Nicotine & Tobacco Research, 2016, S41–S48
doi:10.1093/ntr/ntv150
Original investigation
Original investigation
Does a Race-Gender-Age Crossover Effect Exist
in Current Cigarette Smoking Between NonHispanic Blacks and Non-Hispanic Whites?
United States, 2001–2013
Ralph S. Caraballo PhD1, Saida R. Sharapova MD, MPH1,
Katherine J. Asman MSPH2
Office on Smoking and Health, National Center for Chronic Disease and Health Promotion, Centers for Disease
Control and Prevention, Atlanta, GA; 2Biostatistics and Epidemiology Division, RTI International, Atlanta, GA
1
Corresponding Author: Ralph S. Caraballo, PhD, Epidemiology Branch, Office on Smoking and Health, Centers for Disease
Control and Prevention, 4770 Buford Hwy, MS F-79, Atlanta, GA 30341, USA. Telephone: 770-488-5732; Fax: 770-488-5848;
E-mail: [email protected]
Abstract
Introduction: For years, national US surveys have found a lower prevalence of cigarette smoking
among non-Hispanic (NH) black adolescents and young adults than their NH white counterparts
while finding either similar or higher prevalence in NH blacks among older adults. We present
results from four US surveys, including one supplemented by cotinine data, to determine if a racegender-age crossover effect exists between NH black and NH white current cigarette smokers.
Methods: We present NH black and NH white current cigarette smoking estimates in the National
Youth Tobacco Survey (2004–2013), National Survey on Drug Use and Health (2002–2013), National
Health Interview Survey (2001–2013), and National Health and Nutrition Examination Survey
(2001–2012).
Results: All surveys consistently found that NH black females aged 12–25 years had a lower smoking prevalence than NH white females of the same age while NH black males aged 26 years or
older had a higher smoking prevalence than NH white males of the same age. Results were inconsistent between surveys for current smoking estimates for males 12–25 years and females aged
26 years or older.
Conclusion: Our results are inconclusive in consistently detecting the existence of a race-genderage crossover effect for current cigarette smoking between NH blacks and NH whites. National
birth cohort studies are better suited to detect a race-gender-age crossover effect in smoking prevalence between these two racial groups.
Introduction
In the last few decades, national US surveys have consistently found an
overall lower prevalence of cigarette smoking among non-Hispanic
(NH) black adolescents compared with NH white adolescents.1,2 The
Youth Risk Behavior Surveillance System reported prevalence of past
month cigarette smoking among NH black adolescents in 2013 as
8.2% (95% confidence interval [CI] = 6.3% to 10.7%), which is less
than half that of NH white adolescents (18.6%, 95% CI = 15.7%
to 21.9%).3 Similar patterns have been observed for young adults in
other surveys. For example, data from the 2013 National Survey on
Drug Use and Health (NSDUH) showed that current cigarette smoking among young adults aged 18–25 was more prevalent among NH
whites than NH blacks (35.8% vs. 23.9%).4 In adults, the pattern is
reversed with NH blacks showing higher cigarette smoking than NH
Published by Oxford University Press on behalf of the Society for Research on Nicotine and Tobacco 2016.
This work is written by (a) US Government employee(s) and is in the public domain in the US.
S41
S42
Whites. For example, in 2013, the National Health Interview Survey
(NHIS) showed that the prevalence of cigarette smoking among
NH white adults aged 26 years or older was 18.7% compared with
19.5% among NH blacks (unpublished data, the data are available
at www.cdc.gov/nchs/nhis/nhis_2013_data_release.htm).
In a study published in the early 1990’s by Geronimus, Neidert,
and Bound in which the authors assessed NH black and NH white
smoking prevalence for women aged 18–44 years using data from
the 1987 NHIS Cancer Supplement, the authors concluded that
there was an age crossover effect by age 30.5 The age crossover
effect is described as an age-related reversal in prevalence of current
smoking among NH blacks and NH whites where the prevalence is
higher among NH whites than NH blacks in adolescence but lower
in adulthood. In a more recent publication by Kandel, Schaffran,
Hu, and Thomas,6 the authors, using the 2006–2008 NSDUH crosssectional surveys found a crossover age of 29 years old. Even though
Geronimus and colleagues’ study was specific to women of reproductive age living in the United States and Kandel and colleagues’
study included all adults in the United States aged 12–49 years, the
ages of the crossover are strikingly similar and also point to the fact
that this phenomena of a possible age crossover between NH blacks
and NH whites may have been going on at least from the 1980’s.
This age crossover effect between NH blacks and NH whites may
not be unique to cigarette smoking prevalence. Similar observations
have been found for illicit drug use and prescription drugs for nonmedical purposes,7 and for heavy drinking.7,8 To help explain the
specific age of the crossover between NH blacks and NH whites,
Kandel and colleagues assessed in their study various sociodemographic factors that may help explain it. Those factors were gender,
education, marital status, and work status. They found that education had the greatest effect on the specific age of the crossover.
Kandel and colleagues’ study, even though it controlled for gender,
did not specifically separate analyses for males and females nor used
a survey with a biomarker of cigarette smoking that could capture under-reported smoking. Our study adds to the potential age
crossover smoking prevalence literature by specifically looking at
results from several surveys, including one that uses a biomarker to
detect cigarette smoking under-reporting. Our study only intends to
determine if there is an age crossover effect in smoking prevalence
between NH blacks and NH whites by gender, not to identify the
specific age of the crossover.
Even though national surveys have traditionally estimated smoking prevalence using self-reports, which are considered generally
accurate,2 still, self-reported smoking information are based on retrospective recall for a behavior viewed by many as socially undesirable, which may result in providing less than accurate smoking status
information, and resulting in variation in current smoking prevalence estimates across surveys.9–17 Furthermore, several national surveys use differing definitions of smoking status, interview settings,
mode of survey administration, time of the year when the survey is
administered, which, along with other differences, may contribute
to varying prevalence estimates.18–20 For example, school interviews,
especially those that are anonymous, provide higher smoking prevalence estimates among youth than household interviews.21 A number
of additional factors have been offered to explain the lower smoking prevalence in NH black youth, including later age of smoking
onset, parental and peer influence not to smoke, sports involvement,
attendance at religious services, and the cost of cigarettes.1,6,10,22–27
Cotinine, a biomarker of nicotine intake or exposure, is used to
detect cigarette smokers who deny having smoked.28 This method
Nicotine & Tobacco Research, 2016, Vol. 18, Suppl. 1
has its limitations. First, the half-life of serum cotinine is about
16–20 hours, and it is used for identifying active tobacco users
(including cigarette smokers) within past 5 days. Second, majority of
adolescents and some adult smokers (approximately 25%29) do not
smoke cigarettes on a regular basis.30 Thus, cotinine level (in blood,
saliva, or urine) is not a gold-standard method to detect cigarette
smoking in the past 30 days, especially for those who do not smoke
daily.31 Complementing self-reported data with cotinine, as it is possible to do with the National Health and Nutrition Examination
Survey (NHANES), can help provide a more complete understanding of current cigarette smoking in the United States and help determine if in fact there is an age crossover effect in cigarette smoking
prevalence between NH blacks and NH whites.
The purpose of this study is to determine if the differences in
current cigarette smoking prevalence observed by race-gender-age
indeed exist and if they are consistent among four US national surveys. We present self-reported current cigarette smoking data for NH
blacks and NH whites by gender and age group in the 2004–2013
National Youth Tobacco Survey (NYTS), 2002–2013 NSDUH,
2002–2013 NHIS, and 2001–2012 NHANES. In addition to selfreports, we present smoking measures using cotinine data obtained
in NHANES.
Methods
Surveys and Samples
The NYTS is sponsored both by Centers for Disease Control and
Prevention and US Food and Drug Administration. This is an anonymous, voluntary, school-based survey that uses self-administered
paper-and-pencil questionnaires. Efforts are made to make the survey
as private as possible in the conditions of a classroom, for example,
spreading the desks apart, covering responses with a piece of paper,
etc. Participants are not incentivized to take the survey. The NYTS
collects national data on key indicators for designing, executing, and
evaluating programs to prevent and control tobacco use. The survey measures tobacco-related beliefs, attitudes, behaviors, access to
tobacco, exposure to secondhand smoke and tobacco-related influences, as well as demographic data. The survey’s respondents are both
public and private middle (grades 6th–8th) and high school (9th–
12th) students from a nationally representative sample. Students in
grades 6th–8th are mainly between the ages 11 and 13 years while
students in grades 9th–12th are mainly between the ages of 14 and
18 years. Data are available for 1999, 2000, 2002, 2004, 2006, 2009,
2011, 2012, and 2013. Survey data and methodology are presented at
www.cdc.gov/tobacco/data_statistics/surveys/nyts/index.htm.
The NSDUH is sponsored by the Substance Abuse and Mental
Health Services Administration. The data has been collected from 70
000 randomly selected persons aged 12 and older every 2–3 years since
1971 and annually since 1990. At present time, it is a private confidential
audio, computer-assisted self-interview provided at home after a previous household screening interview. Since 2002, each participant received
an incentive to complete the survey. The survey focuses on tobacco, alcohol and illegal substances assessing levels and patterns of use, tracking
trends, gauging consequences, and identifying risk groups for the substances consumption and abuse. Most NSDUH survey results are provided for three different age groupings: 12–17 years old, 18–25 years
old, and 26 years or older. More information about NSDUH is available
online at https://nsduhweb.rti.org/respweb/homepage.cfm.
The NHIS is a large data collection program of the National
Center for Health Statistics, which is part of the Centers for Disease
Nicotine & Tobacco Research, 2016, Vol. 18, Suppl. 1
Control and Prevention. Since 1957, it has provided data on trends
in diseases, disability, and national health objectives. Respondents of
the NHIS are non-institutionalized civilians aged 18 years or older
residing in the United States who are sampled using a multistage
probability design. Participation is voluntary and a private computer-assisted personal interview is conducted by trained professionals in respondents’ homes. There is no incentive for participation.
NHIS data on smoking are self-reported by respondent. More survey
information is available at www.cdc.gov/nchs/nhis.htm.
The NHANES, like NHIS, is conducted by National Center for
Health Statistics at Centers for Disease Control and Prevention. The
survey, which began in 1960, is unique, in that it combines interview questions with physical examinations in order to evaluate the
health and nutritional status of children and adults in the United
States. Since 1999, NHANES has been conducted continuously, with
data released in 2-year increments. Approximately 5000 civilian
non-institutionalized individuals from birth and older participate
annually. Participation in the survey is voluntary and confidential.
Participants receive reports of study findings and a small incentive for
participation. Respondents who are 20 years and older participate
in an in-home computer-assisted personal interview conducted by a
trained professional. The interview is followed by a standard physical examination, including collecting blood and urine specimens at
a mobile examination center. At these centers, adolescent respondents (12–19 years old) also provide health information in an audio
computer-assisted self-interview in a private environment. Tobacco
use in NHANES is assessed in participants 12 years or older by computer assisted self-administered questionnaires (12–19 years old)
and face-to-face interviews (20 years and older). Biochemical determination of current cigarette smoking was performed by measuring
serum cotinine levels in blood specimens obtained by venipuncture
in the mobile examination center. The cotinine assay involved isotope dilution, liquid chromatography, and tandem mass spectrometry. We used cut-off points of higher than 10 ng/mL and 10 ng/mL or
less cotinine in serum to designate current smoking after excluding
users of other tobacco products who reported not smoking cigarettes
every day or some days as well as those who reported using a nicotine replacement therapy. More details about survey methodology
can be found at www.cdc.gov/nchs/nhanes.htm and more information for each survey can be found in the Supplementary Materials.
Analytical Samples
For this study we have used data available for each survey between
2001 and 2013: NYTS 2004–2013, NSDUH 2002–2013, NHIS
2001–2013, and NHANES 2001–2012. Of all surveys analyzed
here, the NHANES is the survey with the smallest sample size for
every 2 years’ worth of data as it only collects data every 2 years in
about 15–30 US primary sampling units. To ensure that NHANES
sample was representative of the US non-institutionalized civil population and comparable to the other samples, the authors decided to
select a large span of years for this study: 2001 to 2012. Samples for
the remaining surveys match NHANES sample as much as possible,
however, the specific years of data included vary due to each surveys’
methodology and data collection years (eg, NYTS is collected every
other year, NSDUH weighting methodology has changed in 2002).
Measures
Current Cigarette Smoking
A person was defined as a current cigarette smoker based on the
questions used by each survey, thus, the definition of a current
S43
cigarette smoker is not standardized. Each survey has screening questions identifying current smokers; survey participants with negative
responses to these screening questions are not asked further questions related to cigarette smoking. NYTS and NSDUH ask about
cigarette smoking in the past 30 days. NHANES asks similar question from its adolescent respondents. In these surveys a current cigarette smoker is defined as having smoked on at least 1 day in the past
30 days. In contrast, NHIS and NHANES ask if the adult respondents ever smoked 100 cigarettes in a lifetime and among those who
answered yes, asked them if they now smoke every day, on some
days, or not at all. Current smoking for NHIS and NHANES adult
samples was defined as those who now smoke every day or on some
days. Specific screening criteria and questions for each survey can be
found in the Supplementary Materials.
For the NHANES survey we used three definitions for the current
cigarette smoker: one based on self-reports only, a second based on
cotinine level only, and a third based on either self-report or cotinine level. An adult was defined as a current cigarette smoker if: (1)
they reported having smoked 100 cigarettes in their lives and having
smoked every day or some days; (2) if their serum cotinine level was
above 10 ng/mL32; or (3) they reported having smoked 100 cigarettes
in their lives and having smoked every day or some days or if their
serum cotinine level was above 10 ng/mL. An adolescent was defined
as a current cigarette smoker if: (1) they reported having smoked on
at least 1 day in the past 30 days; (2) if their serum cotinine level
was above 10 ng/mL; or (3) they reported having smoked on at least
1 day in the past 30 days or if their serum cotinine level was above
10 ng/mL.
Persons aged 12 years or older who reported not using cigarettes,
but using other tobacco products or using nicotine replacement
therapy were excluded from the analysis in order for them not to be
misclassified as cigarette smokers if their cotinine concentration was
above 10 ng/mL.
Statistical Analyses
Analyses were restricted to NH white and NH black persons only,
thus, estimates presented in this manuscript are not identical to previously reported current smoking estimates for the whole US noninstitutionalized population from each survey (these surveys include
other races as well in their reports). We compared current cigarette
smoking prevalence between NH whites and NH blacks by gender and age groups. We defined age groups as 12–17, 18–25, and
26 years or older.
All data analyses were performed in SUDAAN using procedures
appropriate for each survey’s sample design. Multiple years of data
were combined for each survey and weighted according to each survey’s analytical guidelines. Records with missing data were excluded
from the analyses. We have estimated weighted prevalence of current cigarette smoking, 95% CIs and statistical significance of the
estimate differences (P value).
Results
Table 1 shows sample size and percent distributions by NH white
and NH black respondents, by gender and age for the four national
surveys included in this study for aggregated years between 2001
and 2013. Sample sizes ranged from 27 987 (NHANES) to 509 517
(NSDUH).
Figure 1 shows the prevalence of current cigarette smoking for
girls and boys aged 12–17 years old by race for the aggregated years
Nicotine & Tobacco Research, 2016, Vol. 18, Suppl. 1
S44
Table 1. Demographic Distribution of Study Participants by Sex, Race, and Age by US National Surveys, 2001–2013
NYTSa
(2004–2013)
Characteristic
Race/ethnicity
Non-Hispanic black
Non-Hispanic white
Gender
Male
Female
Age
12–17
18–25
26 and older
Total
NSDUH
(2002–2013)
NHIS
(2001–2013)
NHANES
(2001–2012)
n
%
n
%
N
%
n
%
19 717
51 100
20.2
79.8
85 004
424 513
14.8
85.2
56 107
236 127
14.1
85.9
9 963
18 024
14.7
85.3
34 711
35 852
50.0
50.0
244 147
265 370
48.1
51.9
127 988
164 246
47.8
52.2
13 799
14 188
48.1
51.9
70 817
—
—
70 817
100
—
—
100
161 079
165 367
183 071
509 517
9.3
12.4
78.3
100
—
31 246
260 988
292 234
—
13.7
86.3
100
4 728
3 752
19 507
27 987
9.7
12.4
77.9
100
NHANES = National Health and Nutrition Examination Survey; NHIS = National Health Interview Survey; NSDUH = National Survey on Drug Use and Health;
NYTS = National Youth Tobacco Survey.
a
Age range for NYTS respondents was 9 to 21 years. Respondents aged less than 12 and more than 17 years were excluded from youth analyses.
Figure 1. Cigarette smoking prevalence and 95% confidence intervals among 12–17 years old by gender and race in three US national surveys.
Nicotine & Tobacco Research, 2016, Vol. 18, Suppl. 1
available for each survey as well as their 95% CIs and P values for
racial differences. For 12–17 years old girls and boys, in all surveys
based only on self-reports, NH whites have a higher smoking prevalence than NH blacks (P < .01). When cotinine levels (cotinine only
or self-reports plus cotinine) were used to identify a current smoker,
NH white girls still have a higher smoking prevalence than NH black
girls (P < .01) but no significant difference in cigarette smoking was
found between NH black and NH white boys (P > .05).
Figure 2 shows current cigarette smoking prevalence for women
and men aged 18–25 years, by race. For this age group, and similar
to the results for adolescents, in all surveys based on self-reports, NH
whites have a higher smoking prevalence than NH blacks (P < .001).
When cotinine levels (cotinine only or self-reports plus cotinine)
were used to identify a current smoker, NH white young women still
have a higher smoking prevalence than NH black young women (P <
.05) but no significant difference was found between NH black and
NH white young men (P > .10).
Figure 3 shows current cigarette smoking prevalence for NH
white and NH black adults aged 26 years or older. For adult women
the results were inconsistent across surveys. Two (NSDUH and
S45
NHANES) of three surveys based only on self-reports found no difference (P > .05) in cigarette smoking between NH black and NH
white women while a third survey (NHIS) found that NH white
women had a higher smoking prevalence than NH black women (P
< .001). When cotinine (cotinine only or self-reports plus cotinine)
was used to detect current smoking, results show that NH black
women have a higher smoking prevalence than NH white women (P
< .01). For adult men aged 26 years or older, all surveys (self-reports
or cotinine) consistently found that NH blacks have a higher smoking prevalence than NH whites (P < .0001).
Discussion
The findings from this study confirm that NH black females aged
12–25 years have a lower smoking prevalence than NH white
females of the same age. Given the consistency of results for females
aged 12–25 years old, including results using cotinine as a biomarker
of current smoking, we conclude that the lower smoking prevalence
among NH black females compared with NH white females in this
age group is real. NH black women aged 26 years or older either
Figure 2. Cigarette smoking prevalence and 95% confidence intervals among 18–25 years old by gender and race in three US national surveys.
S46
Nicotine & Tobacco Research, 2016, Vol. 18, Suppl. 1
Figure 3. Cigarette smoking prevalence and 95% confidence intervals among 26+ years old by gender and race in three US national surveys.
have a similar (self-reports) or higher (cotinine) smoking prevalence
than NH white women of the same age. What is clear is that the
lower smoking prevalence observed in NH black females at adolescence and young adulthood in 2001–2013 was not observed for
NH black women aged 26 years or older. The crossover age hypothesis for women aged 26 years or older is inconclusive when taking all surveys (biomarker or not) into consideration. When using
only data from the survey with cotinine for all age-groups, it points
to an age crossover in smoking prevalence for NH black and NH
white women.
As for males, all surveys based on self-reports found that NH
blacks aged 12–25 years have lower cigarette smoking prevalence
than NH whites, however no racial difference in smoking prevalence was observed when determining current smoking prevalence
for this age group using cotinine only or when cotinine data were
added to the self-reports. For men aged 26 years or older, all surveys confirm that NH black men have a higher smoking prevalence
than NH white men. If cotinine is detecting under-reported current
smoking among adolescents and it is more accurately assessing cigarette smoking prevalence than that of self-reports only, then the age
crossover in smoking prevalence is not confirmed for NH black and
NH white males.
The validity of self-reported cigarette smoking information has
been extensively studied over the years.18,23,33–36 In a study of factors
that affect the accuracy of self-reported smoking status among youth,
Brener and colleagues explained that these factors can be categorized
as cognitive (internal) and situational (external).23 Cognitive factors
include not understanding the meaning of the question, not being
able to remember the time period for which the question is asked,
and not being able to provide an accurate response. Situational factors include the setting (ie, school, home) where the survey is conducted, the method of survey administration (ie, self-administered
or interviewer-conducted), social desirability of the behavior in the
community (ie, cigarette smoking seen as a grown-up behavior), and
the perception of the privacy and/or confidentiality of the responses.
Some of these factors may explain some of the discrepancies in the
survey smoking prevalence results.
Self-reports or biomarkers such as cotinine are not completely
accurate to determine current smoking. Not all cigarette smokers
(self-reports) will answer accurately to having smoked in the past
30 days or smoking now some days. It would be expected that nondaily smokers, especially those who smoke infrequently (ie, only a
few days in the past 30 days) will have a harder time to accurately
recall smoking in the past 30 days or some days. Some nondaily
Nicotine & Tobacco Research, 2016, Vol. 18, Suppl. 1
smokers perceive themselves as nonsmokers and respond to the surveys accordingly.37,38 Regarding the biomarker cotinine in serum, it is
only highly sensitive to detect nicotine exposure (inhaled, absorbed,
or secondhand smoke exposure) for a few days (1 to 5 days) since
last smoked. Also, compared to cotinine, self-reported measures of
current smoking are measured for a longer time frame (either for the
past 30 days or “every day” or “some days”). Thus, self-reports nor
serum cotinine levels are gold standards of current smoking status.
There is a possibility that in the future cotinine hair analysis will
provide a standardized external validity measure given that it measures tobacco use or exposure to it over a longer period of time than
serum, saliva, or urine cotinine. However, more research would be
needed before this method can be deemed reliable and valid.39,40
Our presentation of current cigarette smoking status using
several surveys estimates has some limitations. As previously mentioned, differences exist between the surveys with regard to the
definition of current cigarette smoking, the setting (classroom,
home, examination trailer) where the interview was conducted,
the mode (paper and pencil, computer assisted, audio-assisted) of
survey administration, incentives (no incentives, monetary incentives, physical and laboratory tests) provided for participation,
the time of the year when the data were collected (spring, all-year
round), as well as variation on data collection of the use of other
tobacco products and nicotine medications. However, the specific
racial-gender-age comparison we are focusing on in our study is
mainly the same within each survey. Second, it is possible that some
users of other tobacco products and nicotine replacement therapy
who have not used cigarettes may have forgotten or denied use of
these products. If that happened, they would have been mistakenly
identified via cotinine as cigarette smokers who denied smoking.
If so, this may have resulted in higher current cigarette smoking
estimates in NHANES compared to other self-reported surveys and
may have also resulted in coming up with a different conclusion
than that of other surveys. Third, not all surveys used in this study
collected the information during comparable years (annually). By
aggregating 12 or 13 years’ worth of data we attempted to reduce
variability in the current smoking estimates. Because of inherent
limitations of all surveys, we decided to look at the consistency of
the results using different national surveys instead of stating that
some surveys were more accurate than others in estimating current
cigarette smoking.
It is well established that cigarette smoking is strongly associated
with a higher likelihood of developing and dying from a smokingrelated disease such as lung cancer, chronic obstructive pulmonary
disease, and heart attack.41 We found that NH blacks aged 26 years
or older, which compose the majority of NH black smokers, either
have a similar or higher smoking prevalence than NH white smokers.
It is known that NH black smokers are less likely to quit smoking
and more likely to relapse after quitting than NH white smokers.42
Public health implications of this study include the need to encourage and support smoking cessation strategies among NH blacks that
are fully implemented.
Conclusion
The results are inconclusive in consistently detecting the existence
of a race-gender-age crossover effect for current cigarette smoking
between NH blacks and NH whites. National birth cohort studies
are better suited to detect a race-gender-age crossover effect in smoking prevalence between these two racial groups.
S47
Supplementary Material
Supplementary Materials can be found online at http://www.ntr.
oxfordjournals.org
Funding
None declared.
Declaration of Interests
None declared.
Acknowledgments
The authors thank Dr Lucinda England for her assistance in helping with the
initial conceptualization of this manuscript and for reviewing it.
Disclaimer
The findings and conclusions are the author’s, not necessarily the CDC’s.
Supplement Sponsorship
This article appears as part of the supplement “Critical Examination of Factors
Related to the Smoking Trajectory among African American Youth and Young
Adults,” sponsored by the Centers for Disease Control and Prevention ­contract
no. 200-2014-M-58879.
References
1. Pampel FC. Racial convergence in cigarette use from adolescence to the
mid-thirties. J Health Soc Behav. 2008;49(4):484–498. www.ncbi.nlm.
nih.gov/pmc/articles/PMC3163096/pdf/nihms-317390.pdf.
Accessed
March 14, 2014.
2. US Department of Health and Human Services. Preventing Tobacco Use
Among Youth and Young Adults: A Report of the Surgeon General. 2012.
www.ncbi.nlm.nih.gov/books/NBK99237/. Accessed September 25, 2014.
3. Eaton DK, Kann L, Kinchen S, et al. Youth risk behavior surveillance—
United States, 2011. MMWR Surveill Summ. 2012;61(4):1–162. www.
cdc.gov/mmwr/pdf/ss/ss6104.pdf. Accessed March 10, 2014.
4. Center for Behavioral Health Statistics and Quality. 2013 National
Survey on Drug Use and Health: Detailed Tables. Rockville, MD:
Substance Abuse and Mental Health Services Administration; 2014.
www.samhsa.gov/data/sites/default/files/NSDUH-DetTabsPDFWHT
ML2013/Web/PDFW/NSDUH-DetTabsCover2013.pdf.
Accessed
November 26, 2014.
5. Geronimus AT, Neidert LJ, Bound J. Age patterns of smoking in US
black and white women of childbearing age. Am J Public Health.
1993;83(9):1258–1264. doi:10.2105/AJPH.83.9.1258.
6. Kandel D, Schaffran C, Hu MC, Thomas Y. Age-related differences in
cigarette smoking among whites and African-Americans: evidence for the
crossover hypothesis. Drug Alcohol Depend. 2011;118(2–3):280–287.
doi:10.1016/j.drugalcdep.2011.04.008.
7. Watt TT. The race/ethnic age crossover effect in drug use and heavy drinking. J
Ethn Subst Abuse. 2008;7(1):93–114. doi:10.1080/15332640802083303.
8. Vogt Yuan AS. Black-White differences in aging out of substance use and
abuse. Sociol Spectrum. 2010;31(1):3–31. doi:10.1080/02732173.2011.5
25694.
9. Bauman KE, Ennett SE. Tobacco use by black and white adolescents:
the validity of self-reports. Am J Public Health. 1994;84(3):394–398.
doi:10.2105/AJPH.84.3.394.
10.Fendrich M, Johnson TP. Race/ethnicity differences in the validity of
self-reported drug use: results from a household survey. J Urban Health.
2005;82(2)(suppl 3):iii67–81. doi:10.1093/jurban/jti065.
S48
11.Stein LA, Colby SM, O’Leary TA, et al. Response distortion in adolescents
who smoke: a pilot study. J Drug Educ. 2002;32(4):271–286. doi:10.2190/
GL7E-B8MV-P9NH-KCVV.
12.Wagenknecht LE, Burke GL, Perkins LL, Haley NJ, Friedman GD.
Misclassification of smoking status in the CARDIA study: a comparison of
self-report with serum cotinine levels. Am J Public Health. 1992;82(1):33–
36. doi:10.2105/AJPH.82.1.33.
13.Caraballo RS, Giovino GA, Pechacek TF. Self-reported cigarette smoking vs. serum cotinine among U.S. adolescents. Nicotine Tob Res.
2004;6(1):19–25. doi:10.1080/14622200310001656821.
14.Wills TA, Cleary SD. The validity of self-reports of smoking: analyses
by race/ethnicity in a school sample of urban adolescents. Am J Public
Health. 1997;87(1):56–61. doi:10.2105/AJPH.87.1.56.
15.Kandel I, Merrick J. Smoking and adolescence. Int. J. Adolesc. Med.
Health. 2003;15(4):365. doi:10.1515/IJAMH.2003.15.4.365.
16.Kandel DB, Schaffran C, Griesler PC, Hu MC, Davies M, Benowitz N.
Salivary cotinine concentration versus self-reported cigarette smoking: Three patterns of inconsistency in adolescence. Nicotine Tob Res.
2006;8(4):525–537. doi:10.1080/14622200600672732.
17.Yeager DS, Krosnick JA. The validity of self-reported nicotine product use
in the 2001–2008 National Health and Nutrition Examination Survey. Med
Care. 2010;48(12):1128–1132. doi:10.1097/MLR.0b013e3181ef9948.
18.Kann L, Brener ND, Warren CW, Collins JL, Giovino GA. An assessment
of the effect of data collection setting on the prevalence of health risk
behaviors among adolescents. J Adolesc Health. 2002;31(4):327–335.
doi:10.1016/S1054-139X(02)00343-9.
19.Biglan M, Gilpin EA, Rohrbach LA, Pierce JP. Is there a simple correction
factor for comparing adolescent tobacco-use estimates from school- and
home-based surveys? Nicotine Tob Res. 2004;6(3):427–437. doi:10.1080/
14622200410001696592.
20.Klein JD, Thomas RK, Sutter EJ. Self-reported smoking in online surveys: prevalence estimate validity and item format effects. Med Care.
2007;45(7):691–695. doi:10.1097/MLR.0b013e3180326145.
21.Griesler PC, Kandel DB, Schaffran C, Hu MC, Davies M. Adolescents’
inconsistency in self-reported smoking: a comparison of reports in
school and in household settings. Public Opin Q. 2008;72(2):260–290.
doi:10.1093/poq/nfn016.
22.Jackson C. Perceived legitimacy of parental authority and tobacco and
alcohol use during early adolescence. J Adolesc Health. 2002;31(5):425–
432. doi:10.1016/S1054-139X(02)00398-1.
23.Brener ND, Billy JO, Grady WR. Assessment of factors affecting the validity of self-reported health-risk behavior among adolescents: evidence
from the scientific literature. J Adolesc Health. 2003;33(6):436–457.
doi:10.1016/S1054-139X(03)00052-1.
24.Ellickson PL, Perlman M, Klein DJ. Explaining racial/ethnic differences in smoking during the transition to adulthood. Addict Behav.
2003;28(5):915–931. doi:10.1016/S0306-4603(01)00285-4.
25.Oredein T, Foulds J. Causes of the decline in cigarette smoking among
African American youths from the 1970s to the 1990s. Am J Public
Health. 2011;101(10):e4–e14. doi:10.2105/ajph.2011.300289.
26.Ford JA, Hill TD. Religiosity and adolescent substance use: evidence
from the national survey on drug use and health. Subst Use Misuse.
2012;47(7):787–798. doi:10.3109/10826084.2012.667489.
27.US Department of Health and Human Services. Changing adolescent
smoking prevalence: Where it is and why. Smoking and Tobacco Control
Monograph No. 14. Bethesda, MD: US Department of Health and Human
Services, Public Health Service, National Institutes of Health, National
Cancer Institute, NIH Publication No. 02-5086, November 2001. http://
Nicotine & Tobacco Research, 2016, Vol. 18, Suppl. 1
cancercontrol.cancer.gov/Brp/tcrb/monographs/14/m14_14.pdf. Accessed
March 23, 2015.
28.Benowitz NL, Bernert JT, Caraballo RS, Holiday DB, Wang J. Optimal
serum cotinine levels for distinguishing cigarette smokers and nonsmokers
within different racial/ethnic groups in the United States between 1999 and
2004. Am J Epidemiol. 2008;169(2):236–248. doi:10.1093/aje/kwn301.
29.Shiffman S, Tindle H, Li X, Scholl S, Dunbar M, Mitchell-Miland C.
Characteristics and smoking patterns of intermittent smokers. Exp Clin
Psychopharmacol. 2012;20(4):264–277. doi:10.1037/a0027546.
30.Caraballo RS, Novak SP, Asman K. Linking quantity and frequency profiles of cigarette smoking to the presence of nicotine dependence symptoms
among adolescent smokers: findings from the 2004 National Youth Tobacco
Survey. Nicotine Tob Res. 2009;Jan;11(1):49–57. doi:10.1093/ntr/ntn008.
31.Caraballo RS, Giovino GA, Pechacek TF, et al. Racial and ethnic differences in serum cotinine levels of cigarette smokers: Third National Health
and Nutrition Examination Survey, 1988–1991. JAMA. 1998;280(2):135–
139. doi:10.1001/jama.280.2.135.
32.Centers for Disease Control and Prevention. Biomonitoring Summary.
Cotinine. CAS No. 486-56-6. 2013. www.cdc.gov/biomonitoring/
Cotinine_BiomonitoringSummary.html. Accessed June 22, 2015.
33.Gfroerer J, Wright D, Kopstein A. Prevalence of youth substance use: the
impact of methodological differences between two national surveys. Drug
Alcohol Depend. 1997;47(1):19–30. doi:10.1016/S0376-8716(97)00063-X.
34.Hedges B, Jarvis M. Cigarette smoking. In: Prescott-Clarke P, Primatesta
P, eds. Health Survey for England. The Health Of Young People ‘95 - 97
Vol 1, Findings. London, United Kingdom: The Stationery Office; 1998:
191–221.
35.Turner CF, Lessler JT, Devore J. Effects of modes of administration and
working on reporting of drug use. In: Turner CF, Lessler JT, Gfroerer J,
eds. Survey Measurement of Drug Use: Methodological Studies. Rockville,
MD: National Institute on Drug Abuse; 1992:177–220.
36.Aquilino WS. Interview mode effects in surveys of drug and alcohol use: a
field experiment. Public Opin Q. 1994;58(2):210–240. doi:10.1086/269419.
37.Leas EC, Zablocki RW, Edland SD, Al-Delaimy WK. Smokers who report
smoking but do not consider themselves smokers: a phenomenon in need
of further attention. Tob Control. 2014;Jul;24(4):400–403. doi:10.1136/
tobaccocontrol-2013–051400.
38.Lee JK, Boyle RG, D’Silva J, St Claire AW, Whittet MN, Kinney AM.
Smoker identity among occasional smokers: findings from Minnesota.
Am J Health Behav. 2013;37(4):525–530. doi:10.5993/AJHB.37.4.10.
39.Kim S, Apelberg BJ, Avila-Tang E, et al. Utility and cutoff value of hair
nicotine as a biomarker of long-term tobacco smoke exposure, compared
to salivary cotinine. Int J Environ Res Public Health. 2014;11(8):8368–
8382. doi:10.3390/ijerph110808368.
40.Lund HM, Gjerde H, de Courtade SM, Oiestad EL, Christophersen AS.
A Norwegian study of the suitability of hair samples in epidemiological
research of alcohol, nicotine and drug use. J Anal Toxicol. 2013;37(6):362–
368. doi:10.1093/jat/bkt038.
41.
US Department of Health and Human Services. The Health
Consequences of Smoking—50 Years of Progress: A Report of the
Surgeon General. Atlanta, GA: US Department of Health and Human
Services, Centers for Disease Control and Prevention, National Center
for Chronic Disease Prevention and Health Promotion, Office on
Smoking and Health; 2014. www.ncbi.nlm.nih.gov/books/NBK179276/
pdf/TOC.pdf. Accessed May 8, 2014.
42.Caraballo RS, Kruger J, Asman K, et al. Relapse among cigarette smokers: the CARDIA longitudinal study—1985–2011. Addict Behav.
2014;39(1):101–106. doi:10.1016/j.addbeh.2013.08.030.