Nicotine & Tobacco Research, 2015, 186–192

doi:10.1093/ntr/ntu204
Advance Access publication October 19, 2014
Original investigation

Original investigation

Development of a Questionnaire for Assessing

Dependence on Electronic Cigarettes Among a
Large Sample of Ex-Smoking E-cigarette Users
Jonathan Foulds PhD1, Susan Veldheer MS1, Jessica Yingst MS1,
Shari Hrabovsky MSN1, Stephen J. Wilson PhD2, Travis T. Nichols MS 2,

Downloaded from http://ntr.oxfordjournals.org/ at New York University on May 29, 2015

Thomas Eissenberg PhD3
1
Department of Public Health Sciences, Tobacco Center of Regulatory Science, College of Medicine, Penn State
University, Hershey, PA; 2Department of Psychology, Penn State University, State College, PA; 3Center for the Study
of Tobacco Products, Virginia Commonwealth University, Richmond, VA

Corresponding Author: Jonathan Foulds, PhD, Department of Public Health Sciences, 500 University Dr, Rm T3429, MC
CH69, Hershey, PA 17033, USA. Telephone: 717-531-3504; Fax: 717-531-0480; E-mail: [email protected]

Abstract
Introduction: Electronic cigarettes (e-cigs) are becoming increasingly popular, but little is known
about their dependence potential. This study aimed to assess ratings of dependence on electronic
cigarettes and retrospectively compare them with rated dependence on tobacco cigarettes among
a large sample of ex-smokers who switched to e-cigs.
Methods: A total of 3,609 current users of e-cigs who were ex-cigarette smokers completed a
158-item online survey about their e-cig use, including 10 items designed to assess their previous
dependence on cigarettes and 10 almost identical items, worded to assess their current depend-
ence on e-cigs (range 0–20).
Results: Scores on the 10-item Penn State (PS) Cigarette Dependence Index were significantly
higher than on the comparable PS Electronic Cigarette Dependence Index (14.5 vs. 8.1, p < .0001).
In multivariate analysis, those who had used e-cigs longer had higher e-cig dependence scores, as
did those using more advanced e-cigs that were larger than a cigarette and had a manual button.
Those using zero nicotine liquid had significantly lower e-cig dependence scores than those using
1–12 mg/ml, who scored significantly lower than those using 13 or greater mg/ml nicotine liquid
(p < .003).
Conclusions: Current e-cigarette users reported being less dependent on e-cigarettes than they retro-
spectively reported having been dependent on cigarettes prior to switching. E-cig dependence appears
to vary by product characteristics and liquid nicotine concentration, and it may increase over time.

Introduction risks and benefits of e-cigs will require the systematic collection of
data on patterns of use, product characteristics, characteristics of
The popularity of electronic cigarettes (e-cigs) has increased rapidly users, effects on toxicant exposure, and effects on health, includ-
in many countries over the past 5 years, and in some countries, as ing dependence. As e-cigs were invented in 2003 and only became
many as a third of current or recent ex-cigarette smokers have tried popular in recent years, much remains unknown about their safety or
e-cigs.1–3 The most recent data from a large representative sample of dependence potential.
US adults found that 4.2% (of all adults) used electronic cigarettes There are a number of brief questionnaires with proven utility
every day, some days, or rarely.4 Assessment of the likely public health in assessing dependence in cigarette users. The Fagerström Test for

© The Author 2014. Published by Oxford University Press on behalf of the Society for Research on Nicotine and Tobacco. All rights reserved. 186
For permissions, please e-mail: [email protected].
Nicotine & Tobacco Research, 2015, Vol. 17, No. 2 187

Nicotine Dependence (FTND) is the most widely used method of comprising 10 items. Two of these (accounting for 10 points) were
assessing cigarette dependence. Many studies have found that two adapted from the FTND/HSI (requiring participants to write the
specific items (cigarettes/day and time to first cigarette of the day) are actual numbers, rather than check pre-defined range options), five
more predictive of difficulty quitting (and other indices of depend- are from the HONC (covering difficulty quitting, experience of crav-
ence) than the other items and so these two items have come to be ing, and withdrawal symptoms), two assess waking at night to use
widely used and together have come to be known as the Heaviness (adapted from Bover et al.16), and one assesses recent strength of
of Smoking Index (HSI).5–8 urges to use (adapted from Fiddler et al.18). The items and scoring
With the recent proliferation of nicotine containing products, it guide are shown in Box 1.
has become clear that there is a need for measures capable of assess- We then piloted the Cigarette Dependence Index in a smoking
ing nicotine dependence in other products.9,10 However, given that cessation trial in 225 smokers who received nicotine patches and
other nicotine delivery products may have different patterns of group support for a month. As expected, the index predicted 1-month
use, there may be advantages in developing methods of assessing intent-to-treat CO-verified abstinence, such that those scoring 0–10
dependence in ways that can facilitate comparison across products,
but not be as dominated by heaviness or frequency of use as the
FTND or HSI. The Wisconsin Inventory of Smoking Dependence Box 1. Penn State [Electronic] Cigarette Dependence Index
Motives (WISDM) is a longer (68-item) multisubscale assessment,
[For the Electronic Cigarette Dependence Index, substitute the
but neither the full score nor most of the subscales are as predictive
underlined word with the words in square brackets]
of cessation outcomes as the much briefer FTND.11 The Hooked On

Downloaded from http://ntr.oxfordjournals.org/ at New York University on May 29, 2015

1. How many cigarettes [times] per day do you usually smoke [use
Nicotine Checklist (HONC) was developed specifically to assess the your electronic cigarette]?a ([assume that one “time” consists
development of dependence in young people.12 It is a 10-item ques- of around 15 puffs or lasts around 10 minutes])
tionnaire that has good psychometric properties, but many adult (Scoring: 0–4 times/day = 0, 5–9 = 1, 10–14 = 2, 15–19 = 3,
smokers reach the ceiling score and the average smoker trying nico- 20–29 = 4, 30+ = 5)
tine replacement therapy scores 8/10 and so it is less sensitive at the 2. On days that you can smoke [use your electronic cigarette]
high end of the spectrum.13 freely, how soon after you wake up do you smoke your first
The FTND is not designed to help assess DSM-4 or 5 criteria cigarette of the day [first use your electronic cigarette]?a
(Scoring: 0–5 mins = 5, 6–15 = 4, 16–30 = 3, 31–60 = 2,
for nicotine dependence,14 and the FTND does not relate well to
61–120 = 1, 121+ = 0)
DSM-based assessments.15 Given the strengths and weaknesses of
3. Do you sometimes awaken at night to have a cigarette [use
existing measures of nicotine dependence, we developed a new nico- your electronic cigarette]?b
tine dependence index comprising some of the most predictive items (Scoring: Yes = 1, No = 0)
from various prior studies, but designed to be easily adapted for dif- 4. If yes, how many nights per week do you typically awaken to
ferent nicotine products. These items were partly selected to cover smoke [use your electronic cigarette]?b
each of the main components of dependence (consumption, drive, (Scoring: 0–1 nights = 0, 2–3 nights = 1, 4+ nights = 2)
craving, withdrawal, difficulty quitting) and adapted to use a scor- 5. Do you smoke [use an electronic cigarette] now because it is
ing scheme reflective of evidence that dependence can exist at low really hard to quit?c
levels of consumption,12 and to measure aspects of dependence (e.g., (Scoring: Yes = 1, No = 0)
6. Do you ever have strong cravings to smoke [use an electronic
cravings and withdrawal symptoms) that have been recommended
cigarette]?c
as being of most practical use in assessing smokers.14 One of our
(Scoring: Yes = 1, No = 0)
priorities was to create a brief scale, and we acknowledge that for 7. Over the past week, how strong have the urges to smoke [use
some purposes, it may be worth including more items, including an electronic cigarette] been?d
items specifically on withdrawal-related craving or those included (Scoring: None/Slight = 0, Moderate/Strong= 1, Very Strong/
in the WISDM. Extremely Strong = 2)
Recent studies have identified some questionnaire items that 8. Is it hard to keep from smoking [using an electronic cigarette]
appear to assess additional aspects of nicotine dependence very sim- in places where you are not supposed to?c
ply. For example, Bover et al.16 found that 51% of 2,768 cigarette (Scoring: Yes = 1, No = 0)
smokers attending a tobacco dependence treatment clinic reported When you haven’t used tobacco [an electronic cigarette] for a
while or when you tried to stop smoking [using]…
at baseline assessment that they wake at night to smoke, and that a
9. Did you feel more irritable because you couldn’t smoke [use an
single question about this behavior predicted abstinence at 6-month
electronic cigarette]?c
follow-up in both univariate and multivariate analyses, whereas (Scoring: Yes = 1, No = 0)
neither HSI item did. Scharf et al.17 reported that self-reported 10. Did you feel nervous, restless, or anxious because you couldn’t
night smoking correlated well with electronic diary-measured night smoke [use an electronic cigarette]?c
smoking and that night smoking quantity predicted lapse risk even   (Scoring: Yes = 1, No = 0)
when controlling for traditional (less predictive) measures of nico-
tine dependence. This suggests that quantity of night smoking may Total scoring: 0–3= not dependent, 4–8 low dependence, 9–12 medium
capture variance in smoking patterns that is not captured by exist- dependence, 13+ = high dependence.
The original questionnaire asks about cigarette use and electronic ciga-
ing items or scales (FTND or Nicotine Dependence Syndrome Scale
rette use separately, and the wording for e-cig questions is provided here in
items). Fidler et al.18 found that strength of smoking urges during
brackets merely to facilitate comparisons and save journal space.
a recent time period of normal smoking (past 24 hrs or past week) a
From FTND/HSI.
predicted difficulty quitting better than the FTND/HSI scores and b
From Bover et al.16
may be a useful measure of cigarette addiction. c
From Hooked on Nicotine Checklist.
After completing a review of existing questionnaire assess- d
From Fiddler et al.18
ments of dependence, we developed a Cigarette Dependence Index
188 Nicotine & Tobacco Research, 2015, Vol. 17, No. 2

at baseline had a 74% quit rate, those scoring 11–14 had a 52% The dataset was first checked for completeness and likely dupli-
quit rate, and those scoring 15–20 had a 27% quit rate. This is not cate completions. Likely duplications were identified as (a) those
surprising as the scale comprises items that have been shown to pre- with the same demographics and e-mail address or (b) those with
dict difficulty quitting in numerous prior studies. We piloted similar identical state, city, gender, age, race, and occupation. Where likely
questions in face-to-face interviews with 108 electronic cigarette duplicates were identified, neither case was included. Cases with
users19 and found that the only question needing adaptation/clari- incomplete demographics or e-cig history were not included, leaving
fication for electronic cigarette users was the item about frequency 6,745 with complete data. Of these, 5,462 were current regular e-cig
of use. Some participants mentioned that this was difficult to answer users (had used e-cigs on at least 30 days, including the past 7 days).
because, unlike a cigarette that is generally smoked as a whole and A total of 5,363 (98%) had ever been a cigarette smoker of whom
then discarded, e-cigs can be used more frequently but perhaps with 3,859 (72%) had quit smoking for at least 30 days. A total of 193
fewer puffs per “session,” because there is no need to “finish the cig” were excluded as they had recently used another tobacco or smoking
in one sitting. We found that the vast majority of e-cig users are able cessation product and 57 had incomplete data on their prior ciga-
to estimate their total amount of e-cig use in a way that is compara- rette use, leaving a final sample of 3,609 who were exclusive current
ble with cigarettes, by specifying, “How many times per day do you e-cig users and ex-cigarette smokers.
usually use your electronic cigarette? (assume one ‘time’ consists of The data were analyzed using the SAS 9.3 statistical package.
around 15 puffs, or lasts around 10 min).” Means and percentages were calculated in order to characterize the
Having developed an index of nicotine dependence that could be overall sample. Paired t-tests (two-tailed) were used to compare
easily adapted for use in assessing both cigarette and e-cig depend- continuous variables and chi-squared tests were used to compare

Downloaded from http://ntr.oxfordjournals.org/ at New York University on May 29, 2015

ence, this study aimed to compare ratings of dependence on elec- categorical variables (generally comparing the same individuals’ cur-
tronic cigarettes with tobacco cigarettes in a sample of ex-smokers rent e-cig use with their prior cigarette use behavior). When compar-
who have switched to e-cigs. This study also aimed to assess the ing E-Cig Dependence Index scores by response characteristics with
validity of the e-cig dependence index by examining the relation- more than two options, one-way analysis of variance was used with
ship of the e-cig dependence index with the concentration of nicotine Tukey adjustment for multiple comparisons. Multivariate regression
in the liquid being used. It was hypothesized that e-cig dependence was used to identify predictors of e-cig dependence using the SAS
would be higher in those using higher concentrations of nicotine and stepwise procedure (iterative backwards and forwards stepwise).
in those who have used e-cigs for a longer period of time.

Results
Methods The main descriptive characteristics of the sample are shown in
Electronic cigarette users were invited to complete an online survey Table 1. The sample was predominantly White, male, and living in
aiming to improve understanding of the use of electronic cigarettes, the United States. They reported making multiple attempts to quit
including the types of e-cigs being used, how frequently they are smoking and had on average been using e-cigs for over a year. Most
used and whether or not they are replacing other types of tobacco reported that they quit smoking cigarettes shortly after they started
among e-cig users aged ≥18 years old. Participation in this study was using e-cigs, but almost 9% had quit smoking cigarettes prior to
voluntary and respondents could remain anonymous, although indi- using e-cigs. However, the majority of these reported that they were
viduals who wished to volunteer for additional research on e-cigs still using another traditional tobacco product when they started
were invited to enter their contact details at the end of the survey. e-cig use and all reported that they were free of traditional tobacco
The survey was administered and responses were stored on REDCap use at the time of the survey. Only 13% were using first-generation
(Research Electronic Data Capture). REDCap is a secure, web-based e-cigs that are the same size and shape as a regular cigarette (“a ciga-
application designed to support data capture for research studies. like”), and most were using more advanced e-cigs that included the
REDCap is maintained by the REDCap Consortium, which is com- ability to press a button to activate the heating coil manually prior
prised of more than 600 institutional partners including Penn State to puffing and the ability to control the voltage. They had tried an
University. This survey was comprehensive; containing 158 items average of five different e-cig models, and almost three quarters had
including demographics, e-cig use history, types of e-cig used, tobacco switched to their current brand because it gave them a more satisfy-
use history, and various free text responses. Embedded within the sur- ing “hit.”
vey were the 10 questions that made up the PS Electronic Cigarette Table 2 shows a comparison of the total Penn State Nicotine
Dependence Index and its 10-item cigarette equivalent. The cigarette Dependence Index and the individual items for current e-cig use and
version was preceded by the following statement: “Think back to past cigarette use. This table shows that although the e-cig users were
a time when you were primarily a traditional cigarette smoker…. having e-cig sessions with a very similar frequency to the number
before you used e-cigs. To the best of your ability, answer the follow- of cigarettes they had previously consumed each day, their overall
ing questions regarding your cigarette smoking at that time.” This E-Cig Dependence Index was significantly lower than their Cigarette
survey was first posted on the Internet in December 2012 (https:// Dependence Index, as was the individual score on every other item.
redcap.ctsi.psu.edu/redcap/surveys/?s=v94cbA). Links to the survey Over ninety percent reported that they had experienced strong urges
were posted on a variety of websites including www.webMD.com, to smoke and withdrawal symptoms when a smoker, but only 25
and sites frequented by e-cig users, including www.e-cigarette-forum. to 35% reported experiencing these symptoms of dependence as an
com, one of the largest e-cig user websites. Visitors to these sites were e-cig user.
also able to cross-post the survey link to friends and other websites. All the variables in Table 1 were entered in a stepwise regression
This study was approved by the Penn State University Institutional model (ejecting any variables not predictive of E-Cig Dependence
Review Board. Data reported here were collected from December Index at p < .15). The final model found that even when adjusting
2012 to May 2014. for all other significant predictors, women, Whites, those without a
Nicotine & Tobacco Research, 2015, Vol. 17, No. 2 189

college education, who are older, who have used e-cigs for a longer regression, it too was a highly significant predictor of the E-cig
time, who have previously tried more e-cig models, who currently Dependence Index (p < .001) and the two scores had a Pearson cor-
use an e-cig larger than a cigarette, with a button, with more than relation of +0.35 (p < .0001).
one battery, that cost over $50 and who use a higher concentra- Figure 1 shows the number and proportion of e-cig users using
tion of nicotine liquid, tend to have a higher e-cig dependence index different levels of nicotine concentration in their e-liquid. Figure 2
(all p < .05). When Cigarette Dependence Index was entered in the shows the relationship between the nicotine concentration and the
e-cig dependence index, after controlling for all the other significant
predictors. Those using zero nicotine liquid had a significantly lower
Table 1. Demographic, Tobacco Use, and Electronic Cigarette Use e-cig dependence index than those using 1–12 mg/ml (p < .001), who
Characteristics of the Sample were significantly lower than those using 13 or greater mg/ml nico-
Characteristic N = 3,609 tine liquid (p < .001).
The online survey included questions about whether respond-
Male, % (n) 72.2 (2,604) ents had advocated for e-cigs or earned income from selling e-cig-
Mean age (SD) 40.5 (12.6) arette merchandise. It is conceivable that some of the volunteers
White, % (n) 92.0 (3,319)
for this study who were advocates for e-cigarettes may therefore
With college degree, % (n) 40.7 (1,467)
have chosen to underestimate their dependence on e-cigs rela-
Live in United States, % (n) 84.8 (3,062)
Mean number of lifetime cigarette 8.6 (15.5) tive to cigarettes in order to make e-cigs appear more favora-
ble. However, when the question about prior e-cig advocacy was

Downloaded from http://ntr.oxfordjournals.org/ at New York University on May 29, 2015

quit attempts (SD)
Mean number of months since quitting 12.7 (12.2) included in the regression model, it did not alter the pattern of
smoking (SD) (n = 3,281) results. More importantly, those who had advocated for e-cigs
Mean length of e-cig use in months (SD) 13.4 (12.9) online had higher e-cig dependence scores, which is not consistent
Mean E-liquid nicotine concentration, 15.5 (8.6) with the notion that they may have minimized their e-cig depend-
mg/ml (SD) ence ratings.
Mean number of E-cigs used before 4.2 (4.1) The first 2 items of the dependence index comprise a 10-point
current model (n = 3,601)
version comparable with the HSI, whereas the next 8 items com-
Use e-cig liquid >12 mg/ml, % (n) 50.8 (1,834)
prise a 10-point measure of specific dependence symptoms. On the
Who started using E-cigs with intention 74.4 (2,684)
to quit smoking, % (n) Cigarette Dependence Index, the two subscores were similar (7.7 vs.
Who quit smoking cigarettes prior to 8.8 (317) 6.8) and moderately correlated (Pearson correlation = +0.51, p <
using E-cig, % (n) .0001), whereas on the E-cig Dependence Index, there was a lower
Who plan to continue using e-cig for 78.8 (2,845) mean score for the specific symptom subscore (6.1 vs. 2.0) and a
at least another year, % (n) lower correlation between the two subscores (Pearson correlation =
E-cig larger than regular cigarette, % (n) 87.1 (3,145) +0.23, p < .001).
E-cig cost > $50 (n = 3,583), % (n) 55.7 (1,994)
E-cig purchased at gas (petroleum) 9.2 (331)
station, % (n) Discussion
Use only Vegetable Glycerin 8.8 (234)
E-liquid, % (n) This study found that although ex-smoking e-cig users use their
E-cig with button to press prior 84.7 (3,056) e-cigs about as much as they smoked their cigarettes, they appear
to inhalation, % (n) to be significantly less dependent on their e-cigs than they were on
E-cig with button to control battery 61.9 (2,234) their cigarettes, as assessed on a number of questions that have been
voltage, % (n) shown to be good measures of dependence in cigarette smokers. It
E-cig uses single 3.7-V battery, % (n) 72.7 (2,622) is possible that e-cig users could experience less craving and with-
E-cig requires more than 1 battery, % (n) 4.9 (175)
drawal because their use of their product is less widely restricted.
E-cig uses proprietary battery, % (n) 30.3 (1,095)
However, that does not explain the fact that e-cig users are much less
Switched to current brand because it 74.8 (2,699)
gives a more satisfying “hit,” % (n) likely to waken at night to use an e-cig than they had been to smoke,
or that they wait significantly longer before using after waking in

Table 2. Penn State Nicotine Dependence Index: Comparison of Total Score and Individual Item Responses for Smoking Cigarettes and
Using Electronic Cigarettes

Penn State Nicotine Dependence Index Traditional cigarette Electronic cigarette p

Mean PSNDI score (SD) 14.5 (3.7) 8.1 (3.5) <.0001

Mean cigarettes (e-cig sessions) per day (SD) 24.3 (12.4) 24.0 (25.9) .4493
Mean time to first smoke (e-cig use) of the day, in mins (SD) 26.8 (67.3) 44.5 (72.7) <.0001
Awakening at night to smoke (use e-cig), % (n) 41.2 (1,487) 6.8 (247) <.0001
Who smoke (use e-cig) because it is really hard to quit, % (n) 89.4 (3,225) 28.8 (1,039) <.0001
Had strong cravings to smoke (use e-cig), % (n) 92.7 (3,344) 35.4 (1,277) .0037
With strong, very strong, or extremely strong urges to smoke (use e-cig), % (n) 85.7 (3,093) 11.72 (423) .0001
Find it hard to keep from using in places where you are not supposed to, % (n) 64.5 (2,328) 10. 5 (378) .0014
More irritable when they cannot smoke (use e-cig), % (n) 91.7 (3,312) 25.6 (923) <.0001
More nervous when they cannot smoke (use e-cig), % (n) 90.4 (3,263) 26.4 (953) <.0001
190 Nicotine & Tobacco Research, 2015, Vol. 17, No. 2

35 32.9
31.4
30

25
19.9
20

%
15 13.7

10

5 2.2
0
0 mg (n=74) 1-6 mg 7-12 mg 13-18 mg 19+ mg
(n=474) (n=1085) (n=1130) (n=683)

Downloaded from http://ntr.oxfordjournals.org/ at New York University on May 29, 2015

E-cig Liquid Nicone Concentraon (mg/ml)

Figure 1. E-cig liquid nicotine concentration, % (n = 3,478).

10
8.8
9 8.3
7.4 7.7
8
Mean Dependence Score

7
6 5.4
5
4
3
2
1
0
0 mg (n=74) 1-6 mg 7-12 mg 13-18 mg 19+ mg
(n=474) (n=1085) (n=1130) (n=683)
E-cig Liquid Nicone Concentraon (mg/ml)

Figure 2. Adjusteda Penn State Electronic Cigarette Dependence Index by e-cig nicotine concentration* (n = 3,446). aAdjusted for gender, age, race, education
level, days used an e-cig, e-cig size, e-cig button, battery, and number of e-cig. *All between group p values <.003 except between (1) 1–6-mg and 7–12-mg
group, and (2) 13–18 and 19+ group.

the morning than they did as cigarette smokers. Farsalinos et al.20 We also found that users of more advanced e-cigs (in terms of
also found that e-cig users reported lower dependence on e-cigs than size, battery power, ability to control timing of heat being applied
prior cigarette smoking, in a sample of 111 e-cig users with carbox- to the coil, cost of e-cig, and so on), score higher on measures of
yhemoglobin-validated abstinence from smoking. dependence. This is also consistent with existing data, showing that
Overall, the most parsimonious explanation for these results advanced generation models deliver higher blood nicotine levels
is that e-cig users are generally less nicotine dependent than they than current “cigalike” models.25,26
were as cigarette smokers. This is perhaps not surprising given that This study found that scores on the PS Electronic Cigarette
almost all the studies of nicotine absorption from e-cigs suggest that Index were significantly related to the concentration of nicotine in
they deliver markedly lower peak nicotine levels compared with the e-cig liquid being used, even after controlling for other predic-
cigarettes, and that they deliver nicotine more slowly than cigarettes, tors. This may not be very surprising, but is at least consistent with
both of which are characteristics likely leading to less addiction.21–25 the idea that the items selected for the index are somewhat sensitive
Nicotine & Tobacco Research, 2015, Vol. 17, No. 2 191

to nicotine consumption after controlling for frequency of use. It Funding

should be noted that even when we analyzed only those e-cig users
This work was supported by an internal grant from Penn State Social Science
who were using e-cigs that were larger than a regular cigarette, had Research Institute and Cancer Institute (PI: SJW). JF, SV, JY, and SH are
a button, with a nicotine concentration greater than 12 mg/ml, for primarily funded by the National Institute on Drug Abuse of the National
at least 18 months (n = 405), their E-Cig Dependence Index was Institutes of Health and the Center for Tobacco Products of the US Food and
still lower than their Cigarette Dependence Index (9.2 vs. 15.4, p < Drug Administration (P50-DA-036107-01 and P50-DA-0361-05). TE is sup-
.0001). ported by the National Institute on Drug Abuse of the National Institutes
The “uncoupling” of the close relationship between heaviness of of Health (P50-DA-0361-05) and the Center for Tobacco Products of the US
use (as measured by times per day and time to first use of the day) Food and Drug Administration.
and the strength of the other dependence symptoms (e.g., strength
of urges and withdrawal symptoms) in e-cig users may be due to
Declaration of Interests
the wider variation in nicotine absorption from different e-cigs than
exists for cigarettes. This underlines the importance of assessing spe- JF has done paid consulting for pharmaceutical companies involved in produc-
ing smoking cessation medications, including GSK, Pfizer, Novartis, J&J, and
cific dependence symptoms when assessing dependence in e-cig users.
Cypress Bioscience.
Weaknesses of this study include the nonrepresentative nature of
The content is solely the responsibility of the authors and does not neces-
the volunteer sample. Clearly those who found out about the survey
sarily represent the official views of the National Institutes of Health or the US
on specialist websites and took the time to complete the survey are a Food and Drug Administration.
particularly experienced and likely “pro-e-cig” sample of e-cig users,

Downloaded from http://ntr.oxfordjournals.org/ at New York University on May 29, 2015

and it is possible their answers were designed to make e-cigs look
“good” relative to traditional cigarettes. However, we found a simi- References
lar pattern of results in the 97% of participants who had never sold 1. Brown J, West R, Beard E, Michie S, Shahab L, McNeill A. Prevalence
e-cigarette merchandise and also in the 58% of participants who had and characteristics of e-cigarette users in Great Britain: find-
never advocated for e-cigs online. Given that the dependence scales ings from a general population survey of smokers. Addic Behav
were embedded within more than 120 other questions, we also think 2014;39(6):1120–1125.
2. Carroll Chapman SL, Wu L. E-cigarette prevalence and correlates of use
it doubtful that those completing these items were aware that they
among adolescents versus adults: a review and comparison. J Psychiatr
would be used to assess dependence and compare products. Another
Res 2014;54:43–54.
potential weakness is the lack of prior studies of validity and reli-
3. Goniewicz ML, Zielinska-Danch W. Electronic cigarette use among teen-
ability of the PS Electronic Cigarette Dependence Index. However, agers and young adults in Poland. Pediatrics 2012;130(4):e879–e885.
the fact that these items all have a good track record as indicators of 4. Agaku IT, King BA, Husten CG, et al. Tobacco product use among
dependence in cigarette users and were positively associated with use adults--United States, 2012–2013. MMWR Morb Mortal Wkly Rep
of higher nicotine concentration e-liquid in this study are supportive 2014;63(25):542–547.
of the validity of the measure. However, with regards to the direct 5. Fagerström K-O, Russ C, Yu CR, Yunis C, Foulds J. The Fagerström Test
comparison between current dependence on e-cigs and retrospective for Nicotine Dependence as a predictor of smoking abstinence: a pooled
ratings of dependence on traditional cigarettes, we acknowledge that analysis of Varenicline Clinical Trial Data. Nicotine Tob Res 2012.
6. Heatherton TF, Kozlowski LT, Frecker RC, Fagerström K-O. The
there is an inherent weakness in this design and a possibility the
Fagerström Test for Nicotine Dependence: a revision of the Fagerström
respondents could attempt to justify their current behavior by mak-
Tolerance Questionnaire. Br J Addic 1991;86(9):1119–1127.
ing it seem more acceptable (less addiction-based) than their for-
7. Kozlowski LT, Porter CQ, Orleans CT, Pope MA, Heatherton T. Predicting
mer smoking. Prospective studies that include biomarkers of intake smoking cessation with self-reported measures of nicotine dependence:
and concurrent ratings of cravings and withdrawal symptoms will FTQ, FTND, and HSI. Drug Alcohol Depen 1994;34(3):211–216.
be required to more fully compare dependence in those switching 8. Baker TB, Piper ME, McCarthy DE, et al. Time to first cigarette in the
from cigarettes to e-cigs. Finally, given that the participants were morning as an index of ability to quit smoking: implications for nicotine
being asked to retrospectively recall their smoking behavior a year dependence. Nicotine Tob Res 2007;9(suppl 4):S555–S570.
or more earlier, it may have been inaccurate. Hudmon et al.27 studied 9. De Leon E, Smith KC, Cohen JE. Dependence measures for non-cigarette
the recall of FTND nicotine dependence retrospective assessments tobacco products within the context of the global epidemic: a systematic
review. Tob Control 2014;23(3):197–203.
5–12 years later and concluded that these have acceptable reliability.
10. Fagerström K-O, Eissenberg T. Dependence on tobacco and nicotine
Strengths of this study lie in the large sample size and the com-
products: a case for product-specific assessment. Nicotine Tob Res
prehensiveness of the overall survey, enabling analyses that can
2012;14(11):1382–1390.
assess and control for a range of factors that may influence e-cig 11. Piper ME, McCarthy DE, Bolt DM, et al. Assessing dimensions of nicotine
dependence, including the advocacy involvement of the respondents. dependence: an evaluation of the Nicotine Dependence Syndrome Scale
Controlling for these variables did not make substantial differences (NDSS) and the Wisconsin Inventory of Smoking Dependence Motives
to the pattern of results, suggesting that the findings are fairly robust. (WISDM). Nicotine Tob Res 2008;10(6):1009–1020.
12. DiFranza JR, Savageau JA, Fletcher K, et al. Measuring the loss of
autonomy over nicotine use in adolescents: the DANDY (Development
Conclusions and Assessment of Nicotine Dependence in Youths) study. Arch Pediatr
Adolesc Med 2002;156(4):397–403.
Current e-cigarette users report being less dependent on e-cigarettes
13. Wellman RJ, Savageau JA, Godiwala S, et al. A comparison of the Hooked
than they retrospectively report having been dependent on cigarettes on Nicotine Checklist and the Fagerström Test for Nicotine Dependence
prior to switching. E-cig dependence appears to be related to length in adult smokers. Nicotine Tob Res 2006;8(4):575–580.
of e-cig use, characteristics of the device used that may facilitate 14. Baker TB, Breslau N, Covey L, Shiffman S. DSM criteria for tobacco use
nicotine absorption, and is related to the concentration of nicotine disorder and tobacco withdrawal: a critique and proposed revisions for
used in the e-liquid. DSM-5. Addiction 2012;107(2):263–275.
192 Nicotine & Tobacco Research, 2015, Vol. 17, No. 2

15. Moolchan ET, Radzius A, Epstein DH, et al. The Fagerström Test for and withdrawal, user preferences and nicotine delivery: randomised cross-
Nicotine Dependence and the diagnostic interview schedule: do they diag- over trial. Tob Control 2010;19(2):98–103.
nose the same smokers? Addic Behav 2002;27(1):101–113. 22. Dawkins L, Corcoran O. Acute electronic cigarette use: nicotine deliv-
16. Bover MT, Foulds J, Steinberg MB, Richardson D, Marcella SW. Waking at ery and subjective effects in regular users. Psychopharmacology (Berl)
night to smoke as a marker for tobacco dependence: patient characteristics 2014;231(2):401–407.
and relationship to treatment outcome. Int J Clin Prac 2008;62(2):182–190. 23. Nides MA, Leischow SJ, Bhatter M, Simmons M. Nicotine blood levels
17. Scharf DM, Dunbar MS, Shiffman S. Smoking during the night: prevalence and short-term smoking reduction with an electronic nicotine delivery sys-
and smoker characteristics. Nicotine Tob Res 2008;10(1):167–178. tem. Am J Health Behav 2014;38(2):265–274.
18. Fidler JA, Shahab L, West R. Strength of urges to smoke as a measure of 24. Vansickel AR, Cobb CO, Weaver MF, Eissenberg TE. A clinical labora-
severity of cigarette dependence: comparison with the Fagerström Test for tory model for evaluating the acute effects of electronic “cigarettes”:
Nicotine Dependence and its components. Addiction 2011;106(3):631–638. nicotine delivery profile and cardiovascular and subjective effects. Cancer
19. Foulds J, Veldheer S, Berg A. Electronic cigarettes (e-cigs): views of Epidemiol Biomarkers Prev 2010;19(8):1945–1953.
aficionados and clinical/public health perspectives. Int J Clin Prac 25. Vansickel AR, Eissenberg T. Electronic cigarettes: effective nicotine deliv-
2011;65(10):1037–1042. ery after acute administration. Nicotine Tob Res 2013;15(1):267–270.
20. Farsalinos KE, Romagna G, Tsiapras D, Kyrzopoulos S, Voudris V. 26. Farsalinos KE, Spyrou A, Tsimopoulou K, Stefopoulos C, Romagna G,
Evaluating nicotine levels selection and patterns of electronic cigarette use Voudris V. Nicotine absorption from electronic cigarette use: comparison
in a group of “vapers” who had achieved complete substitution of smok- between first and new-generation devices. Sci Rep 2014;4:4133.
ing. Subst Abuse 2013;7:139–146. 27. Hudmon KS, Pomerleau CS, Brigham J, Javitz H, Swan GE. Validity of
21. Bullen C, McRobbie H, Thornley S, Glover M, Lin R, Laugesen M. Effect retrospective assessments of nicotine dependence: a preliminary report.

Downloaded from http://ntr.oxfordjournals.org/ at New York University on May 29, 2015

of an electronic nicotine delivery device (e cigarette) on desire to smoke Addict Behav 2005;30(3):613–617.