandon TH,
      2. Herzog TA,
      3. Irvin JE, et al
      . Cognitive and social learning models of drug dependence: implications for the assessment of tobacco dependence in adolescents. Addiction 2004;99:51–77. doi:10.1111/j.1360-0443.2004.00737.x
      OpenUrlCrossRefPubMedWeb of Science
    16. ↵
      1. Gwaltney CJ,
      2. Metrik J,
      3. Kahler CW, et al
      . Self-efficacy and smoking cessation: a meta-analysis. Psychol Addict Behav 2009;23:56–81. doi:10.1037/a0013529
      OpenUrlCrossRefPubMedWeb of Science
    17. ↵
      1. Bandura A
      . Self-efficacy: toward a unifying theory of behavioral change. Psych Review 1977;84:191–215. doi:10.1037/0033-295X.84.2.191
      OpenUrlCrossRef
    18. ↵
      1. Kahler CW,
      2. Spillane NS,
      3. Metrik J
      . Alcohol use and initial smoking lapses among heavy drinkers in smoking cessation treatment. Nicotine Tob Res 2010;12:781–5. doi:10.1093/ntr/ntq083
      OpenUrlAbstract/FREE Full Text
    19. ↵
      1. Dunbar MS,
      2. Scharf D,
      3. Kirchner T, et al
      . Do smokers crave cigarettes in some smoking situations more than others? Situational correlates of craving when smoking. Nicotine Tob Res 2010;12:226–34. doi:10.1093/ntr/ntp198
      OpenUrlAbstract/FREE Full Text
    20. ↵
      1. Niaura RS,
      2. Rohsenow DJ,
      3. Binkoff JA, et al
      . Relevance of cue reactivity to understanding alcohol and smoking relapse. J Abnorm Psychol 1988;97:133–52. doi:10.1037/0021-843X.97.2.133
      OpenUrlCrossRefPubMedWeb of Science
    21. ↵
      1. Gwaltney C,
      2. Shiffman S,
      3. Norman G, et al
      . Does smoking abstinence self-efficacy vary across situations? Identifying context-specificity within the Relapse Situation Efficacy Questionnaire. J Consult Clin Psych 2001;69:516–27. doi:10.1037/0022-006X.69.3.516
      OpenUrlCrossRefPubMedWeb of Science
    22. ↵
      1. Banks H,
      2. Rehm KL,
      3. Sutton KL, et al
      . Dynamic modeling of behavior change in problem drinkers. Center for Research in Scientific Computation Technical Report, CRSC-TR10-19. 2010.
    23. ↵
      1. Barta W,
      2. Tennen H
      . An idiographic-nomothetic approach to personality and psychological influences on health risk behavior. Person Soc Psych Res 2008:13–59.
    24. ↵
      1. Nelson RO,
      2. Hayes SC
      . The nature of behavioral assessment: a commentary. J Appl Behav Anal 1979;12:491–500. doi:10.1901/jaba.1979.12-491
      OpenUrlCrossRefPubMedWeb of Science
    25. ↵
      1. Shiffman S
      . Ecological momentary assessment (EMA) in studies of substance use. Psychol Assess 2009;21:486–97. doi:10.1037/a0017074
      OpenUrlCrossRefPubMedWeb of Science
    26. ↵
      1. Kazdin AE,
      2. Nock MK
      . Delineating mechanisms of change in child and adolescent therapy: methodological issues and research recommendations. J Child Psychol Psyc 2003;44:1116–29. doi:10.1111/1469-7610.00195
      OpenUrl
    27. ↵
      1. Todd M,
      2. Armeli S,
      3. Tennen H, et al
      . Drinking to cope: a comparison of questionnaire and electronic diary reports. J Stud Alcohol 2005;66:121–9.
      OpenUrlPubMed
    28. ↵
      1. Carney MA,
      2. Tennen H,
      3. Affleck G, et al
      . Levels and patterns of alcohol consumption using timeline follow-back, daily diaries and real-time. J Stud Alcohol 1998;59:447–54.
      OpenUrlPubMedWeb of Science
    29. ↵
      1. Shiffman S,
      2. Hufford M,
      3. Hickcox M, et al
      . Remember that? A comparison of real-time versus retrospective recall of smoking lapses. J Consult Clin Psych 1997;65:292–300. doi:10.1037/0022-006X.65.2.292.a
      OpenUrlCrossRefPubMedWeb of Science
    30. ↵
      NIAAA. The National Institute on alcohol abuse and alcoholism five year strategic plan: FY09–14 “Alcohol Across the Lifespan”. National Institute on Alcohol Abuse and Alcoholism, 2008.
    31. ↵
      1. Heatherton TF,
      2. Kozlowski LT,
      3. Frecker RC, et al
      . The Fagerström test for nicotine dependence: a revision of the Fagerstrom Tolerance Questionnaire. Brit J Addict 1991;86:1119–27. doi:10.1111/j.1360-0443.1991.tb01879.x
      OpenUrlCrossRef
    32. ↵
      1. Pomerleau CS,
      2. Carton SM,
      3. Lutzke ML, et al
      . Reliability of the Fagerstrom tolerance questionnaire and the Fagerstrom test for nicotine dependence. Addict Behav 1994;19:33–9. doi:10.1016/0306-4603(94)90049-3
      OpenUrlCrossRefPubMedWeb of Science
    33. ↵
      1. First MB,
      2. Spitzer RL,
      3. Gibbon M, et al
      . Structured clinical interview for DSM-IV axis I disorders, patient edition, January 1995 FINAL. (SCID-I/P Version 2.0). New York, NY: Biometrics Research Department, New York State Psychiatric Institute, 1995.
    34. ↵
      1. Biener L,
      2. Abrams D
      . The contemplation ladder: a measure of intention to stop smoking. Health Psychol 1991;10:360–5. doi:10.1037/0278-6133.10.5.360
      OpenUrlCrossRefPubMedWeb of Science
    35. ↵
      1. Biener L,
      2. Abrams DB
      . The Contemplation Ladder: validation of a measure of readiness to consider smoking cessation. Health Psychol 1991;10:360–5. doi:10.1037/0278-6133.10.5.360
      OpenUrlCrossRefPubMedWeb of Science
    36. ↵
      1. Amodei N,
      2. Lamb R
      . Convergent and concurrent validity of the Contemplation Ladder and URICA scales. Drug Alcohol Dependence 2004;73:301–6. doi:10.1016/j.drugalcdep.2003.11.005
      OpenUrlCrossRefPubMedWeb of Science
    37. ↵
      1. Rollnick S,
      2. Heather N,
      3. Gold R, et al
      . Development of a short ‘readiness to change’ questionnaire for use in brief, opportunistic interventions among excessive drinkers. Brit J Addict 1992;87:743–54. doi:10.1111/j.1360-0443.1992.tb02720.x
      OpenUrl
    38. ↵
      1. Heather N,
      2. Rollnick S,
      3. Bell A
      . Predictive validity of the Readiness to Change Questionnaire. Addiction 1993;88:1667–77. doi:10.1111/j.1360-0443.1993.tb02042.x
      OpenUrlCrossRefPubMedWeb of Science
    39. ↵
      1. Miller WR,
      2. Tonigan JS
      . Assessing drinkers’ motivation for change: the Stages of Change Readiness and Treatment Eagerness Scale (SOCRATES). Psychol Addict Behav 1996;10:81–9. doi:10.1037/0893-164X.10.2.81
      OpenUrlCrossRefWeb of Science
    40. ↵
      1. Velicer WF,
      2. DiClemente CC,
      3. Rossi JS, et al
      . Relapse situations and self-efficacy: an integrative model. Addict Behav 1990;15:271–83. doi:10.1016/0306-4603(90)90070-E
      OpenUrlCrossRefPubMedWeb of Science
    41. ↵
      1. Holt LJ,
      2. Litt MD,
      3. Cooney NL
      . Prospective analysis of early lapse to drinking and smoking among individuals in concurrent alcohol and tobacco treatment. Psychol Addict Behav 2012;26:561–72. doi:10.1037/a0026039
      OpenUrlCrossRefPubMed
    42. ↵
      1. Sobell LC,
      2. Sobell MB
      . Timeline Followback user's guide: a calendar method for assessing alcohol and drug use. Toronto: Addiction Research Foundation, 1996.
    43. ↵
      1. Brown RA,
      2. Burgess ES,
      3. Sales SD, et al
      . Reliability and validity of a smoking timeline follow-back interview. Psychol Addict Behav 1998;12:101–12. doi:10.1037/0893-164X.12.2.101
      OpenUrlCrossRefWeb of Science
    44. ↵
      1. Mundt JC,
      2. Searles JS,
      3. Perrine M, et al
      . Conducting longitudinal studies of behavior using interactive voice response technology. Int J Speech Technol 1997;2:21–31. doi:10.1007/BF02539820
      OpenUrl
    45. ↵
      1. Sheffer CE,
      2. Stitzer M,
      3. Brandon T, et al
      . Effectiveness of adding relapse prevention materials to telephone counseling. J Subst Abuse Treat 2010;39:71–7. doi:10.1016/j.jsat.2010.03.013
      OpenUrlCrossRefPubMed
    46. ↵
      1. Glasgow RE,
      2. Morray K,
      3. Lichtenstein E
      . Controlled smoking versus abstinence as a treatment goal: the hopes and fears may be unfounded. Behav Ther 1990;20:77–91. doi:10.1016/S0005-7894(89)80119-4
      OpenUrl
    47. ↵
      1. Hufford M,
      2. Shields A,
      3. Shiffman S, et al
      . Reactivity to ecological momentary assessment: an example using undergraduate problem drinkers. Psychol Addict Behav 2002;16:205–11. doi:10.1037/0893-164X.16.3.205
      OpenUrlCrossRefPubMedWeb of Science
    48. ↵
      1. Barta W,
      2. Tennen H,
      3. Litt MD
      . Measurement reactivity in diary research. In: Mehl MR, Conner TS, eds. Handbook of research methods for studying daily life. New York: Guilford Press, 2011:108–23.
    49. ↵
      1. Tennen H,
      2. Affleck G
      . The challenge of capturing daily processes at the interface of social and clinical psychology. J Soc Clin Psychol 2002;21:610–27. doi:10.1521/jscp.21.6.610.22796
      OpenUrlCrossRefWeb of Science
    50. ↵
      1. Corkrey R,
      2. Parkinson L
      . Interactive voice response: review of studies 1989–2000. Behav Res Methods Instrum Comput 2002;34:342–53. doi:10.3758/BF03195462
      OpenUrlCrossRefPubMedWeb of Science
    51. ↵
      1. Corkrey R,
      2. Parkinson L
      . A comparison of four computer-based telephone interviewing methods: getting answers to sensitive questions. Behav Res Methods Instrum Comput 2002;34:354–63. doi:10.3758/BF03195463
      OpenUrlPubMedWeb of Science
    52. ↵
      1. Mundt JC,
      2. Bohn MJ,
      3. King M, et al
      . Automating standard alcohol use assessment instruments via interactive voice response technology. Alcohol Clin Exp Res 2002;26:207–11. doi:10.1111/j.1530-0277.2002.tb02525.x
      OpenUrlCrossRefPubMedWeb of Science
    53. ↵
      1. Searles JS,
      2. Helzer JE,
      3. Walter DE
      . Comparison of drinking patterns measured by daily reports and timeline follow back. Psychol Addict Behav 2000;14:277–86. doi:10.1037/0893-164X.14.3.277
      OpenUrlCrossRefPubMedWeb of Science
    54. ↵
      1. Oslin DW,
      2. Cary M,
      3. Slaymaker V, et al
      . Daily ratings measures of alcohol craving during an inpatient stay define subtypes of alcohol addiction that predict subsequent risk for resumption of drinking. Drug Alcohol Depen 2009;103:131–6. doi:10.1016/j.drugalcdep.2009.03.009
      OpenUrlPubMed
    55. ↵
      1. Shiffman S,
      2. Paty JA,
      3. Gnys M, et al
      . First lapses to smoking: within-subjects analysis of real-time reports. J Consult Clin Psych 1996;64:366–79. doi:10.1037/0022-006X.64.2.366
      OpenUrlCrossRefPubMedWeb of Science
    56. ↵
      1. Stone AA,
      2. Shiffman S,
      3. Schwartz JE, et al
      . Patient compliance with paper and electronic diaries. Control Clin Trials 2003;24:182–99. doi:10.1016/S0197-2456(02)00320-3
      OpenUrlCrossRefPubMedWeb of Science
    57. ↵
      1. Hughes JR
      . Reduced smoking: an introduction and review of the evidence. Addiction 2000;95:3–7. doi:10.1080/09652140032008
      OpenUrlCrossRef
    58. ↵
      1. Hughes JR,
      2. Callas PW,
      3. Peters EN
      . Interest in gradual cessation. Nicotine Tob Res 2007;9:671–5. doi:10.1080/14622200701365293
      OpenUrlAbstract/FREE Full Text
    59. ↵
      1. Hughes JR,
      2. Carpenter MJ
      . Does smoking reduction increase future cessation and decrease disease risk? A qualitative review. Nicotine Tob Res 2006;8:739–49. doi:10.1080/14622200600789726
      OpenUrlAbstract/FREE Full Text
    60. ↵
      1. Hughes JR,
      2. Carpenter MJ
      . The feasibility of smoking reduction: an update. Addiction 2005;100:1074–89. doi:10.1111/j.1360-0443.2005.01174.x
      OpenUrlCrossRefPubMedWeb of Science
    61. ↵
      1. Farkas AJ
      . When does cigarette fading increase the likelihood of future cessation? Ann Behav Med 1999;21:71–6. doi:10.1007/BF02895036
      OpenUrlCrossRefPubMedWeb of Science
    62. ↵
      1. Cohen J
      . A power primer. Psychol Bulletin 1992;112:155–9. doi:10.1037/0033-2909.112.1.155
      OpenUrl
    63. ↵
      1. Bryk A,
      2. Raudenbush S
      . Hierarchical linear models. Newbury Park: Sage, 1992.
    64. ↵
      1. Raudenbush S,
      2. Bryk A
      . Hierarchical linear models: applications and data analysis methods. 2nd edn. Thousand Oaks, CA: Sage, 2002.
    65. ↵
      1. Raudenbush S,
      2. Bryk A,
      3. Cheong Y, et al
      . HLM 6: hierarchical linear and nonlinear modeling. Lincolnwood, IL: Scientific Software International, 2004.
    66. ↵
      1. Little RJ,
      2. Rubin D
      . Statistical analysis with missing data. New York: John Wiley & Sons, 1987.
    67. ↵
      1. Little RJ,
      2. Rubin DB
      . Statistical analysis with missing data. 2002.

    Footnotes

    • Contributors AC and TB contributed to study concept; SA and AC will contribute to statistical analyses; MB and SE contributed to data acquisition; all authors contributed to the review of this manuscript and provided comments. All authors read and approved the final manuscript.

    • Funding This work was supported by the National Cancer Institute grant number R03CA175870.

    • Competing interests None.

    • Ethics approval Institutional Review Board approval for this study was provided by Schulman Associations (IRB # 201304003).

    • Provenance and peer review Not commissioned; peer reviewed for ethical and funding approval prior to submission.

    • Data sharing statement Data are sensitive and first priority in sharing data will be protection of study participants’ privacy. Thus, this will not be a public use dataset. Data will be available for certain types of sharing in accordance with the terms of a data sharing agreement, and only after publication of major findings of the study. Only researchers certified at their institution in protection of human subjects will be considered for access to the data.

    • ↵i We had initially excluded individuals who used other tobacco products. However, on initial screening, we found that a large percentage of smokers who responded to our screening advertisement for ‘smokers who regularly drink’ did report using other tobacco products, such as little cigars/cigarillos or large cigars. As a result, we relaxed this exclusion criterion. We will examine other tobacco product use as a potential covariate in analyses.