Drug and Alcohol Dependence 87 (2007) 1019

Predictive validity of four nicotine dependence

measures in a college sample
Eve M. Sledjeski a, , Lisa C. Dierker a , Darce Costello a ,
Saul Shiffman b , Eric Donny c , Brian R. Flay d ,
Tobacco Etiology Research Network (TERN)
a

Wesleyan University, Psychology Department, 207 High Street, Middletown, CT 06459, United States
b Department of Psychology, University of Pittsburgh, 130 N. Belleeld Avenue, Suite 510,
Pittsburgh, PA 15260, United States
c Department of Psychology, University of Pittsburgh, 4119 Sennott Square,
210 South Bouquet Street, Pittsburgh, PA 15260, United States
d Department of Public Health, Oregon State University, College of Health and Human Sciences,
254 Waldo, Corvallis, OR 97331, United States
Received 27 February 2006; received in revised form 5 July 2006; accepted 5 July 2006

Abstract
Background: The present study compared the predictive and incremental validity of four commonly used dependence measures (Diagnostic and
Statistical Manual-IV [DSM-IV] nicotine dependence criteria, Fagerstrom Test for Nicotine Dependence [FTND], Hooked On Nicotine Checklist
[HONC], Nicotine Dependence Syndrome Scale [NDSS]) in a first year college sample reporting light smoking patterns.
Methods: Nicotine dependence measures were administered at the end of the first semester and follow-up smoking behavior (i.e. continued
smoking, quantity, frequency, and length of abstinence) was assessed at the end of the first and second academic years.
Results: Higher levels of dependence as measured by the HONC and DSM-IV predicted smoking behavior at both follow-up assessments. While
higher scores on some of the NDSS factors predicted heavier smoking behavior during follow-up assessments, higher scores on other NDSS factors
predicted lighter smoking behavior. The DSM-IV, NDSS-priority, and HONC measures provided some evidence for incremental validity. Higher
dependence scores on all four measures were related to shorter lengths of smoking abstinence.
Conclusions: The four dependence measures were differentially related to smoking behavior outcomes in a light smoking sample. These findings
suggest that nicotine dependence can predict a variety of smoking behaviors in light smokers.
2006 Elsevier Ireland Ltd. All rights reserved.
Keywords: Smoking; Nicotine dependence; Predictive validity; Incremental validity; Light smokers

1. Introduction
Nicotine dependence has been shown to predict smoking
maintenance and unsuccessful quit attempts in adulthood (Colby
et al., 2000a,b). Four commonly used nicotine dependence measures are the Diagnostic and Statistical Manual-IV (DSM-IV)
nicotine dependence criteria (APA, 1994; WHO, 1994), Fagerstrom Tolerance Questionnaire (FTQ) and its modified forms
(mFTQ) and Fagerstrom Test for Nicotine Dependence (FTND:

Corresponding author. Tel.: +1 860 685 2609; fax: +1 860 685 2761.
E-mail address: esledjeski@wesleyan.edu (E.M. Sledjeski).

0376-8716/$ see front matter 2006 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.drugalcdep.2006.07.005

Fagerstrom, 1978; Heatherton et al., 1991; Prokhorov et al.,

1996), Hooked On Nicotine Checklist (HONC: DiFranza et al.,
2000, 2002a,b; OLoughlin et al., 2002a,b,c), and the Nicotine
Dependence Syndrome Scale (NDSS: Shiffman et al., 2004).
Despite some overlap, these measures tend to assess different
aspects of nicotine dependence. The FTQ and its modified forms
focus on smoking intensity and the amount of effort used to
maintain desired blood nicotine levels while the DSM-IV taps
more of the cognitive and behavioral components of dependence (Colby et al., 2000b; Fagerstrom, 1978; Strong et al.,
2003). The HONC measures nicotine dependence by assessing
an individuals loss of autonomy over smoking behavior resulting in difficulty quitting (DiFranza et al., 2000, 2002a,b), and the

E.M. Sledjeski et al. / Drug and Alcohol Dependence 87 (2007) 1019

NDSS is a new measure designed to assess multiple dimensions

of nicotine dependence: drive, tolerance, priority, stereotypy and
continuity (Shiffman et al., 2004).
Recent cross-sectional and longitudinal research has suggested that the FTQ/FTND, NDSS, DSM-IV and HONC were
related to smoking quantity and frequency (Cohen et al., 2002;
Kandel et al., 2005; Prokhorov et al., 1998; Shiffman and
Sayette, 2005; Shiffman et al., 2004), failed quit attempts
(Breslau and Johnson, 2000; Haddock et al., 1999; Shiffman
et al., 2004; Wellman et al., 2005), and biochemical markers of
nicotine (Chen et al., 2002; Prokhorov et al., 2000; Shiffman et
al., 2004) in adolescent and adult moderate to heavy smokers.
While there is some evidence of validity for each of these
dependence measures, few studies have examined more than
one measure of nicotine dependence, making direct comparisons difficult. Among studies that have employed more than
a single dependence measure, comparisons suggest that some
measures may be superior at predicting smoking behavior. For
example, three studies have found the mFTQ to be a better predictor of smoking quantity/frequency compared to the DSM-IV
in adolescents (Cohen et al., 2002; Strong et al., 2003) and adults
(Hughes et al., 2004). In addition, two studies have shown the
NDSS to continue to predict 1 year follow-up smoking behavior in adolescents and adult smokers after controlling for FTQ
scores, suggesting that the NDSS accounted for additional variance not measured by the FTQ (Clark et al., 2005; Shiffman et
al., 2004).
Aside from the dearth of comparative data, few longitudinal
studies have been conducted to examine the ability of dependence measures to predict smoking quantity and frequency over
and above initial smoking behavior (incremental validity: Clark
et al., 2005; Etter et al., 1999; Shiffman et al., 2004; Wellman et
al., 2005). Longitudinal studies controlling for initial smoking
behavior have suggested that the mFTQ/FTND fails to predict smoking behavior while the NDSS and HONC continue
to predict. For example, Etter et al. (1999) found no relationship between the FTND and continued smoking 7 months later
after controlling for initial smoking quantity, suggesting that
the FTND may represent a proxy measure for the number of
cigarettes smoked per day and may not be related to subsequent
change in smoking behavior. In a more recent study, the HONC
continued to predict smoking cessation at the 6- and 12-month
follow-ups after controlling for baseline smoking frequency and
quantity while the mFTQ did not (Wellman et al., 2006). In
addition, the NDSS has demonstrated incremental validity over
and above initial smoking behavior when predicting follow-up
smoking quantity in adolescent daily smokers (Clark et al., 2005)
and difficultly with abstaining in adult daily smokers (Shiffman
et al., 2004). More longitudinal research is needed to examine
the predictive validity of these measures as well as the independent contribution of nicotine dependence when compared to
easily obtainable indices of smoking behavior (i.e. quantity and
frequency).
Finally, although nicotine dependence measures have demonstrated some validity among samples of established smokers, few
studies have examined the utility of these measures in smokers
with irregular and/or light smoking patterns. Notable excep-

11

tions have included studies of validity of the HONC, which

have examined samples of adolescents with a range of smoking
behaviors, the majority of whom were relatively light non-daily
smokers (DiFranza et al., 2002a,b; OLoughlin et al., 2003,
2002c; Wellman et al., 2006; Wheeler et al., 2004). In these
studies, HONC scores were associated with smoking frequency
(OLoughlin et al., 2003, 2002c; Wellman et al., 2006; Wheeler
et al., 2004) and predicted continued smoking at follow-up,
daily use, and experiencing a failed quit attempt (DiFranza et
al., 2002b). The predictive validity of other nicotine dependence measures is yet to be examined within a light smoking
sample.
The present study compared the predictive and incremental validity of four commonly used dependence measures in a
sample of first year college students who have reported relatively light smoking patterns (i.e. 78% non-daily, average of <5
cigarettes per day). Our goals were to (1) determine whether
nicotine dependence measures administered at the end of the
first college semester predict follow-up smoking behavior at the
end of the first and second college year (predictive validity)
and (2) assess whether dependence measures continue to predict smoking behavior after controlling for baseline smoking
behavior (incremental validity).
2. Methods
2.1. Participants
The selection and retention of participants for the present study have been
previously reported in detail (Clayton, 2004; Dierker et al., 2006a; Tiffany et
al., unpublished). Briefly, eligible participants were selected from a pool of first
year college students who completed a screening survey during an orientation
program in the summer of 2002 at Purdue University (n = 4690, 71% response
rate). From these, 2001 first year students who had atleast some prior experience with smoking (i.e. one or more puffs lifetime) were invited to participate
in the study. In total, 912 students (45%) agreed to participate in the baseline
survey and weekly web-based surveys throughout their first year and one survey during their second year. Participants who reported smoking during the
past week completed the nicotine dependence measures at the end of the first
semester (n = 112). The present analyses included 95 participants who completed the web-based surveys at the end of their first semester and the follow-up
survey at the end of their first year (retention rate = 85%). In addition, 55 participants completed the follow-up survey at the end of the second year (retention
rate = 58%). All participants lived in on-campus housing their freshman year
and 42% continued to live on-campus during their sophomore year. The present
sample was 48% female and 94% Caucasian. Attrition rates were similar for
males and females and did not differ based on FTND, HONC, NDSS-T, NDSSdrive, NDSS-priority, NDSS-continuity and DSM-IV continuous and diagnosis
scores. However, participants who completed the end of the first year assessment reported lower NDSS-tolerance scores and participants who completed the
end of the second year assessment reported lower NDSS-stereotypy scores and
lower smoking frequency compared to participants who did not complete the
follow-up assessments (p < 0.05).

2.2. Measures
2.2.1. Smoking behavior. Smoking behavior was assessed using a 7-day timeline follow-back report. This procedure had participants think back over the past
7 days and report the number of cigarettes they smoked each day. Continued
smoking at the end of the first and second years was assessed by these 7-day
retrospective reports of cigarette smoking using a web-based protocol. Aggregate past week smoking variables addressing both the quantity (total number
of cigarettes smoked this past week) and frequency (number of days smoked

12

E.M. Sledjeski et al. / Drug and Alcohol Dependence 87 (2007) 1019

during the past week) were created from these responses. In addition, an objective measure of length of abstinence was created to assess the longest period
of consecutive days that participants reported not smoking. This variable was
created by using the continuous timeline follow-back reports of smoking from
the end of the first semester to the end of the first year.
2.2.2. Nicotine dependence.
2.2.2.1. Fagerstrom Test for Nicotine Dependence (FTND). The FTND is a
modified version of the Fagerstrom Tolerance Questionnaire (FTQ: Fagerstrom,
1978) consisting of six items designed to assess nicotine dependence (Heatherton
et al., 1991). Item scoring was based on procedures developed by Heatherton et
al. (1991) and items were summed to yield a total score (possible range = 010).
The smoking quantity item (i.e. average number of cigarettes per day) on
the FTND was excluded in analyses concerning quantity and frequency to
avoid confounding (possible range = 07: Cohen et al., 2002; Lichtenstein and
Mermelstein, 1986; Prokhorov et al., 1998). One item assessing whether the
participant smoked more frequently during the first hours after waking was not
included in reliability analyses since all participants responded no. The FTND
demonstrated low reliability in the present sample (Cronbachs = 0.585). Prior
research has also suggested poor internal consistency using the FTQ, FTND and
other modified versions in adolescent and adult samples (Burling and Burling,
2003; Cohen et al., 2002; Heatherton et al., 1991; Lichtenstein and Mermelstein,
1986; Payne et al., 1994; Pomerleau et al., 1994).
2.2.2.2. Hooked On Nicotine Checklist (HONC). The HONC is a 10-item measure rated on a dichotomous scale (i.e. yes or no) designed to test an individuals
loss of autonomy over tobacco use (DiFranza et al., 2002a,b). Two methods
of scoring have been proposed for the HONC including a continuous measure
consisting of the sum of endorsed responses (degree of lost autonomy) and a
dichotomous measure categorizing participants into loss (endorsement of one
or more items) versus no loss of autonomy groups (DiFranza et al., 2002a). In
the present sample only 11 participants reported no loss of autonomy; therefore,
the continuous HONC scoring method was used (possible range = 010). The
measure demonstrated acceptable reliability in the present sample (Cronbachs
= 0.877).
2.2.2.3. Diagnostic and Statistical Manual of mental disorders-IV (DSM-IV).
A self-administered version of the Composite International Diagnostic Interview
Tobacco Module (CIDI: WHO, 1994) was used to assess the seven criteria
of nicotine dependence as specified by the DSM-IV (APA, 1994): tolerance
(two items), withdrawal (nine items), smoking in larger amounts or longer than
intended (two items), persistent desire or unsuccessful efforts to cut down (one
item), great deal of time spent to obtain, use or recover from smoking (one item),
activities given up or reduced (one item), and continued use despite physical or
psychological problems caused or exacerbated by smoking (two items). The
complete questionnaire can be found in Dierker et al. (2006).
Given our interest in the association between a broad range of smoking
quantity and frequency and the endorsement of nicotine dependence criteria,
daily use of nicotine was not required for the assessment of symptoms. Further, unlike instruments that assess withdrawal symptoms only among smokers
who have tried to quit or cut down, we assessed withdrawal among all smokers based on any periods in which smoking behavior has been limited for any
reason.
Based on the difficulty in self-reporting decreased heart rate, this withdrawal symptom was not assessed. Though craving is not listed as a symptom
of nicotine dependence or withdrawal in DSM-IV (APA, 1994), craving is the
most frequently reported withdrawal symptom among young smokers thus it
was assessed (Colby et al., 2000a; DiFranza et al., 2000). Response categories
included not at all, a little bit, somewhat, and quite a bit. Symptoms
were coded as present if reported at any level (i.e. a little bit through quite a bit).
An individual was classified as dependent if he/she experienced atleast three of
the seven dependence criteria associated with their smoking behavior.
Although the DSM was developed to provide a dichotomous measure of
nicotine dependence, modern theories maintain, explicitly or implicitly, that
dependence varies on a continuum (Tiffany et al., 2004). That continuum is
linked in turn to a trajectory of smoking behavior, with the basic elements of

dependence processes evident even in the early episodes of cigarette use. Given
that we were interested in nicotine dependence among light irregular smokers, we computed a continuous score in addition to the traditional dependence
diagnosis. Similar to prior research, the number of DSM-IV criteria met were
summed to yield a continuous dependence score (possible range = 07) (Cohen
et al., 2002). The continuous measure demonstrated acceptable reliability in the
present sample (Cronbachs = 0.752).
2.2.2.4. Nicotine Dependence Syndrome Scale (NDSS). The NDSS is a 19-item
multidimensional measure consisting of five factors (drive, priority, continuity, stereotypy, and tolerance) that assess nicotine dependence (Shiffman et
al., 2004). Drive measures craving and withdrawal symptoms while tolerance
assesses reduced sensitivity to tobacco products. Priority assesses the preference for smoking over other reinforcers. Continuity assesses the regularity of
smoking while stereotypy measures the sameness of smoking contexts. Items
were rated on a scale from one (not at all true) to five (extremely true). The total
score (NDSS-T) and five factor scores were computed using the regression-based
algorithms described in Shiffman et al. (2004). These algorithms were designed
to reduce the intercorrelations among the five factors as well as standardizing
the scores (mean = 0, S.D. = 1 on the normative sample). The total scores and
most factors demonstrated acceptable reliability (Cronbachs = 0.8230.945);
however, reliability for the stereotypy subscale was lower (Cronbachs
= 0.686).

2.3. Statistical analyses

Chi-square analysis (dichotomous measures) and analysis of variance
(ANOVA: continuous measures) were used to examine group differences
between participants that completed the follow-up assessments and those that
did not as well as gender differences in smoking behavior and nicotine dependence. Logistic regression analyses were used to assess the relationship between
dependence measures and the dichotomous measures of continued smoking
at the follow-up assessments. Linear regression analyses for smoking quantity, frequency, and length of abstinence were used to assess the relationships
between the dependence measures and smoking behavior. Hierarchical linear regressions (continuous) were used to examine the incremental validity of
the dependence measures in predicting future smoking quantity and frequency
after controlling for baseline smoking quantity and frequency (Hunsley and
Meyer, 2003). Hierarchical linear regression analyses were used to examine the
interaction between nicotine dependence levels and baseline smoking levels in
predicting follow-up smoking quantity and frequency. The Bonferroni method
was used to adjust p values for measures employing multiple scoring procedures (i.e. NDSS and DSM-IV). Significance levels were set at 0.008 for the
NDSS and its factor scores and 0.025 for DSM-IV continuous and diagnosis
scores.

3. Results
3.1. Prevalence of nicotine dependence and smoking
behavior
Table 1 displays descriptive statistics for the four dependence measures and smoking behaviors at each of the three
waves of assessment (baseline and end of year 1 and 2). Participants reported relatively light and irregular smoking patterns.
On average at baseline, participants smoked 30.62 (S.D. = 34.88)
cigarettes during the past week and the majority were non-daily
smokers (M = 4.58, S.D. = 2.21 days; 22% daily smokers). Males
and females showed similar rates of nicotine dependence and
smoking quantity and frequency (ps > 0.05). Dependence levels as measured by the FTND and NDSS were low. However,
64% of participants met diagnostic criteria based on the DSM-IV
and average HONC scores indicated moderate levels of nicotine
dependence.

E.M. Sledjeski et al. / Drug and Alcohol Dependence 87 (2007) 1019

Table 1
Descriptive statistics for nicotine dependence measures and smoking behavior,
mean (S.D.) and frequency (%)
Baseline assessment

Total (n = 95)

Female (%)
Caucasian (%)

48
94

Dependence measures
FTND
DSM-IV
Continuous
Diagnosis
NDSS-total
Drive
Priority
Tolerance
Continuity
Stereotypy
HONC

3.57 (2.03)
61 (64%)
1.04 (0.886)
1.67 (1.02)
0.401 (0.502)
0.812 (1.11)
1.12 (1.25)
0.711 (0.752)
4.78 (3.13)

Smoking behavior
Quantity
Frequency

30.62 (34.88)
4.58 (2.21)

13

62% increase for the second year follow-up. Finally, for each
unit increase in the NDSS-stereotypy score there was an associated 60% decrease in the likelihood of continued smoking for
the first year follow-up.
3.3. Dependence measures and smoking quantity and
frequency

0.71 (1.26)

Second semester

Total (n = 95)

Continued smoking
Quantity
Frequency
Length of abstinence (days)

58 (61%)
25.36 (31.85)
3.39 (3.06)
4.63 (4.21)

Second year

Total (n = 55)

Female (%)
Caucasian (%)
Continued smoking
Quantity
Frequency

54
98
39 (71%)
29.25 (39.58)
3.75 (3.00)

Note: quantity = number of cigarettes smoked in the past week; frequency =

number of days smoked in the past week.

3.2. Dependence measures and continued smoking

Continued smoking was reported by 58 (61%) participants
during the second semester follow-up and 39 (71%) participants
during the second year follow-up. Logistic regression analyses
were used to evaluate the relationship between the dependence
measures (assessed at the end of the first college semester) and
continued smoking (yes/no) at the end of the first and second year
of college. A separate regression was conducted for each dependence measure with the dependence measure/subscale serving
as the predictor and continued smoking serving as the outcome
variable in the model. The HONC, NDSS-stereotypy, and DSMIV continuous score and diagnosis predicted continued smoking
at the end of the first year and the DSM-IV measures continued
to predict smoking at the end of sophomore year (see Table 2).
More specifically, for each unit increase in the HONC score
there was an associated 21% increase in the likelihood of continued smoking for the first year follow-up and 20% increase for
the second year follow-up. In addition, for each unit increase in
the DSM-IV score there was an associated 40% increase in the
likelihood of continued smoking for the first year follow-up and

Separate linear regression analyses were conducted to assess

the relationship between the dependence measures and smoking quantity and frequency at the follow-up assessments (see
Tables 3 and 4). Higher scores on the HONC, NDSS-T, NDSSdrive, NDSS-tolerance, and DSM-IV measures predicted a
greater number of cigarettes smoked at the end of the first year
while lower scores on the NDSS-priority predicted higher quantity at the end of the first year. In addition, the HONC, NDSSdrive, and DSM-IV measures predicted smoking frequency at the
end of the first year. The HONC, NDSS-T, NDSS-tolerance and
DSM-IV measures predicted second year quantity and frequency
while the FTND and NDSS-drive predicted quantity only.
Separate hierarchical linear regressions were used to assess
the incremental validity of the dependence measures (see
Tables 3 and 4). Only NDSS-priority scores predicted quantity at the end of the first year after controlling for baseline
smoking behavior. In addition, the HONC and DSM-IV measures continued to predict frequency at the end of the first year
after controlling for baseline smoking behavior. Only DSMIV continuous scores and diagnosis predicted frequency at the
end of the second year after controlling for baseline smoking
behavior.
3.4. Interaction between baseline smoking behavior and
nicotine dependence
Separate hierarchical linear regressions were conducted to
test for an interaction between baseline quantity and frequency
and dependence levels on all four measures. These analyses were
conducted to examine whether the relationship between nicotine
dependence and follow-up smoking quantity and frequency varied depending on initial smoking levels. The only significant
interaction was between DSM-IV continuous scores and baseline smoking quantity predicting smoking quantity at the end of
the first year. DSM-IV continuous scores and baseline smoking
quantity were entered in the first step and the interaction between
DSM-IV scores and quantity was entered in the second step of
the model. Smoking quantity at the first year follow-up served as
the dependent variable. The interaction between DSM-IV dependence scores and baseline smoking quantity significantly added
to the model, F change (1, 91) = 8.66, p = 0.004. Decomposition of the interaction revealed that at low levels of baseline
smoking (1 S.D. below the mean, 5 cigarettes per week),
higher DSM-IV dependence levels (5 dependence criteria) predicted higher smoking quantity at first follow-up, t(91) = 2.74,
p = 0.007 (see Fig. 1). In contrast, the DSM-IV did not predict
smoking quantity when baseline smoking quantity was high (1
S.D. above the mean, 60 cigarettes per week), t(91) = 1.63,
p = 0.107.

14

E.M. Sledjeski et al. / Drug and Alcohol Dependence 87 (2007) 1019

Table 2
Logistic regressions: dependence measures predicting continued smoking at the end of the first and second college year
Predictor

First year (n = 95)

Second year (n = 55)

OR (CI)

FTND
HONC

0.189
0.192

0.253
0.009

0.827 (0.5981.15)
1.21 (1.051.40)*

NDSS-T
Drive
Priority
Tolerance
Continuity
Stereotypy

0.268
0.571
1.26
0.276
0.099
0.920

0.279
0.016
0.017
0.169
0.559
0.004

1.31 (0.8052.12)
1.77 (1.132.82)+
0.284 (0.1010.798)+
1.32 (0.8891.95)
1.10 (0.7921.54)
0.398 (0.2140.713)*

DSM-IV
Continuous
Diagnosis

0.336
1.54

0.003
0.001

1.40 (1.121.75)*
4.54 (1.8511.14)*

OR (CI)

0.223
0.182

0.516
0.078

1.25 (0.6382.45)
1.20 (0.9801.47)+

0.835
0.619
0.009
0.443
0.113
0.382

0.052
0.067
0.992
0.135
0.603
0.413

2.31 (0.9915.36)+
1.86 (0.9583.60)+
0.991 (0.1606.14)
1.56 (0.8712.79)
0.893 (0.5831.37)
1.45 (0.5873.66)

0.480
1.99

0.006
0.003

1.62 (1.142.28)*
7.33 (2.0126.73)*

Note: Separate logistic regressions were conducted for each dependence measure/subscale. OR, odds ratio; CI, 95% confidence interval. + Marginally significant,
p < 0.1. * Significant at 0.05 or Bonferroni corrected level.

Table 3
Linear regressions: dependence measures predicting end of first year smoking quantity and frequency (n = 95)
Hierarchical analysesa

Univariate analyses

R2

R2

1.38
3.77

0.172
0.000

0.010
0.123*

0.034
0.205

0.333
1.97

0.740
0.052

0.001
0.033+

20.88
3.03
3.29
3.33
0.615
0.375

0.005
0.003
0.001
0.001
0.540
0.708

0.072*
0.080*
0.095*
0.097*
0.007
0.009

0.043
0.120
0.309
0.153
0.036
0.174

0.364
1.13
3.50
1.44
0.382
1.81

0.717
0.262
0.001
0.154
0.703
0.074

0.001
0.011
0.095*
0.018
0.001
0.028+

1.21
3.34

0.002
0.001

0.090*
0.097*

0.137
0.197

1.28
2.00

0.203
0.049

0.014
0.034+

0.062
0.331

0.597
3.38

0.552
0.001

0.007
0.109*

0.151
0.273

1.41
2.50

0.164
0.014

0.019
0.059*

NDSS-T
Drive
Priority
Tolerance
Continuity
Stereotypy

0.173
0.282
0.250
0.189
0.117
0.209

1.70
2.84
2.49
1.85
1.14
2.06

0.093
0.006
0.014
0.067
0.259
0.042

0.030+
0.070*
0.053+
0.025+
0.003
0.033+

0.044
0.190
0.227
0.077
0.115
0.239

0.348
1.73
2.34
0.693
1.16
2.47

0.729
0.087
0.021
0.490
0.248
0.016

0.001
0.029+
0.051+
0.005
0.013
0.057+

DSM-IV
Continuous
Diagnosis

0.390
0.406

4.09
4.29

0.000
0.000

0.143*
0.156*

0.351
0.350

3.11
3.31

0.003
0.001

0.088*
0.099*

Quantity
FTND
HONC

0.142
0.364

NDSS-T
Drive
Priority
Tolerance
Continuity
Stereotypy

0.286
0.300
0.323
0.327
0.064
0.039

DSM-IV
Continuous
Diagnosis

0.351
0.327

Frequency
FTND
HONC

Note: R2 = adjusted R2 . + Marginally significant, p < 0.1. * Significant at 0.05 or Bonferroni corrected level.
a Controlling for baseline quantity and frequency.

3.5. Dependence measures and smoking abstinence

Results of linear regression analyses evaluating the relationship between dependence measures and the longest period of
abstinence are presented in Table 5. Univariate linear regressions were conducted separately for each dependence measure/subscale. A longer length of abstinence was related to

lower FTND, HONC, NDSS-drive, NDSS-tolerance, and DSMIV diagnosis and continuous scores.
4. Discussion
The present study examined the predictive and incremental
validity of four nicotine dependence measures in a sample of first

E.M. Sledjeski et al. / Drug and Alcohol Dependence 87 (2007) 1019

15

Table 4
Linear regressions: dependence measures predicting second year smoking quantity and frequency (n = 55)
Hierarchical analysesa

Univariate analyses

Quantity
FTND
HONC

0.274
0.405

R2

2.07
3.23

0.043
0.002

0.058*
0.148*

0.171
0.139

4.25
3.68
1.93
4.32
0.140
0.375

0.000
0.001
0.059
0.000
0.889
0.709

0.240*
0.189*
0.048+
0.246*
0.000
0.000

0.204
0.134
0.049
0.281
0.054
0.065

R2

1.62
1.12

0.112
0.269

0.028
0.015

1.48
0.983
0.420
2.35
0.487
0.582

0.144
0.330
0.677
0.023
0.628
0.563

0.025
0.011
0.002
0.060
0.003
0.004

2.23
2.28

0.030
0.027

0.055+
0.057+

NDSS-T
Drive
Priority
Tolerance
Continuity
Stereotypy

0.504
0.452
0.256
0.510
0.019
0.051

DSM-IV
Continuous
Diagnosis

0.551
0.488

4.80
4.07

0.000
0.000

0.290*
0.223*

0.291
0.268

0.075
0.296

0.551
2.26

0.584
0.028

0.006
0.070*

0.024
0.105

0.193
0.729

0.848
0.469

0.001
0.008

NDSS-T
Drive
Priority
Tolerance
Continuity
Stereotypy

0.381
0.332
0.058
0.355
0.005
0.131

3.00
2.57
0.423
2.77
0.039
0.964

0.004
0.013
0.674
0.008
0.969
0.339

0.129*
0.094+
0.000
0.110*
0.000
0.000

0.178
0.106
0.093
0.188
0.034
0.079

1.11
0.670
0.695
1.32
0.266
0.613

0.271
0.506
0.490
0.192
0.792
0.542

0.019
0.007
0.008
0.027
0.001
0.006

DSM-IV
Continuous
Diagnosis

0.500
0.558

4.20
4.90

0.000
0.000

0.235*
0.298*

0.371
0.455

2.48
3.47

0.017
0.001

0.086*
0.154*

Frequency
FTND
HONC

Note: R2 = adjusted R2 . + Marginally significant, p < 0.1. * Significant at 0.05 or Bonferroni corrected level.
a Controlling for baseline smoking quantity and frequency.

year college students. The sample was comprised of relatively

light, mostly non-daily (78%) smokers, smoking an average of
30 cigarettes during the past week. Higher levels of dependence
as measured by the HONC and DSM-IV continuous scores and
diagnosis predicted continued smoking, smoking quantity and

smoking frequency at the end of the first academic year. Both

the HONC and DSM-IV measures continued to predict smoking quantity and frequency during the second year follow-up.
While higher scores on the NDSS-T, NDSS-drive, and NDSStolerance predicted heavier smoking behavior during follow-up
assessments, higher scores on the NDSS-priority and NDSSstereotypy subscales predicted lighter smoking behavior. Higher
dependence scores on all four measures were related to shorter

Table 5
Linear regressions: dependence measures predicting length of abstinence
(n = 95)

Fig. 1. Interaction between continuous DSM-IV scores and baseline smoking

quantity predicting end of first year smoking quantity. Quantity represents the
number of cigarettes smoked during the past week. DSM-IV dependence predicted follow-up smoking quantity only at lighter baseline smoking levels.

Length of abstinence

R2

FTND
HONC

0.213
0.371

2.10
3.02

0.039
0.004

0.046*
0.122*

NDSS-T
Drive
Priority
Tolerance
Continuity
Stereotypy

0.507
0.379
0.075
0.447
0.041
0.124

5.67
3.95
0.726
4.82
0.400
1.20

0.000
0.000
0.469
0.000
0.690
0.233

0.249*
0.135*
0.000
0.191*
0.000
0.005

DSM-IV
Continuous
Diagnosis

0.473
0.457

4.05
3.88

0.000
0.000

0.210*
0.195*

Note: R2 = adjusted R2 . + Marginally significant, p < 0.1. * Significant at 0.05 or

Bonferroni corrected level.

16

E.M. Sledjeski et al. / Drug and Alcohol Dependence 87 (2007) 1019

lengths of smoking abstinence. In addition, DSM-IV, HONC,

and NDSS-priority measures provided some evidence for incremental validity. Finally, DSM-IV continuous scores predicted
future smoking quantity when baseline smoking quantity was
low, but not when baseline smoking quantity was high.
4.1. Predictive validity of nicotine dependence
The HONC was designed to measure the onset of nicotine
dependence by assessing lost autonomy over smoking behavior in neophyte smokers (DiFranza et al., 2002a). Our findings
support previous research demonstrating the predictive validity
of the HONC in adolescents reporting light smoking patterns
(DiFranza et al., 2002b; Wellman et al., 2006). Mean HONC
(M = 4.78, S.D. = 3.13) scores among our predominantly light
smoking sample were similar to prior research assessing dependence in adolescent smokers across a range of smoking patterns
(M = 4.304.77; OLoughlin et al., 2003). Despite our small
sample size, the odds ratios for HONC scores predicting continued smoking for the first and second year follow-ups were almost
identical to those reported by Wellman et al. (2006). Based on
prior research and the present findings, the HONC appears to
be a valuable measure of nicotine dependence and subsequent
smoking behavior among less experienced smokers.
Despite our light smoking sample, DSM-IV continuous
scores and diagnosis demonstrated predictive validity during
follow-up assessments. Our findings support prior research
demonstrating a relationship between DSM-IV measures of
nicotine dependence and concurrent daily smoking quantity in
adolescents (Kandel et al., 2005). In addition, mean DSM-IV
dependence scores (M = 3.57, S.D. = 2.03) were similar to prior
research assessing dependence in adolescent and adult smokers (M = 3.44.5; Hughes et al., 2004; Strong et al., 2003). To
our knowledge, this is the first study to support the predictive
validity of DSM-IV measures of nicotine dependence in a sample of relatively light smokers and highlights the utility of using
DSM-IV criteria in college-age samples.
The NDSS-stereotypy was negatively associated with continued smoking during the first year while the NDSS-T, NDSSdrive, NDSS-priority, and NDSS-tolerance factors predicted
quantity and NDSS-drive predicted smoking frequency at the
end of the first year. Only the NDSS-T, NDSS-drive, and NDSStolerance continued to predict smoking quantity and frequency
during the second year. While the NDSS-T, NDSS-drive, and
NDSS-tolerance scores demonstrated the expected positive relationship with smoking behavior, higher levels of dependence
on the NDSS-stereotypy and NDSS-priority factors predicted
lower levels of smoking behavior. Stereotypy is a measure of
the invariance of smoking which reflects the idea that nicotine
dependence would be characterized by smoking patterns that
are not particularly affected by the time of day, day of the week,
or different situations/emotions. Priority represents a measure of
ones preference for smoking over other reinforcers (Shiffman
et al., 2004). These findings may reflect the unstable nature of
these factors in light smoking samples which is also evidenced
by the poor internal consistency of the stereotypy subscale in
the present sample. However, one recent study found positive

relationships between the stereotypy and priority subscales and

several measures of smoking behavior within a group of adult
light smokers (chippers, Shiffman and Sayette, 2005). It is
also possible that social networks played a role in decreasing
smoking behavior at follow-up. For example, individuals who
scored high on priority and stereotypy but lacked a supportive
smoking environment may have been more likely to reduce their
smoking behavior at follow-up. Numerous studies have shown
peers smoking status to impact adolescents smoking behavior (Mayhew et al., 2000). Despite the unexpected relationship
between stereotypy, priority and smoking behavior, these findings support the use of separate factor scores when using the
NDSS to measure nicotine dependence.
With the exception of second year smoking quantity, the
FTND failed to predict smoking behavior in both univariate and
hierarchical analyses. In the present sample, the internal consistency of the FTND was quite low ( = 0.585); however, this low
reliability has been reported in a number of studies (Cohen et
al., 2002; Heatherton et al., 1991; Lichtenstein and Mermelstein,
1986; Payne et al., 1994; Pomerleau et al., 1994). Researchers
have suggested that the low internal consistency could be due
to the small number of items (6), limited response range of the
items, or the presence of two separate constructs (Colby et al.,
2000b; Etter, 2005; Haddock et al., 1999; Payne et al., 1994).
In addition, FTND scores (M = 0.71, S.D. = 1.26) in the present
sample were much lower compared to rates reported in previous research (M = 1.845.3; (Burling and Burling, 2003; Etter et
al., 1999; Haddock et al., 1999; Hughes et al., 2004). Sixty-four
(67%) participants did not endorse any FTND item resulting in
a greatly skewed distribution of scores, while scores on the other
measures of dependence were normally distributed.
Although cross-sectional studies have demonstrated a relationship between FTQ/FTND/mFTQ scores and smoking
behavior (Cohen et al., 2002; Fagerstrom and Schneider, 1989;
Prokhorov et al., 2000, 1998, 1996; Strong et al., 2003),
our results mirror the findings reported by other longitudinal
research demonstrating the inability of the FTND and modified
versions to predict smoking behavior even prior to adjusting for
baseline smoking behavior (Etter et al., 1999; Wellman et al.,
2006). These authors have suggested that the FTND may act as
a proxy for smoking quantity. While the FTND was developed
for and typically used in samples of heavy smokers, our findings
confirm that it may not be an appropriate measure of nicotine
dependence among light non-daily smokers.
Lower dependence scores on all four measures predicted
longer periods of abstinence, suggesting that more dependent
individuals have a harder time abstaining and tend to report more
regular smoking habits. These findings support prior research
demonstrating a relationship between the number of failed quit
attempts and nicotine dependence (Breslau and Johnson, 2000;
DiFranza et al., 2000; Haddock et al., 1999; Shiffman et al.,
2004; Wellman et al., 2006). The use of smoking abstinence, as
opposed to quit attempts, may be more appropriate for light
smoking samples given the inherent variability in smoking
patterns among non-daily smokers. In addition, retrospective
reports of quit attempts may be biased due to problems with
recall and a lack of an operational definition of a quit attempt (e.g.

E.M. Sledjeski et al. / Drug and Alcohol Dependence 87 (2007) 1019

defined period of time without smoking). By using 7 day timeline follow-back reports, our study allowed an objective measure
of smoking abstinence by examining the longest period of time
participants reported not smoking between the end of the first
semester and end of the first college year.
4.2. Incremental validity of nicotine dependence
To ensure that the measures of dependence were not merely
assessing cigarette use, we examined the incremental validity
of each (Colby et al., 2000b; Hunsley and Meyer, 2003; Tiffany
et al., unpublished). When controlling for baseline quantity and
frequency, lower NDSS-priority scores predicted higher smoking quantity while higher dependence scores on the HONC and
DSM-IV predicted higher smoking frequency at the end of the
first year. The bivariate relationships between end of first year
quantity and the HONC, DSM-IV, NDSS-drive, and NDSStolerance scores were no longer significant after adjusting for
baseline smoking behavior. By the end of the second year, only
the DSM-IV added incremental validity when predicting smoking frequency.
Notably, a hierarchical linear regression predicting end of
first year smoking quantity demonstrated an interaction between
DSM-IV scores and baseline smoking quantity. Decomposition
of the interaction revealed that dependence scores predicted
smoking quantity at follow-up but only in low level smokers.
It has been historically assumed that nicotine dependence is
present only in heavy daily smokers (APA, 1994). However,
more recent research has contradicted this assumption, finding
the presence of dependence symptoms in new adolescent smokers (DiFranza et al., 2000, 2002a,b). Our findings build on this
research by establishing the importance of dependence symptoms among light smokers in predicting future smoking over
and above number of cigarettes smoked. Research into the emergence of nicotine dependence is necessary in order to identify
individual differences that predict sensitivities to nicotine and
consequent chronic use (DiFranza et al., 2000; Shiffman, 1991).
4.3. Strengths and limitations
Since most dependence research has been limited to daily
smokers, the major strength of our study was the assessment
of nicotine dependence and its ability to predict later smoking
behavior across a continuum of use including very light and
non-daily smoking. Typically dependence criteria are assessed
in individuals meeting an arbitrary threshold of use (e.g. daily
use). In order to better understand the emergence of nicotine
dependence and predict future smoking behavior, it is necessary to assess dependence in all current smokers (Colby et al.,
2000a,b; Strong et al., 2003).
In addition, current nicotine dependence measures have
largely been validated in adult heavy smokers, which raises questions regarding their utility in light smoking samples (Tiffany
et al., 2004). Given that research has begun to focus on the
emergence of nicotine dependence, it is particularly important to
assess dependence at low levels of use to determine the validity
of current measures. The present study administered multiple

17

measures of nicotine dependence and assessed smoking behavior at several time points allowing for the examination of both
predictive and incremental validity of the measures in a light
smoking sample. Future research is needed to better understand
the emergence of nicotine dependence as well as the individual
differences present in smoking and dependence trajectories.
Finally, smoking behavior was assessed weekly based on retrospective recall of past week behavior. While the accuracy of
these reports have been questioned, research has shown selfreported smoking behavior to be a valuable index of smoking
heaviness, being positively correlated with biochemical measures of tobacco use (Heatherton et al., 1989). Further, accuracy
of smoking behavior was maximized by limiting recall to the
past week, rather than several weeks or even lifetime use.
The present results should be interpreted within the context of
study limitations. First, our relatively homogenous sample (i.e.
first year college students with the majority being 18 years of age
and predominantly Caucasian) limited our ability to generalize
to more diverse populations. However, increased smoking rates
among college students highlight the importance of assessing
smoking behavior and nicotine dependence in this vulnerable
population (Kear, 2002). Second, nicotine dependence was only
assessed among students who reported smoking within the past 7
days. Given the non-daily smoking behavior in this sample, light
smokers who did not smoke during the previous week may have
been excluded from completing these measures. Third, including
subscales and scoring methods, we examined the validity of
ten dependence scales increasing our chance of Type I error. A
more conservative interpretation of the results would be to use
a p value of 0.005. Fourth, our inability to find relationships
between dependence measures and second year smoking could
be due to our small sample size during the follow-up (n = 55).
Despite these limitations, to our knowledge this study is the
first to administer four dependence questionnaires to mostly nondaily, light smokers. While not all measures performed well,
there was evidence of nicotine dependence as well as predictive
and incremental validity in a sample of light smokers. Although
smoking initiation typically begins in adolescence (SAMHSA,
2002), little research has examined the importance of emerging
dependence in light smokers and its consequent role in smoking
maintenance into adulthood (Colby et al., 2000a,b; DiFranza et
al., 2000; Tiffany et al., 2004). Our findings suggest that nicotine
dependence measures, in particular the DSM-IV dependence
criteria, could be used as a tool to examine the emergence of
dependence as well as predict smoking behavior prior to the
development of more established smoking patterns.
Given that current measures of nicotine dependence were
developed for adult heavy smokers, Colby et al. (2000b) have
suggested that they may not be tapping the appropriate constructs that would predict smoking trajectories in light smokers.
Thus, more qualitative research (e.g. focus groups, individual interviews) may be warranted to determine factors related
to smoking maintenance among light smokers (Nichter et al.,
2002, 1997). Additional prospective studies employing multiple
measures of nicotine dependence are needed to identify factors that consistently predict smoking persistence among light
smokers.

18

E.M. Sledjeski et al. / Drug and Alcohol Dependence 87 (2007) 1019

Acknowledgments
This research was sponsored by the Robert Wood Johnson
Foundation, Tobacco Etiology Research Network (TERN). Data
analyses were supported by grant K01 DA 15454-01 from the
National Institute of Drug Abuse (Dierker) and an Investigator Award from the Patrick & Catherine Weldon Donaghue
Medical Research Foundation (Dierker). The Tobacco Etiology
Research Network (TERN) includes Richard Clayton, David
Abrams, Robert Balster, Linda Collins, Ronald Dahl, Brian
Flay, Gary Giovino, Jack Henningfield, George Koob, Robert
McMahon, Kathleen Merikangas, Mark Nichter, Saul Shiffman,
Stephen Tiffany, Dennis Prager, Melissa Segress, Christopher
Agnew, Craig Colder, Lisa Dierker, Eric Donny, Lorah Dorn,
Thomas Eissenberg, Brian Flaherty, Lan Liang, Nancy Maylath, Mimi Nichter, Elizabeth Richardson, William Shadel, and
Laura Stroud.
References
APA, 1994. Diagnostic and Statistical Manual of Mental Disorders. American
Psychiatric Association, Washington, DC.
Breslau, N., Johnson, E.O., 2000. Predicting smoking cessation and major
depression in nicotine-dependent smokers. Am. J. Public Health 90,
11221127.
Burling, A.S., Burling, T.A., 2003. A comparison of self-report measures of
nicotine dependence among male drug/alcohol-dependent cigarette smokers.
Nicotine Tob. Res. 5, 625633.
Chen, X., Zheng, H., Sussman, S., Gong, J., Stacy, A., Xia, J., Gallaher, P.,
Dent, C., Azen, S., Shan, J., Unger, J.B., Johnson, C.A., 2002. Use of
the Fagerstrom Tolerance Questionnaire for measuring nicotine dependence
among adolescent smokers in China: a pilot test. Psychol. Addict. Behav.
16, 260263.
Clark, D.B., Wood, D., Martin, C.S., Cornelius, J.R., Lynch, K.G., Shiffman, S.,
2005. Multidimensional assessment of nicotine dependence in adolescents.
Drug Alcohol Depend. 77, 235242.
Clayton, R.R., 2004. The importance of adolescence in the development of nicotine dependence: introduction to part V. Ann. NY Acad. Sci. 1021, 162166.
Cohen, L.M., Myers, M.G., Kelly, J.F., 2002. Assessment of nicotine dependence among substance abusing adolescent smokers: a comparison of the
DSM-IV criteria and the modified Fagerstrom Tolerance Questionnaire. J.
Psychopathol. Behav. 24, 225.
Colby, S.M., Tiffany, S.T., Shiffman, S., Niaura, R.S., 2000a. Are adolescent
smokers dependent on nicotine? A review of the evidence. Drug Alcohol
Depend. 59, 83.
Colby, S.M., Tiffany, S.T., Shiffman, S., Niaura, R.S., 2000b. Measuring nicotine
dependence among youth: a review of available approaches and instruments.
Drug Alcohol Depend. 59, S23S39.
Dierker, L., Lloyd-Richardson, E., Stolar, M., Flay, B., Tiffany, S., Collins, L.,
Bailey, S., Nichter, M., Nichter, M., Clayton, R., 2006a. The proximal association between smoking and alcohol use among first year college students.
Drug Alcohol Depend. 81, 19.
Dierker, L.C., Donny, E., Tiffany, S., Colby, S.M., Perrine, N., Clayton,
R., TERN, 2006. The association between cigarette smoking and DSMIV nicotine dependence among first year college students. Drug Alcohol
Depend.
DiFranza, J.R., Rigotti, N.A., McNeill, A.D., Ockene, J.K., Savageau, J.A.,
Cyr, D.S., Coleman, M., 2000. Initial symptoms of nicotine dependence in
adolescents. Tob. Control 9, 313319.
DiFranza, J.R., Savageau, J.A., Fletcher, K., Ockene, J.K., Rigotti, N.A.,
McNeill, A.D., Coleman, M., Wood, C., 2002a. Measuring the loss of
autonomy over nicotine use in adolescents: the DANDY (Development and
Assessment of Nicotine Dependence in Youths) study. Arch. Pediatr. Adolesc. Med. 156, 397403.

DiFranza, J.R., Savageau, J.A., Rigotti, N.A., Fletcher, K., Ockene, J.K.,
McNeill, A.D., Coleman, M., Wood, C., 2002b. Development of symptoms
of tobacco dependence in youths: 30-month follow-up data from the DANDY
study. Tob. Control 11, 228235.
Etter, J.-F., 2005. A comparison of the content-, construct- and predictive validity of the cigarette dependence scale and the Fagerstrom Test for Nicotine
Dependence. Drug Alcohol Depend. 77, 259268.
Etter, J.-F., Duc, T.V., Perneger, T.V., 1999. Validity of the Fagerstrom Test
for Nicotine Dependence and of the heaviness of smoking index among
relatively light smokers. Addiction 94, 269281.
Fagerstrom, K.-O., 1978. Measuring degree of physical dependence to tobacco
smoking with reference to individualization of treatment. Addict. Behav. 3,
235241.
Fagerstrom, K.O., Schneider, N.G., 1989. Measuring nicotine dependence: a
review of the Fagerstrom Tolerance Questionnaire. J. Behav. Med. 12,
159.
Haddock, C., Lando, H., Klesges, R.C., Talcott, G., Renaud, E.A., 1999. A
study of the psychometric and predictive properties of the Fagerstrom Test
for Nicotine Dependence in a population of young smokers. Nicotine Tob.
Res. 1, 5966.
Heatherton, T.F., Kozlowski, L.T., Frecker, R.C., Fagerstrom, K.-O., 1991. The
Fagerstrom Test for Nicotine Dependence: a revision of the Fagerstrom Tolerance Questionnaire. Br. J. Addict. 86, 11191127.
Heatherton, T.F., Kozlowski, L.T., Frecker, R.C., Rickert, W., et al., 1989. Measuring the heaviness of smoking: using self-reported time to the first cigarette
of the day and number of cigarettes smoked per day. Br. J. Addict. 84,
791799.
Hughes, J.R., Oliveto, A.H., Riggs, R., Kenny, M., Liguori, A., Pillitteri, J.L.,
MacLaughlin, M.A., 2004. Concordance of different measures of nicotine
dependence: two pilot studies. Addict. Behav. 29, 15271539.
Hunsley, J., Meyer, G.J., 2003. The incremental validity of psychological testing
and assessment: conceptual, methodological, and statistical issues. Psychol.
Assess. 15, 446455.
Kandel, D., Schaffran, C., Griesler, P., Samuolis, J., Davies, M., Galanti, R.,
2005. On the measurement of nicotine dependence in adolescence: comparisons of the mFTQ and a DSM-IV-based scale. J. Pediatr. Psychol. 30,
319332.
Kear, M.E., 2002. Psychosocial determinants of cigarette smoking among college students. J. Community Health Nurs. 19, 245257.
Lichtenstein, E., Mermelstein, R.J., 1986. Some methodological cautions in the
use of the Tolerance Questionnaire. Addict. Behav. 11, 439442.
Mayhew, K.P., Flay, B.R., Mott, J.A., 2000. Stages in the development of adolescent smoking. Drug Alcohol Depend. 59, 6181.
Nichter, M., Nichter, M., Thompson, P.J., Shiffman, S., Moscicki, A.-B., 2002.
Using qualitative research to inform survey development on nicotine dependence among adolescents. Drug Alcohol Depend. 68, 41.
Nichter, M., Nichter, M., Vuckovic, N., Quintero, G., Ritenbaugh, C., 1997.
Smoking experimentation and initiation among adolescent girls: qualitative
and quantitative findings. Tob. Control 6, 285295.
OLoughlin, J., DiFranza, J., Tarasuk, J., Meshefedjian, G., McMillan-Davey,
E., Paradis, G., Tyndale, R.F., Clarke, P., Hanley, J., 2002a. Assessment of
nicotine dependence symptoms in adolescents: a comparison of five indicators. Tob. Control 11, 354360.
OLoughlin, J., DiFranza, J., Tyndale, R.F., Meshefedjian, G., McMillan-Davey,
E., Clarke, P.B.S., Hanley, J., Paradis, G., 2003. Nicotine-dependence symptoms are associated with smoking frequency in adolescents. Am. J. Prev.
Med. 25, 219225.
OLoughlin, J., Kishchuk, N., DiFranza, J., Tremblay, M., Paradis, G., 2002b.
The hardest thing is the habit: a qualitative investigation of adolescent
smokers experience of nicotine dependence. Nicotine Tob. Res. 4, 201
209.
OLoughlin, J., Tarasuk, J., Difranza, J., Paradis, G., 2002c. Reliability of
selected measures of nicotine dependence among adolescents. Ann. Epidemiol. 12, 353.
Payne, T.J., Smith, P.O., McCracken, L.M., McSherry, W.C., Antony, M.M.,
1994. Assessing nicotine dependence: a comparison of the Fagerstrom Tolerance Questionnaire (FTQ) with the Fagerstrom Test for Nicotine Dependence (FTND) in a clinical sample. Addict. Behav. 19, 307.

E.M. Sledjeski et al. / Drug and Alcohol Dependence 87 (2007) 1019

Pomerleau, C.S., Carton, S.M., Lutzke, M.L., Flessland, K.A., Pomerleau, O.F.,
1994. Reliability of the Fagerstrom Tolerance Questionnaire and the Fagerstrom Test for Nicotine Dependence. Addict. Behav. 19, 33.
Prokhorov, A.V., De Moor, C., Pallonen, U.E., Hudmon, K.S., Koehly, L., Hu,
S., 2000. Validation of the modified Fagerstrom Tolerance Questionnaire
with salivary cotinine among adolescents. Addict. Behav. 25, 429433.
Prokhorov, A.V., Koehly, L.M., Pallonen, U.E., Hudmon, K.S., 1998. Adolescent nicotine dependence measured by the modified Fagerstrom Tolerance
Questionnaire at two time points. J. Child Adol. Subst. 7, 3547.
Prokhorov, A.V., Pallonen, U.E., Fava, J.L., Ding, L., et al., 1996. Measuring
nicotine dependence among high-risk adolescent smokers. Addict. Behav.
21, 117127.
SAMHSA, 2002. Results from the 2001 National Household Survey on Drug
Abuse, vol. I, Summary of National Findings (Office of Applied Studies,
NHSDA Series H-17, DHHS Publication No. SMA 02-3758). Rockville,
MD.
Shiffman, S., 1991. Refining models of dependence: variations across persons
and situations. Addiction 86, 611615.
Shiffman, S., Sayette, M.A., 2005. Validation of the Nicotine Dependence Syndrome Scale (NDSS): a criterion-group design contrasting chippers and
regular smokers. Drug Alcohol Depend. 79, 45.
Shiffman, S., Waters, A.J., Hickcox, M., 2004. The Nicotine Dependence Syndrome Scale: a multidimensional measure of nicotine dependence. Nicotine
Tob. Res. 6, 327348.

19

Strong, D.R., Brown, R.A., Ramsey, S.E., Myers, M.G., 2003. Nicotine dependence measures among adolescents with psychiatric disorders: evaluating
symptom expression as a function of dependence severity. Nicotine Tob.
Res. 5, 735746.
Tiffany, S.T., Agnew, C., Maylath, N., Dierker, L.C., Flaherty, B.P., Richardson,
E., Segress, M., Clayton, R., TERN, unpublished. Smoking and College
Freshman: University Project of the Tobacco Etiology Research Network
(UPTERN). Technical Report.
Tiffany, S.T., Conklin, C.A., Shiffman, S., Clayton, R.R., 2004. What can dependence theories tell us about assessing the emergence of tobacco dependence?
Addiction 99, 7886.
Wellman, R.J., DiFranza, J.R., Pbert, L., Fletcher, K.E., Flint, A., Young, M.H.,
Druker, S., 2006. A comparison of the psychometric properties of the Hooked
On Nicotine Checklist and the modified Fagerstrom Tolerance Questionnaire. Addict. Behav. 31, 486495.
Wellman, R.J., DiFranza, J.R., Savageau, J.A., Godiwala, S., Friedman, K.,
Hazelton, J., 2005. Measuring adults loss of autonomy over nicotine
use: the Hooked On Nicotine Checklist. Nicotine Tob. Res. 7, 157
161.
Wheeler, K.C., Fletcher, K.E., Wellman, R.J., Difranza, J.R., 2004. Screening
adolescents for nicotine dependence: the Hooked On Nicotine Checklist. J.
Adolesc. Health 35, 225.
WHO, 1994. Composite International Diagnostic Interview (CIDI) Researchers
Manual. American Psychiatric Association, Washington, DC.