Liking of e-liquid flavors: correlation between smelling and vaping

It is well known that smoking is associated with severe health effects.
Recently, the e-cigarette was introduced. Whereas smoking burns tobacco, the
use of an e-cigarette (vaping) only heats e-liquids. E-liquids consist of a
base liquid with flavorings and potentially nicotine. E-liquids are available
in a high variety of flavors, which increase sensory appeal. Therefore, usage
of the e-cigarette has exponentially increased over the past couple of years,
and governments aim to regulate the availability of e-liquid flavors. Within
Europe, no regulation on the availability of e-liquid flavors currently exists.
In order to advice the government on how to regulate e-liquids flavors,
research is required. On the one hand, certain flavors may help currently
addicted adult smokers switch to potentially less harmful forms of
nicotine-containing tobacco products. At the same time, e-cigarettes are the
most commonly used tobacco product among middle and high school students, and
flavors are identified as one of the top three reasons for use. Given these
findings, governments need to be wary of the role flavors play in attracting
youth to initiate on any tobacco product that could lead to regular use *
something that clearly should be avoided. No child should use any tobacco
products, including e-cigarettes.

Regulation could focus on liking (and disliking) of particular e-liquid flavors
and/or flavor categories among these different consumer groups (i.e.,
adults/adolescents; smokers/non-smokers). On behalf of the Dutch Ministry of
Health, Welfare, and Sports, Wageningen University and RIVM collaborate in
order to investigate sensory characteristics of e-liquids flavors within these
target groups. In this way, we aim to obtain insight in attractiveness and
preferences of different e-liquid flavors and addictive behavior - also within
the brain. The Smell-e study is the first step in this 4-years research
project. Eventually, we will advice the government in how to best regulate
e-liquid flavors, contributing to the overall aim for a 'smoke-free
generation'.

Research on e-liquid flavors could be performed through smelling or tasting
(*vaping*) experiments. The main objective of the Smell-e study is to determine
if there is a correlation between liking of e-liquid flavors through smelling
and liking of e-liquid flavors through vaping, for smokers as well as for
non-smokers. If we find a correlation between liking by smelling and liking by
vaping, we may be able to assess liking through vaping in future research by
performing smelling experiments only. This removes potential health
implications, and will save time, effort and money. In addition, performing
smelling experiments allows to include youth as well (<18 years old). Results
in flavor liking among this study population will be of huge importance to
policy makers worldwide in developing their regulations on e-cigarette flavors
in order to prevent tobacco use among non-smoking youth.

The Smell-e study consists of an experimental within-research subjects design.,
in which the smell and taste of 25 e-liquids will be evaluated by smelling and
vaping. The study consists of 4 testsessions (two smelling and two vaping
sessions). Test sessions will take place in sensory booths of the Restaurant of
the Future at Wageningen University. The order of sessions A * D will be
counterbalanced across research subjects. Sessions will be scheduled in
accordance with panelists. Minimal time between two sessions is 1 week.

The screening session (30 minutes) will include olfactory screening using the
Sniffin* Sticks identification test. Research subjects with normal olfactory
function according to this test will be included in the study. Research
subjects will subsequently be trained in using an e-cigarette by vaping an
unflavored e-liquid (maximal 5 puffs), and in using a visual analogue scale
(VAS) by scoring liking of the unflavored e-liquid. This will avoid vaping
results to be influenced by incorrect usage of the e-cigarette or VAS.

The test sessions consist of two smelling and two vaping sessions (each 60
minutes).
For the smelling sessions, 10 drops of each e-liquid will be put into a glass
vial, which will be covered in tinfoil to avoid visual cues and labelled with a
random three-digit code. In one smelling session, research subjects will smell
12 of the 25 e-liquid flavors. The other 13 e-liquid flavors will be assessed
in the other smelling session. During the smelling session, research subjects
will first assess each of the 12 or 13 samples in randomized order on liking
(100mm VAS). Subsequently, they will assess each of the same samples in
randomized order on intensity, familiarity, and irritation (100mm VAS). In this
way, research subjects do not have to continuously switch between hedonic and
analytical assessment for each sample. Research subject will smell each sample
2 times. A one-minute break will be incorporated between each sample to
neutralize the research subjects* smell and prevent adaptation.
For the vaping sessions, two sets of 25 e-cigarettes will be filled with each
of the 25 e-liquids. The e-cigarette cartridge (which contains the e-liquid)
will be covered in tinfoil to avoid visual cues and labelled with a random
three-digit code. In one vaping session, research subjects will evaluate 12 of
the 25 e-liquid flavors. The other 13 e-liquid flavors will be assessed in the
other vaping session. During the vaping session, research subjects will first
assess each of the 12 or 13 samples in randomized order on liking (100mm VAS).
Subsequently, they will assess each of the same samples in randomized order on
intensity, familiarity, and irritation (100mm VAS). Each research subject will
use 2 puffs for each sample. Research subjects themselves decide whether to
inhale the vape over their lungs or whether to keep the vape only in their
mouth before exhaling; as long as they inhale each sample in the same way.

Research subjects will be exposed to the odor of 25 e-liquids (divided over two
smelling sessions) and the vapor of the same 25 e-liquids (divided over two
vaping sessions).

The burden for research subjects includes 6 visits to the unversity for the
information meeting, screening, and four test ssessions respectively. The study
will take approximately 5.5 hours in total. During the information meeting,
research subjects fill out the screening questionnaire. During the screening
session, research subjects smell 16 odors (Sniffin' Sticks). During the test
sessions, research subjects evaluate each sample on four aspects (liking,
intensity, familiarity, and irritation) while they smell the sample twice (25
samples divided over 2 smelling sessions) and they take two puffs of each
sample (25 samples divided over two vaping sessions).

The health risk associated with smelling is negligible. Potential risks
associated with vaping can be divided into three parts: addictiveness,
attractiveness, and toxicity.
The risks on addictiveness is not applicable, as we only use e-liquids without
nicotine (this will be confirmed using chemical-analytical research). The risk
of attractiveness of e-cigarette use will be minimized by using the most
straightforward and least attractive e-cigarette model (far from the look of a
regular tobacco cigarette and not the most popular e-cigarette model),
selecting e-liquids with a low amount of vegetable glycerin (in order to mimize
the amount of vapor clouds that is produced), and by performing the vaping
experiments in laboratory setting (while vaping might be more associated with a
social setting).

The vaping part of this study will be associated with a very low health risk,
because research subjects will take a low number of puffs (50 in total, divided
over two sessions) and because the e-liquids do not contain nicotine. The RIVM
has published a report on health effects of e-cigarette exposure, of which we
can conclude that potential health risks of the Smell-e study are very low.
Only with a much higher exposure to e-cigarette vapor than the exposure of this
study, users might be at risk of irritation or harm to airways and
palpitations. In any way, these effects are less sever than the health effects
that are associated with smoking of regular tobacco cigarettes (such as lung
cancer, heart attack, and stroke).

Potential health effects will be elaborated during the information meeting. A
more detailed risk analsyis is provided in chapter 13 of our reseach protocol.