E cigarettes an assessment of the so-called gateway effect based on product classification

Visit our science website:
www.fontemscience.com
Since electronic cigarettes (e-cigs) have become popular as alternatives to
conventional cigarettes (CC), with subsequent market growth amongst adult smokers,
there is currently a debate as to whether e-cigs may serve as a gateway to CC smoking
or not. Such fears are related to the possibility that prior use of e-cigs could conceivably
result in CC smoking initiation amongst never smokers. Although the common definition
of a “gateway effect” is based on the concern that current use of a potential low-risk
product could facilitate the use of higher-risk products in the future, there is currently no
agreed method for assessing the so-called “gateway effect” for e-cig products. This
creates a lack of clarity and confusion amongst researchers, politicians, media, vapers
and smokers, which often leads to misleading study interpretations and conclusions
being drawn with a negative impact on policy and regulatory decisions.
To that end, we propose and describe a framework based on product classification to
assess any so-called “gateway effect”: ‘alternative product’, ‘transition product’,
‘substitution product’ or ‘gateway product’.
Each of these four categories corresponds to a different probability of a consumer
switching from a potential low-risk product to a high-risk product, and vice versa, based
on the motives for using them. Using an approach such as dynamic population
modelling, it is possible to classify e-cigs in one of these four product categories and
thereby to assess whether e-cigs are a gateway to or a gateway from CC smoking.
Here we describe this innovative approach.
1. Introduction/context
To evaluate the gateway hypothesis for e-cigs, it is necessary to investigate:
1) the probabilities to switch from e-cigs to cigarettes consumption
2) The motivation of people to vape and/or to smoke
Considering these two key points, four different product groups can be defined (Figure
1): substitution products and transition products that are used by consumers who
are motivated to smoke CCs and alternative products and gateway products that
are used by consumers who are not necessary motivated to smoke CCs.
2. Framework definition
The gateway effect is a concept that requires an accurate specific study design and from which data shall be interpreted carefully. The framework proposed here allows
product classification in one out of the four described categories: ‘alternative product’, ‘transition product’, ‘substitution product’ or ‘gateway product’.
Our next step is to define the states and to set up appropriate questionnaires to estimate the probabilities of transitions according to relevant factors (age, gender,
consumption…). In this way, collected data will be used in our computational model in order to estimate the probabilities Pii, Pid and determine the product category.
5. Conclusion
To allocate a product shown in Figure 1, appropriate surveys and specific analytical
tools such as population modelling are required and need to be considered with and
without the presence of e-cigs.
Definitions of all the possible states of the base and current case (see figure 2) related
to product consumption have to be established; and the transition probabilities from
moving from one state to another have to be estimated according to relevant individual
characteristics (for example: age, gender, time in the state, quantity of products used).
Comprehensive and flexible computational models such Dynamic Population Modelling
(DPM) should be used to model population behaviours using the transition probabilities
estimated above.
Two cases are considered:
• the base-case with three states: NS, “never smoker”; S, “current smoker”; and
FS, “former smoker” and;
• the current case where an e-cig is available on the market.
Additional states include: EU “e-cig user” or DU “dual user”; some states may also
change such as FS to FU “former user” and NS to NU “never user”.
Figure 2 highlights the important transitions (P1, P2, P3 and P4) to consider in order to
allocate a product to a specific category.
3. Framework evaluation
In the state diagrams, the probabilities P1 , P2, P3 and P4 are direct or indirect
probabilities of transition from one state to another. These probabilities can integer
multiple information such as user history (time spent in the different states),
consumption levels, socio-demographic factors (for example: age, education,
gender…).
To classify a product according the different probabilities, an indicator P can be
evaluated as follow:
𝑷 = 𝑷𝒊𝒊 − 𝑷𝒊𝒅
With
Pii = P3 + P4 , the probability to initiate CC smoking indirectly (via an e-cig)
Pid = P1 − P2 , the change of probability to initiate CC smoking directly after the
introduction of e-cig.
Figure 3 shows the product classification according to the probabilities Pii and Pid.
4. Product classification
Figure 1. Product classification based on the switching probabilities.
Figure 2. State diagram with
transition probabilities with and
without e-cig on the market .
Figure 3.
Product classification
according the probabilities
of transition from one
state to another one.
E-CIGARETTES:
AN ASSESSMENT OF THE SO-CALLED “GATEWAY
EFFECT” BASED ON PRODUCT CLASSIFICATION
Thomas Verron1, Xavier Cahours1, Liz Cerson2, Grant O’Connell3 and Stéphane Colard1
1 SEITA-Imperial Brands, 45404 Fleury-les-Aubrais, France; 2 Imperial Tobacco Ltd, 121 Winterstoke Road, Bristol, BS3 2LL, UK; 3 Fontem Ventures B.V., HN Amsterdam 1083, Netherlands
Base-case Current case
CORESTA "Bridging the gaps", Berlin, Germany 9-13 October 2016

E cigarettes an assessment of the so-called gateway effect based on product classification

More Related Content

Similar to E cigarettes an assessment of the so-called gateway effect based on product classification (20)

More from Fontem Ventures (16)

Recently uploaded (20)

E cigarettes an assessment of the so-called gateway effect based on product classification