j.1360-0443.2010.03201.x

8/13/2019 j.1360-0443.2010.03201.x

1/9

The relationships between menthol cigarette

preference and state tobacco control policies on

smoking behaviors of young adult smokers in the

200607 Tobacco Use Supplements to the Current

Population Surveys (TUS CPS)

Karen Ahijevych & Jodi Ford

The Ohio State University, College of Nursing, Columbus, Ohio, USA

ABSTRACT

Aim To examine relationships between the preference for menthol cigarettes and young adult smoking behaviors,including the extent to which state tobacco control policies moderate these relationships. Design Cross-sectional

design using secondary data from the 200607 Tobacco Use Supplements to the Current Population Surveys (TUS

CPS) surveys appended with 2006 state-policy data. Setting United States nationally representative survey.Partici-

pants A total of 2241 young adult daily smokers and 688 young adult non-daily smokers. Measurements The two

dependent variables of smoking behaviors were smoking first cigarette within 30 minutes of waking (TTF) and number

of cigarettes smoked per day (cpd). Primary independent variables included menthol brand preference and state

tobacco control policies (youth access laws, clean indoor air laws and cigarette excise taxes), adjusting for controls.

Findings Among daily smokers, there were no significant associations between menthol brand preference andTTF or

cpd. However, lower educational attainment, not being in the labor force and the lack of home smoking rules were

associated positively with shorter TTF, being white and the lack of home smoking rules were associated positively with

cpd. Among daily smokers, state excise taxes were associated negatively with higher cpd. Among non-daily smokers,

menthol brand preference was associated positively with shorter TTF, but associations did not vary with state tobaccocontrol policies. Menthol brand preference was not associated significantly with cpd, but male gender, unmarried

status and the lack of home smoking rules were associated positively with greater cpd among non-daily smokers.

Conclusions Young adult non-daily smokers who preferred menthol cigarettes were significantly more dependent

than those who preferred non-menthol cigarettes, as shown through the shorter TTF. Associations between menthol

brand preference and smoking behaviors did not vary with state tobacco control policies.

Keywords Mentholated cigarettes, nicotine dependence, young adult smokers, tobacco control policies.

Correspondence to: Karen Ahijevych, The Ohio State University, College of Nursing, 1585 Neil Avenue, Columbus, Ohio 43210, USA. E-mail:

[email protected]

Submitted 29 July 2010; initial review completed 12 March 2010; final version accepted 6 August 2010

INTRODUCTION

Young adults are at high risk for smoking due to increas-

ing autonomy from parents, reaching legal age to pur-

chase tobacco products, increasing use of other addictive

substances and direct targeting by the tobacco industry

to sustain tobacco use and recapture those who quit

smoking [13]. According to national estimates from

200507, nearly 24% of young adults aged 1824 years

reported currently smoking every day or on some days

the highest smoking prevalence among all age groups

[4]. In addition, young adult smokers preference for

menthol cigarettes increased from 34.1% to 40.8%

between the years 200408 [5], which is concerning,

because research suggests that menthol cigarettes may

be a starter tobacco product and increase nicotine depen-

dence [6]. Evidence of the role of menthol brand prefer-

ence in shaping the smoking behaviors of young adults is

limited, but research has found that adolescent menthol

smokers have45% higherodds of being above themedian

on nicotine dependence and are significantly less likely

to consider quitting smoking than teens who smoked

RESEARC H REPO RT doi:10.1111/j.1360-0443.2010.03201.x

2010 The Authors, Addiction 2010 Society for the Study of Addiction Addiction,105(Suppl. 1), 4654

8/13/2019 j.1360-0443.2010.03201.x

2/9

non-menthol cigarettes [6]. Research has found that

people who quit smoking before the age of 35 have a life

expectancy similar to those who never smoked [7]. There-

fore, identifying variables related to nicotine dependence

among young adult smokers may inform cessation strat-

egies as recommended in the Institute of Medicine (IOM)

report on ending the tobacco problem [8].

However, our understanding of the factors that

prevent smoking or reduce nicotine dependence among

young adults is limited due to a scarcity of research

among this population [9], particularly among higher-

risk groups who already smoke on a daily or occasional

(non-daily) basis, as well as between menthol and non-

menthol users. Research among adolescents and young

adults suggests that state tobacco control policies, par-

ticularly cigarette taxes, may be effective in reducing the

prevalence of smoking and increasing tobacco cessation

[10,11]. However, research also found that increased

cigarette taxes were associated with a reduction in thenumber of cigarettes smoked among young adults

[12,13], but the increased tax also was associated with

young adults switching to cigarette brands with higher

amounts of tar and nicotine [12]. Strong clean indoor

air laws and increasing cigarette prices were associated

with decreased cigarette consumption in a study among

current smokers aged 1580 years [13], while another

found that young adults were more likely to quit smoking

when exposed to more restrictive smoking bans in the

work-site and public places [11]. Youth access policies

and enforcement aim to reduce sales of cigarettes to

minors, but studies provide only limited support of theireffectiveness [14]. However, adolescents obtain most of

their cigarettes from non-retail sources, such as older

peers and parents or through theft; thus the effectiveness

of state policy controls may be attenuated [14].

Because young adulthood is a critical time for the

establishment of nicotine dependence [15], efforts to

reduce smoking among this vulnerable population are

imperative. Further research on the role of state tobacco

control policies and nicotine dependence among young

adults is needed, particularly among specific population

groups and types of users. Therefore, the purpose of

this research was twofold: [1] to examine associationsbetween menthol brand preference and the smoking

behaviors of young adult daily and non-daily smokers

and [2] to examine the extent to which the associations

between menthol brand preference and smoking behav-

iors were moderated by state tobacco control policies.

METHODS

Study design and data sources

A cross-sectional design was employed using secondary

data from the Tobacco Use Supplements to the Current

Population Surveys (TUS CPS)May 2006, August

2006 and January 2007 surveys appended with 2006

state-policy data on youth access laws, clean indoor air

laws and cigarette excise taxes. The TUS is sponsored by

the National Cancer Institute (NCI) and the Centers for

Disease Control and Prevention (CDC) and surveys US

households on smoking and other tobacco use [16]. The

TUS is a supplement to the core CPS that the US Census

Bureau conducts monthly to examine employment status

and socio-demographic factors of US households. For this

study, data on state-level youth access and clean indoor

air laws were derived from NCIs State Cancer Legislative

Database Program (http://www.scld-nci.net) and the

data on state-level cigarette excise taxes were obtained

from the American Lung Association (ALA; State of

Tobacco Control).

Sample

The 200607 TUS CPS surveys were administered to

approximately 267 000 civilian, non-institutionalized

people. The sampling frame for our study was young

adults aged 1824 years who reported smoking daily

(n =2339) or non-daily (n =795), which was defined as

smoking 129 of the last 30 days.This study utilized only

data from self-respondents, because certain items of

interest related to tobacco use were not asked of proxies.

Observations with missing data were excluded from

analysis for a final sample size of 2241 young adult daily

smokers (missing n =98) and 688 non-daily smokers

(missing n = 107) who lived across the 50 states and

Washington, DC. The sample for this study was stratified

by daily smoking and non-daily smoking for theoretical

and methodological reasons. First, because menthol is

thought to be a starter product to greater nicotine depen-

dence [6], we wanted to capture the unique associations

along this continuum. Secondly, non-daily and daily

smokers most probably smoke for different reasons, with

variations in risk and protective characteristics. Thirdly,

although the TUS asks daily and non-daily smokers iden-

tical smoking behavior questions, items for the latter

group refer to the average only on the days that they

smoke, which complicates measurement and estimation

if analyzed simultaneously.

Measures

Dependent variables

Smoking behaviors were operationalized as smoking

within 30 minutes upon waking and the average number

of cigarettes smoked per day. Time to first cigarette was

measured as a categorical variable, with individuals who

smoked their first cigarette within 30 minutes of waking

categorized as yes. The distribution for the average

Young adult smokers and menthol cigarettes 47


8/13/2019 j.1360-0443.2010.03201.x

3/9

number of cigarettes smoked was non-normal even after

taking the log; thus we analyzed this as a continuous

count variable.

Independent variables

Menthol brand preference was measured as a categorical

variable, with yes indicating that young adults preferred

to smoke only menthol brand cigarettes. Individual-level

control variables included key socio-demographic char-

acteristics and household smoking rules. Race and eth-

nicity were categorized as Hispanic, non-Hispanic white,

non-Hispanic black and non-Hispanic other, with the

final category including those participants who self-

identified as non-Hispanic Asian, American Indian and

multi-racial, due to small sample sizes. Foreign birth was

categorized as born in United States versus born outside

United States. Socio-economic position was measured via

three indicatorsemployment status, household incomeand educational attainment. Employment status was cat-

egorized as employed, unemployed and not in the labor

force during the week before the survey. Individuals were

classified by the TUS CPS as not in the labor force if they

were retired, disabled or other status, which included stu-

dents or those participants who were running a house-

hold. Income in the TUS CPS is an ordinal measure of

annual family income that was accrued during the 12

months preceding the survey, including income obtained

from employment, business, farm or rent, dividends,

social security, pensions and interest. We categorized the

income measure into four brackets (as in Table 1). Lowereducational attainment was measured as those partici-

pants who were aged 18 years with less than 11th grade

education or aged 1924 years without a high school

diploma. Presence of smoking rules in the household

were measured categorically as no smoking rules in the

home, some smoking allowed and no smoking allowed at

all. Additional controls included age, gender, marital

status (categorical measure of married versus never

married, divorced, widowed or separated), and for non-

daily smokers a control variable for the number of days

that they smoked per month.

Four state-level variables were examined: 2006 youthaccess tobacco laws, 2006 clean indoor air laws, 2006

cigarette excise tax and 200607 smoking prevalence.

Youth access laws were measured using a composite

developedby Alciati andcolleagues [17] to rate theexten-

siveness to which states legislate tobacco use among

youth. The composite rates nine tobacco control mea-

sures, including minimum age for purchase, sealed

packaging, clerk intervention for purchase, photo identi-

fication for purchase, ban on the sale of tobacco products

through vending machines, ban of free tobacco samples

or rebates, graduated penalties for retailers violating the

youth access laws, random inspections of retailers and

state-wide enforcement of the laws. Six of the items are

ratedfrom0 (noeffectiveprovision) to4 (meetstarget)and

threehavethepossibilitytoscore5forexceedingthetarget

goal for a possible total of 39 on the composite. The state

clean indoorair laws also were measured using a compos-

itedeveloped byChriqui andcolleagues [18] that rates the

extensiveness to which states restrict indoor tobacco use.

The composite rates nine tobacco control measures,

including smoking restrictions in seven site-specific areas

and two provisions on enforcement. Scoring of items

range from 0 (no effective provision) to 4 (meets target)

and if the state exceeds thetarget on six of thenine items,

then they areeligibleto receivean additional 6 pointsfor a

maximum scoreof 42. Dataon stateexcisetaxesrepresent

the tax added to the purchaseof a packof cigarettes inUS

dollars. The variable was logged for the purposes of this

analysis to achieve a normal distribution. Because state

excise taxes increased among several states between datacollection points in the 200607 TUS, we disaggregated

this measure to the individual level to adjust for these

changes. All state-level policy variables were continuous

measures. A state-level measure of the prevalence of

young adults (aged 1829 years) who smoked daily or

non-daily was included as a control [19].

Analysis

Descriptive and multi-level analyses were conducted

using SAS version 9.1 statistical software (SAS Institute,

Cary, NC, USA) and HLM 6.07 (SSI, Inc., Lincolnwood,IL, USA). Multicollinearity was examined and no influ-

ential correlation between variables was found. All

analyses were unweighted as multi-level modeling soft-

ware packages are unable to accommodate replicate

weights [20]. Descriptive analyses were conducted to

yield characteristics of participants. Multi-level analyses

were conducted to examine associations between

menthol brand preferences and the dependent variables

measuring smoking behaviors, including cross-level

interactions with state-level tobacco control policies.

Multi-level random intercept models were analyzed

using the Bernoulli distribution for the binary depen-dent variable (first cigarette smoked within 30 minutes

of waking) and a Poisson distribution corrected for over-

dispersion with a negative binomial distribution for the

count dependent variable (average number of cigarettes

smoked). [We chose this analytical technique because it

allows for random effect modeling inherent to the nested

nature of our data. However, this model will produce a

prediction for the zero count but a zero is not included

in the distribution of the dependent variable, which

ranges from 1 to 60. We analyzed the data using a

zero-truncated negative binomial regression model with

48 Karen Ahijevych & Jodi Ford


8/13/2019 j.1360-0443.2010.03201.x

4/9

robust standard errors and the findings were consistent

between the two techniques.]

RESULTS

Table 1 presents descriptive characteristics of young

adult daily and non-daily smokers. Among young adult

daily smokers, half the sample smoked their first ciga-

rette within 30 minutes of waking and the mean

number of cigarettes smoked was 13.5 (range 160).

Approximately 30% of the young adult daily smokers

reported that they preferred to smoke menthol ciga-

rettes. The majority was non-Hispanic white (80%) and

only 4.5% was foreign-born. In relation to socio-

economic position, 68% were employed, but 43%

reported a low household income and 24% had lower

educational attainment. The sample was nearly 50%male and only 15% was married. Nearly half lived in

households where smoking was not allowed anywhere,

21% lived in households where smoking was allowed

only in certain places and 30% lived in households

where smoking was allowed in all places.

Among non-daily smokers, on the days that they

smoked approximately 10% smoked their first cigarette

within30 minutes upon waking andthe mean numberof

cigarettes smoked was 4.4 (range 130). Nearly 26% of

non-daily smokers preferred menthol cigarettes. Approxi-

mately 70% of the sample was non-Hispanic white and

Table 1 Descriptive findings of socio-demographic characteristics and menthol brand preference of young adult daily and non-daily

smokers, including state tobacco control policiesa.

Daily smokers Non-daily smokers

n = 2241 n = 688

% Mean (SD) % Mean (SD)

Dependent variablesFirst cigarette smoked within 30 minutes of waking 50.3 9.7

Average number of cigarettes smoked daily 13.5 (7.4) 4.4 (4.2)

Independent variables

Menthol brand preference 29.9 25.7

Race

Black 6.4 8.1

Hispanic 7.0 14.5

Other 6.7 8.0

White 79.9 69.4

Foreign birth 4.5 11.5

Socio-economic position

Household income

Missing 7.0 6.1

$024 999 43.1 43.5$2549 999 28.5 26.4

$50 000+ 21.4 24.0

Lower education 24.1 16.7

Employment

Unemployed 12.7 10.8

Not in labor force 19.2 17.7

Employed 68.1 71.5

Age 21.5 (1.9) 21.8 (1.8)

Male 48.9 48.8

Married 14.9 11.9

Lives with parent 28.9 24.7

Number of days smoked per month 13.1 (7.1)

Home smoking rules

Smoking allowed 30.5 11.8Some smoking allowed 21.3 20.6

No smoking allowed 48.2 67.6

State youth access laws 19.3 (6.4) 19.4 (6.3)

State clean air laws 23.3 (12.2) 24.3 (12.1)

State cigarette excise tax $ 0.96 (0.59) 0.99 (0.57)

State logged cigarette excise tax $ -0.27 (0.74) -0.21 (0.71)

State prevalence of current smoking 24.6 (5.7) 23.3 (5.5)

aUnweighted analyses. SD: standard deviation.



8/13/2019 j.1360-0443.2010.03201.x

5/9

11% was foreign-born. Socio-economic composition,

gender and marital status were similar to daily smokers.

Nearly 68% of non-daily smokers lived in households

where smoking was not allowed anywhere, 21% lived inhouseholds where smoking was allowed only in certain

places and 12% lived in households where smoking was

allowed in all places.

Table 2 presents the findings of the multivariate

analyses for both of the smoking behaviors among the

sample of young adult daily smokers. Specifically, in rela-

tion to time to first cigarette, we found no associations

between menthol brand preference and smoking within

30 minutes upon waking.In addition,no significant asso-

ciations between the state tobacco control policies and

timing to first cigarette were found, nor were there any

significant moderating effect of policies on the associa-

tions between menthol brand preference and cigarette

timing. Among control variables, we found that Hispanic

(-0.49, P

8/13/2019 j.1360-0443.2010.03201.x

6/9

significantly to the timing of first cigarette smoked.

Specifically, compared to young adult daily smokers who

lived in households where no smoking was allowed,

young adult daily smokers who lived in households

with no smoking rules were more likely to smoke their

first cigarette within 30 minutes of waking (0.752;

P

8/13/2019 j.1360-0443.2010.03201.x

7/9

smokers of both races [22]. Potential explanations for

different effects of menthol cigarettes and TTF between

daily and non-daily smokers may be due to differences inthe measurement of time to first cigarette and laboratory

settings versus survey design. Specifically, we used a cat-

egorical measure of having smoked the first cigarette

within the first 30 minutes of waking in accordance with

Faganet al. [23], while the two aforementioned studies

utilized continuous measures. Secondly, in relation to

menthol exposure, research has foundbrand variabilityin

the amount of menthol contained in cigarettes [24].

Because menthol additive masks harshness and discom-

fort of inhaling smoke, differences in the amount of

menthol in a cigarette in combination with variation in

smoking topography behaviors may yield variability in

menthol exposure.This maynot be captured adequatelyin

the categorical measure of menthol versus non-mentholcigarette information.

No significant variations in smoking behaviors were

found between menthol users and non-menthol users

based on differences in state tobacco control policies,

which suggests that policies may have similar effects on

smoking behaviors between these two groups. State poli-

cies related to clean air, youth access and cigarette excise

tax had few significant associations with the smoking

behaviors of both daily and non-daily young adult

smokers in our study overall. However, among non-daily

smokers, clean air laws were associated with shorter time

Table 3 Random effects models of menthol brand preference and smoking behaviors of young adult non-daily smokers; 200607

Tobacco Use Supplement to the Current Population Survey (n = 688)a,b.

First cigarette within 30 minutes

of waking

Average number of cigarettes

smoked daily

Binary logit model Poisson model

Coefficient (SE) Coefficient (SE)

Level I fixed effects

Menthol brand preference 0.709 (0.317)* -0.018 (0.110)

Race

Black 0.262 (0.457) -0.022 (0.165)

Hispanic -1.41 (0.574)* -0.445 (0.064)***

Other -0.298 (0.567) -.140 (0.119)

White (reference) ref ref

Foreign birth -0.245 (0.495) -0.054 (0.120)

Socio-economic position

Household income

Missing 0.545 (0.746) 0.218 (0.151)

$024 999 0.697 (0.383) -0.106 (0.114)

$2549 999 0.244 (0.506) 0.100 (0.117)

$50 000+(reference) ref ref

Lower education 0.498 (0.317) 0.123 (0.093)Employment

Unemployed -0.485 (0.507) 0.070 (0.137)

Not in labor force 0.326 (0.347) 0.239 (0.086)**

Employed (reference) ref ref

Age -0.052 (0.083) -0.002 (0.018)

Male 0.727 (0.257)** 0.234 (0.074)**

Married 0.032 (0.553) -0.299 (0.091)**

Lives with parent 0.056 (0.376) 0.051 (0.086)

Number of days smoked per month 0.050 (0.027) 0.022 (0.005)***

Home smoking rules

Smoking allowed 0.620 (0.427) 0.199 (0.097)*

Some smoking allowed 0.330 (0.359) 0.225 (0.081)**

No smoking allowed (reference) ref ref

Level II fixed effectsState youth access laws -0.025 (0.024) -0.004 (0.005)

State clean air laws 0.031 (0.015)* -0.007 (0.004)

State logged cigarette excise tax $ -0.153 (0.257) 0.018 (0.049)

State prevalence of current smoking 0.007 (0.034) 0.006 (0.006)

Random effects

Intercept -3.63 (0.422)*** 1.26 (0.109)***

*P < 0.05; **P < 0.01; ***P < 0.001. aUnweighted analyses. bFinal models. Cross-level interactions non-significant, thus not depicted in model. Available

upon request. SE: standard error.



8/13/2019 j.1360-0443.2010.03201.x

8/9

to first cigarette. In this case, individuals may smoke

earlier in the day in anticipation of being in clean air

facilities. Cigarette tax was the only state policy that had

an effect on cigarettes per day, albeit a very slight reduc-

tion. The limited effect of state tobacco control policies in

our study warrants furtherexamination. First, in relation

to cigarette taxes, price-sensitive smokers may use high

price avoidance strategies such as a low or untaxed venue,

discount or generic cigarette brands and coupons, which

dampen the health impact of higher cigarette prices [25].

Secondly, policy effectiveness may vary based on psycho-

socialfactors, as in onenational study findingthatadoles-

cents who had low self-control were largely unresponsive

to cigarette price[26].Consequently, furtherstudy on how

individual locus of control influences smoking behaviors

and the effectiveness of additional state tobacco control

policies are warranted.Thirdly, young adults maybe more

responsive to social control measures against smoking if

they occur in their immediate environment. Specifically,home smoking bans had a significant impact on smoking

behaviors among our sample of young adults, which is in

accordance with previous research [27]. Thus, increased

emphasis on home smoking bans by peers, parents or

partner may be an effective strategy to modify smoking

behavior in the 1824-year-old age group of daily and

non-daily smokers. Lastly, further research examining

other contextual factors should be considered. For

example, social norms related to menthol brand prefer-

ence, tobacco industry marketing of menthol cigarettes,

and anti-smoking media, including menthol messages

may have more of an influence on the smoking behaviorsof young adults.

Several of our findings related to our control variables

and smoking behaviors of young adult daily and non-

daily smokers warrant further discussion. First, socio-

economic disparities were found with lower educational

attainment associated with shorter time to first cigarette

among daily smokers, while non-daily smokers who were

not in the work force smoked more cigarettes on average

when they smoked compared to their employed peers.

These findings are in accordance with previous research

[9,15,2830] and highlight the need to direct smoking

cessation interventions to socio-economically disadvan-taged populations. Additionally, we found that minorities,

particularly young adult Hispanic smokers and those of

foreign birth, were less likely than their white peers to be

more dependent on nicotine, consistent with previous

research [31]. However, more detailed investigations of

Hispanics are needed, as previous research has found

that health behaviors and health outcomes vary

across Hispanics based on their country of origin, socio-

economic opportunities and acculturation [32,33].

Our study had several limitations. First, the study is

cross-sectional,which precludescausal inferences.Longi-

tudinal analyses on associations between menthol brand

preference and lag effects of tobacco policies on smoking

behaviors may yield more robust findings. Secondly, a

proportion of the TUS CPS respondents were accessed

using random digit dialing computer-assisted telephone

surveys, which may under-represent low-income popula-

tions andthose engaging in higherrisk behaviors, includ-

ing smoking [33]. In both our samples, approximately

50%of the young adults were interviewed in personusing

computer-assisted devices, 43% were interviewed via tele-

phone and 7% were missing on the response to type of

interview. Increasingly, young adults are more likely than

any other age group to have cell phone service only, thus

the use of land-line telephone surveys for this population

should be usedwith caution.Thirdly, ourmeasuresof time

to first cigarette and menthol use were limited to those in

the TUS CPS, which did not include measures of smoking

topography, exhaled carbon monoxide, plasma nicotine

or cotinine concentrations. Consequently, our ability todetect significant associations may have been restricted,

andfuture studies should consider morerefined measures.

Despite these limitations, this study enhances our under-

standing of the role of menthol brand preference in the

smoking behaviors of young adult daily and non-daily

smokers, including no significant differences in smoking

behaviors between menthol and non-menthol users based

on state tobacco control policies. Future research that

examines the role of menthol cigarettes as starter-

products to greater nicotine dependence are needed,

including policy analyses of state tobacco control efforts

targeting menthol use.

Declarations of interest

None.

References

1. Arnett J. J. Emerging adulthood: a theory of development

from the late teens through the twenties. Am Psychol 2000;

55: 46980.

2. Park M. J., Mulye T. P., Adams S. H., Brindis C. D., Irwin C. E.

The health status of young adults in the United States.

J Adolesc Health2006;39: 30517.3. Ling P. M., Glantz S. A. Tobacco industry research

on smoking cessation: recapturing young adults and

other recent quitters. J Gen Intern Med 2004; 19: 419

26.

4. Schoenborn C. A., Adams P. F. Health behaviors of adults:

United States, 20052007. National Center for Health

Statistics. Vital Health Stat 2010; 10(245). Available

at: http://www.cdc.gov/nchs/data/series/sr_10/sr10_245.

pdf. Accessed 4 June 2010. (Archived at WebCite at

http://www.webcitation.org/5tevN98vL on 21 October

2010).

5. Substance Abuse and Mental Health Services Administra-

tion, Office of Applied Studies.The NSDUH Report: Use of



8/13/2019 j.1360-0443.2010.03201.x

9/9

Menthol Cigarettes. Rockville, MD: Substance Abuse and

Mental Health Services Administration, Office of Applied

Studies; 2009.

6. Hersey J. C., Ng S. W., Nonnemaker J. M., Mowery P.,

Thomas K. Y., Vilsaint M.-Y. et al. Are menthol cigarettes a

starter product for youth. Nicotine Tob Res2006; 8: 40313.

7. Doll R., Peto R., Boreham J., Sutherland I. Mortality in rela-

tion to smoking: 50 years observations on male British

doctors.BMJ2004;328: 151927.

8. Institute of Medicine. Ending the Tobacco Problem: A Blue-

print for the Nation. Washington, DC: The National Acad-

emies Press; 2007.

9. Ling P. M., Neilands T. B., Glantz S. A. Young adult smoking

behavior. A national survey. AmJ PrevMed2009; 36: 389

94.

10. Tworek C., Yamaguchi R., Kloska D. D., Emery S., Barker D.

C., Giovino G. A. et al. State-level tobacco control policies

and youth smoking cessation measures. Health Policy 2010;

97: 13644. e-pub ahead of print.

11. Tauras J. A. Public policy and smoking cessation among

young adults in the United States. Health Policy 2004; 68:

32132.

12. Evans W. N., Farrelly M. C. The compensating behavior ofsmokers: taxes, tar, and nicotine. Rand J Econ 1998; 29:

57895.

13. Chaloupka F. J., Wechsler H. Price, tobacco control policies

and smoking among young adults. J Health Econ1997;16:

35973.

14. Dinno A., Glantz S. Tobacco control policies are egalitarian:

a vulnerabilities perspective on clean indoor air laws, ciga-

rette prices, and tobacco use disparities. Soc Sci Med2009;

68: 143947.

15. Green P., McCausland K. L., Xiao H., Duke J. C., Vallone D.

M., Healton C. G. A closer loom at smoking among young

adults: where tobacco controlshouldfocus itsattention. Am

J Public Health 2007;97: 142733.

16. US Department of Commerce, Census Bureau. NationalCancer Institute and Centers for Disease Control and

Prevention Co-sponsored Tobacco Use Supplement to the

Current Population Survey (200607). 2008. http://

riskfactor.cancer.gov/studies/tus-cps/. Data files (AND/OR)

technical documentation (technical documentationweb-

site). Available at: http://www.census.gov/apsd/techdoc/

cps/cpsmayaug06.pdf and/or http://www.census.gov/

apsd/techdoc/cps/cpsjan07.pdf. Accessed 11 November

2009. (Archived at WebCite at http://www.webcitation.

org/5tex2ZfvB and at http://www.webcitation.org/5tex

7W8bu on 21 October 2010).

17. Alciati M. H., Frosh M., Green S. B., Brownson R. C., Rosher

P. H., Hobatrt R. et al. State laws on youth access to tobacco

in the United States: measuring their extensiveness with anew rating system.Tob Control1998;7: 34552.

18. Chriqui J. F., Frosh M., Brownson R. C., Shelton D. M.,

Sciandra R. C., Gobert R. et al. Application of a rating

system to state clean indoor air laws (USA). Tob Control

2002;11: 2634.

19. Giovino G. A., Chaloupka F. J., Hartman A. M. Cigarette

Smoking Prevalence and Policies in 50 States: An Era of

ChangeThe Robert Wood Johnson Foundation ImpacTeen

Tobacco Chart Book. Buffalo, NY: University of Buffalo State

University of New York; 2009.

20. Pierre F., Sadi A. 2008. Implementing resampling methods

for design-based variance estimation in multilevel models:

Using HLM6 and SAS together. 2008 Joint Statistical

Meeting, Section on SurveyResearch Methods. p. 788795.

Available at: http://www.amstat.org/Sections/Srms/

Proceedings/y2008/Files/300724.pdf. Accessed 11

November 2009. (Archived at WebCite at http://www.

webcitation.org/5teyFJqB3 on 21 October 2010).

21. Collins C. C., Moolchan E. T. Shorter time to first cigarette of

the day in menthol adolescent cigarette smokers. Addict

Behav2006;31: 146064.

22. Ahijevych K.,Parsley L. A. Smoke constituent exposure and

stage of change in black and white women cigarette

smokers.Addict Behav1999;24: 11520.

23. Fagan P., Augustson E., Backinger C. L., OConnell M. E.,

Vollinger R. E., Kaufman A.et al. Quite attempts and inten-

tion to quit cigarette smoking among young adults in the

United States.Am J Public Health 2007;97: 141220.

24. Kreslake J. M.,WayneG. F., Alpert H. R.,Koh H. K.,Connolly

G. N. Tobacco industry control of menthol in cigarettes and

targeting adolescents and young adults.Am J Public Health

2008;98: 168592.

25. Hyland A., Bauer J. E., Li Q., Abrams S. M., Higbee C.,

Peppone L.et al. Higher cigarette prices influence cigarettepurchase patterns.Tob Control2005;14: 8692.

26. Fletcher J. M.,Deb P., Sindelar J. L. Tobaccouse, taxation and

self control in adolescence. NBER Working Paper Series,

2009. Available at: http://www.nber.org/papers/w15130

Accessed 12 November 2009. (Archived at WebCite at

http://www.webcitation.org/5teuSLXu0 on 21 October

2010).

27. Clark P. I., Schooley M. W., Pierce B., Schulman J., Hartman

A. M., Schmitt C. L. Impact of home smoking rules on

smoking patterns among adolescents and young adults.

Preventing Chronic Disease [serial online], 2006. Available

from http://www.cdc.gov/pcd/issues/2006/apr/05_0028.

htm Accessed 12 November 2009. (Archived at WebCite

at http://www.webcitation.org/5tex5yyBx on 21 October2010).

28. Hu M.-C., Davies M., Kandel D. B. Epidemiology and corre-

lates of daily smoking and nicotine dependence among

young adults in the United States.Am J Public Health2006;

96: 299308.

29. Yang S., Lynch J., Schulenberg J., Diez-Roux A. V., Raghu-

nathan T. Emergence of socioeconomic inequalities in

smoking and overweight and obesity in early adulthood:

the national longitudinal study of adolescent health. Am J

Public Health2008;98: 46877.

30. Office of Applied Studies. Past month cigarette use among

racial and ethnic groups. National Survey on Drug Use and

Health. 2006. Available at: http://www.oas.samhsa.gov/

2k6/raceCigs/raceCigs.htm (accessed 21 October 2010).31. Zambrana R. E., Carter-Pokras O. Role of acculturation

research in advancing science and practice in reducing

health care disparities among Latinos.Am J Public Health

2010;100: 1823.

32. Lara M., Gamboa C., Kahramanian M. I., Morales L. S.,

Bautista D. E. H. Acculturation and Latino health in the

UnitedStates: a reviewof theliterature andits sociopolitical

context.Annu Rev Public Health 2005;26: 36797.

33. Blumberg S. J., Luke J. V. Coverage bias in traditional

telephone surveys of low-income and young adults. Public

Opin Q2007;71: 73449.



j.1360-0443.2010.03201.x

Documents