Developing Genetic Education for Smoking Cessation

Background. Reducing cigarette smoking among adults is a major public health objective.
Although the target prevalence for adult smoking is < 12% by 2010 (DHHS 2000), current
estimates indicate that 20.8%, or 45.3 million adults, currently smoke (CDC, 2007). An
estimated 44.2% of smokers tried to quit at least one day in the last 12 months (CDC, 2007),
yet <10% of smokers who try to quit remain abstinent (Benowitz, 2008). Even with intensive
treatment, only 11-20% of smokers are abstinent for 6 months (Sutherland, 2003). Smoking
also results in an enormous economic burden, with health-care and lost productivity costs
totaling approximately $167 billion per year (CDC, 2008). A limitation of current treatment
guidelines is a lack of specificity for individual smokers. Completion of the sequencing of
the human genome and the current focus on genomics (Collins, Green, Guttmacher, & Guyer,
2003) provides an unprecedented opportunity to study the biobehavioral factors influencing
smoking. This research will contribute to individualized treatment by providing a better
understanding of the effects of genetic, psychological, and environmental factors and their
interaction on smoking initiation, maintenance, cessation, and relapse (Lerman, Schnoll &
Munafo, 2007; Morgan et al., 2003; Swan et al., 2003). Both candidate gene studies to
identify specific genes that affect smoking and pharmacogenomic studies that investigate the
effects of selected genes on treatment response contribute to a growing body of knowledge
about genetic influences on smoking. The anticipated outcome of these studies will be the
ability to individualize type, dose, and duration of treatment based on smokers' genotypes
(Lerman, Schnoll & Munafo). Ultimately, the long-term goal of this research is to decrease
relapse and promote sustained abstinence.

The Genetic Education and Smoking Model provides a framework for investigating variables
that potentially influence the ways smokers respond to information about: (a) genetic
contributions to smoking and (b) the potential for genetically-informed cessation treatment.
Based on social-cognitive theory and the Common Sense Model (CSM)(Cervone, 2004; Brownlee,
Leventhal, & Leventhal, 2000), the Genetic Education and Smoking Model proposes that genetic
education will influence smokers' mental representations and appraisals about genetics and
smoking in addition to their smoking-related behaviors and affective responses.

As genetically-informed cessation treatment becomes a part of research and clinical care,
smokers will need a degree of genetic literacy to make informed decisions (McInerney, 2002;
Shields, Lerman, & Sullivan, 2004). There is a growing body of scientific knowledge to
guide an educational program about genetic contributions to smoking and the potential for
genetically-informed cessation treatment. Epidemiologic twin studies consistently support a
genetic cause for smoking, with an estimated 56% of the variance for smoking initiation and
67% of progression to nicotine dependence due to additive genetic effects (Sullivan and
Kendler, 1999). Findings from several genome-wide scans using linkage analysis have
identified chromosomes 1, 2, 4, 5, 6, 9, 10, 11, 14, 17, 18, and 21 as containing possible
susceptibility loci for smoking-related genes (Li,Cheng, Ma, & Swan, 2003; Li, Ma, & Beuten,
2004). Several candidate-genes for smoking heritability have been identified, including
nicotinic acetylcholine receptors (nAChRs), dopamine receptor (DRD2 and DRD4) and
transporter (SLC6A3) genes, the catechol-o-methyl-transferase (COMT) gene, the mu opioid
receptor (OPRM1) gene, and the serotonin transporter (5-HTT) gene (Lerman, Schnoll, &
Munafo, 2007; Schnoll, Johnson, & Lerman, 2007). Polymorphisms of CYP2A6, a cytochrome P450
enzyme, play a major role in the inactivation of nicotine to its metabolite, cotinine
(Messina, Tyndale, & Sellers, 1997; Benowitz, 2008).

Specific Aims. The primary specific aim is to compare the effects of the experimental group
to the attention control group on smoking-related mental representations, appraisals,
behaviors, and affective responses over time. The primary aim will be achieved using a
randomized, longitudinal, two-group repeated measures design to determine the direct effects
of the intervention on smoking-related mental representations, appraisals, behaviors, and
affective responses.

The secondary aim is to explore whether personality characteristics (trait negative
affectivity and curiosity) and educational level moderate the effects of the genetic
educational program on smoking-related mental representations, appraisals, behaviors, and
affective responses.

The secondary aim will be achieved with multiple and logistic regression to determine
whether personality traits and educational level moderate the effects of the intervention on
the identified smoking-related outcomes.

Methods. Participants will be assigned randomly to either the experimental or attention
control groups. The groups will be balanced for gender and number of cigarettes smoked per
day using stratified randomization (Friedman, Furberg, & DeMets, 1996). Age and baseline
data for major study variables will be treated as covariates.

The experimental group, referred to as the Genetic Education Session (GES) group, will
receive education about two smoking-related genotypes, DRD2 and CYP2A6. The content of GES
session one is basic genetics and includes concepts and terminology, the impact of the Human
Genome Project on health-care and the use of genetic counseling and testing. The content of
GES session two is research about smoking and genetics. The content focuses on the
multifactorial nature of smoking; research findings about genetic contributions to smoking,
specifically for candidate genotypes DRD2 and CYP2A6; and potential applications of this
research for cessation treatment, including legal, ethical and social implications of
genotyping. To control for an attention placebo effect (Meltzoff, 1998), the control group
will receive information about nutritional guidelines as established by the USDA and FDA.
The attention control group will be referred to as the Nutritional Education Session (NES)
group. The content of NES sessions one and two are aimed at understanding and using dietary
and food safety guidelines. Following the genetic or nutritional education sessions, both
groups will receive a standard cognitive-behavioral smoking cessation intervention with NRT
in the form of transdermal nicotine (TN) patch. This intervention will be referred to as the
standard smoking cessation (SSC) sessions. SSC content focuses on the: (a) nature of
nicotine addiction; (b) use of self-change strategies in the form of self-monitoring,
stimulus identification and control, stress management, and relapse prevention; and (3) use
of nicotine replacement to manage craving and withdrawal symptoms. The SSC sessions consists
of four 2-hour group counseling sessions and one follow-up telephone call after the last
session. The intervention will occur over a 5-week period after the GES or NES. Participants
will receive a 6-week supply of TN patches in the following dosages: 21 mg/24hours for 4
weeks and 14 mg/24 hours for 2 weeks (Fiore Jaen, Baker al., 2008). A 1-week supply of
patches will be distributed during SSC sessions 2-3 and a 2-week supply will be distributed
at session 4. The final two weeks of free TN in 14 mg/24hours strength will be mailed to
participants' homes during week 7. Participants will be encouraged to purchase a two-week
supply of the 7 mg/24hours patches for the final two weeks of an 8-week course of TN. They
will also have the option of using combination therapy as recommended in the 2008 guideline,
to augment the TN with other NRT (gum, spray, inhaler) or bupropion SR. This option will be
explained at the time of obtaining informed consent so participants can visit their primary
care provider to obtain prescription medication (nicotine spray or inhaler or bupropion SR)
prior to the SSC. Data will be collected on use of TN alone or in the recommended
combinations. Expired air CO measurements will be conducted at each Session as
feedback/motivation for participants. Session 1 CO measurement will be used as the baseline
CO data.

Data Collection. Data collection in the form of self report questionnaires will occur at
four time points, Time 1 through Time 4 (T1-T4). Baseline data collection (T1) will be
followed by randomization to the experimental or control group. T2 data will be collected in
two parts (T2a&b) to obtain initial post-intervention data for variables pertaining to each
of these sessions: T2a will occur at the end of the education sessions (GES or NES) and T2b
will occur at the end of the SSC sessions (i.e., 6 weeks). T2a Questionnaires will be
distributed at the end of these sessions and return by the start of the SSC (Week 3). T2b
questionnaires will be distributed at the end of the SSC. Questionnaires will be returned by
postage-paid mail by Week 8. If needed, participants will be reminded to do this at the
follow-up telephone session. Two additional data collections (T3-T4) will occur at twelve
weeks and 6 months after the end of the SSC. T3-T4 data will consist of self-report
questionnaires and in-person measurement of expired air CO with salivary cotinine
verification of abstinence for participants not using NRT and whose CO measurement is <8-10
ppm (Benowitz et al., 2002).

Data Analysis. Preliminary analyses of the dependent variables will include frequency
distributions to look for outliers, descriptive statistics and plots to assess normality,
and bivariate scatterplots to evaluate linearity of relationships among study variables.
Statistical methods of dealing with nonnormality, such as transformations of the raw data,
may be employed prior to analysis. For all scales, internal consistency reliability
coefficients will be estimated. All primary analyses will be based on an intent-to-treat
paradigm. Each participant's data will be analyzed according to their randomized assignment,
regardless of whether they completed all intervention sessions. The analysis plan will be
carried out in order to work out any methodological issues, but because this is not a fully
powered study, significant findings are not expected at the .05 level. A more liberal level
of .10 will be used for significance tests, and emphasis will be placed on descriptive
statistics and estimating effect sizes.

For Aim 1, the primary endpoint of mean change over time relative to baseline will be
compared for the experimental and control groups using Generalized Estimating Equations
(GEE) methodology (Laing & Zeger, 1986). A model will be fit incorporating terms to test the
effects of the intervention and time, covarying for baseline values, for age, and for amount
smoked. An interaction term will also be included to test whether the rate of change over
time is the same for both groups. If difficulties occur with implementing GEE because of the
relatively small number of cohorts per group in the pilot study, the model may be simplified
to a cross-sectional analysis with individuals nested within cohorts at each time point.

For Aim 2, multiple and logistic regression will be used to predict the outcome variables
from the independent variables and tests of moderation will be conducted according to
procedures described by Baron & Kenny (1986).