This research forms part of our assessment framework which we use to test our potentially reduced-risk products.
User product interaction assessment involves both objective (e.g. puffing topography) and subjective (e.g. risk perception) measurements.
Measuring consumers’ puffing topography provides the distribution of use parameters like puff volume, puff duration and puff interval, which can then be used to establish the machine puffing regime for various laboratory testing of potentially reduced-risk products.
In addition to assessing the chemical emissions of potentially reduced-risk products, it is important to understand whether consumers are using these products in a manner that supports the potential reduction in overall toxicant exposure compared with smoking cigarettes. Consumer use behaviour studies therefore play a key part in the proposed scientific framework for evaluating the reduced risk potential of these new products.
First, we measure average daily consumption of products through market surveys, asking consumers to self-report their usage (e.g. the amount of e-liquid they use). This works well for products already on the market. We also collect consumers’ puffing behaviour (or topography) to measure parameters such as puff volume, frequency, profile, puff number, as well as estimating mouth-level exposure (MLE) to components such as nicotine, vegetable glycerine (VG), propylene glycol (PG) and NFDPM (nicotine-free dry particulate matter) in central location tests. Occasionally, these test are preceded by a home use test, to let the user get familiar with the product (pre-market) and also collect their average daily consumption. Most of these tests also conduct sensory evaluation to understand the experience of taste, throat hit and other sensory characteristics of interest.
Measuring MLE and puffing topography across a range of product categories will provide additional, useful information that will help to enable the placement of tobacco and nicotine products on a risk continuum and aid the individual risk assessment of these products.
In the past, research into the MLE of conventional cigarettes has been widely based around the relationship between the materials trapped within the filter and user’s mouth level exposure to smoke1
Using this relationship, researchers can collect used cigarette filters from participants and analyse them to estimate MLE for a particular product. We continue to use this method for cigarettes used as reference products in our potentially reduced-risk product studies. There are fundamental differences in the function of the mouth-end sections between tobacco heating products (THPs) and cigarettes, meaning additional research is necessary before MLE from THPs can be estimated using this approach. Furthermore, as vapour products do not contain filters like cigarettes, the part-filter method for estimating MLE is inappropriate.
More recently, we measure puffing topography and MLE in real-time using the SA7, a bespoke smoking analyser that we originally designed to help us understand more about smoking cigarettes but have recently modified for use with both vapour and THPs. The SA7 consists of three main parts: a product holder (which the product is vaped or smoked through), a Data Acquisition and Transmission (DAT) unit and a computer.
The flow of vapour or aerosol to the consumer’s mouth is measured by recording the pressure on either side of a stainless-steel cavity inside the holder and converting the change in pressure to indicate volume of aerosol or vapour, the time taken to draw it into the mouth, and the length of time between puffs. Simultaneously, the optical density of the smoke or vapour passing through the holder is measured using an LED and a detector: when no aerosol is present, 100% of the light from the LED reaches the detector, but when aerosol or vapour is present, some light is blocked and less reaches the detector. These data can be used to estimate the MLE of constituents in the aerosols, which in turn allows us to estimate what consumers are being exposed to.
Product satisfaction helps us to determine how our products are perceived and used by consumers; this includes sensorial attributes like taste, as well as consumption patterns. Product satisfaction is assessed in observational studies where consumer behaviour can be discreetly witnessed first-hand, this can be with trained panellists, with semi-naïve panellists, or with recruited participants who have no training or prior knowledge of such studies/products. Studies can be designed to include both passive observations of behaviours and questionnaires, such as the Modified Cigarette Evaluation Questionnaire.2
In conjunction with objectively measuring consumer behaviour, we can aim to understand the level of acceptability towards product usage from a sensorial standpoint. For example, we can measure the amount of draw effort the consumer needs to apply to the product to obtain enough aerosol in their mouth – however, it is also valuable to understand whether this level is acceptable to the consumer, which can in turn inform product development. Other factors that we may measure include mouth fill, throat hit, taste likeability, aerosol delivery and irritation.
Our population studies programme has been designed to assess the impact of our potentially reduced-risk products on public health and claims messaging. This helps us ensure consumers are informed about our potentially reduced-risk products and monitor the health and behavioural effects of them after they have been made available. Outcomes from these studies will support help evidence-based regulation of these new categories of products.
Population studies involve monitoring both user and non-user behaviour over a long period of time to understand the complex transitions in use behaviour (e.g. dual and poly use). The US Food and Drug Administration (FDA) requires population studies that measure consumer perception, comprehension of ‘claims messaging’ and use behaviour in both pre and post-market scenarios to understand the potential public health benefits of our new products on individuals and the population. This data can be used to help model the impact on population health. Before conducting our studies, where applicable, we seek to gain Independent Ethics approval.
Measuring consumer and non-user comprehension of product claims and messaging can give insights into attitudes or behaviours towards potentially reduced-risk products, and the potential for uptake and switching. For instance, clear comprehension of product messaging should persuade current smokers to transition towards potentially reduced-risk products but should dissuade or be unappealing to never and former smokers. Correct comprehension of product messaging and claims will affect attitudes towards different products, which may also affect an individual’s risk perception of using the product. As with comprehension, the correct risk perception will also have an impact on product uptake and switching.
Likelihood of uptake is a type of market research that assesses consumers’ interest in and willingness to try potentially reduced-risk products. Consumers are shown a stimulus, which consists of publicly available information explaining what the product is, how it works and any claims or messaging. After viewing the stimulus, consumers are asked how likely they will purchase and try the product for experimental purposes. This enables us to find out how our potentially reduced-risk products would be received in a particular market.
Quality of life measures mental and physical wellbeing. For example, an individual might be good in physical wellbeing but poor in emotional wellbeing. Health status assesses the 12-month prevalence of smoking-related short-term health effects, for example, teeth staining. Changes, more specifically improvements, in quality of life and smoking-related health status can be correlated with consumer switching patterns away from cigarettes towards potentially reduced-risk products.
Risk perception measures an individual’s perceived level of general risk of developing smoking-related diseases by using various tobacco or nicotine products. This can be asked for different product categories and/or after seeing various information on the product, such as images, descriptions of use, product messaging or claims.
Our population studies programme covers both active and passive post-market surveillance. Active post-market surveillance is conducted using cross-sectional surveys and may also include longitudinal cohort studies. Currently, we are not planning to conduct additional post-marketing studies (e.g. health outcome studies, case-control studies, or studies in disease-specific populations) as a part of the population studies programme. However, these types of studies will be considered if there is a further need to evaluate potentially reduced-risk products based on the outcomes from the current programme. Passive post-market surveillance usually takes the form of safety tracking, which can be conducted within cross-sectional surveys.
We use safety tracking to collect and record self-reported unanticipated and adverse events related to potentially reduced-risk products use, on consumer use behaviour, perception and health after potentially reduced-risk products introduction to a market.
Safety tracking is an FDA requirement, who define an adverse event as 'any health-related event associated with the use of a tobacco product in humans that is adverse or unfavourable, whether or not it is considered tobacco-product related'.
Our consumers are informed that they should contact us via our toll-free careline if they feel unwell during the use of one of our products. Any adverse event that is reported via our careline is recorded, reviewed and tracked by our internal Quality team.
After we launch a product, we want to know who is buying our product, why they are buying it, and how they are using it. Behaviours are dynamic and are likely to change over time. Post-market surveillance assesses real-life use-behaviour over time in a real-world setting. It is an FDA requirement to conduct post-market surveillance and report the outcomes at regular intervals to evaluate the real-world effectiveness of the introductions of potentially reduced-risk products.
We are conducting our first post-market surveillance study, which is in Japan. This study assesses the impact of the introduction of our tobacco heating product (THP) glo by administering a point-prevalence cross-sectional survey.
The objective of cross-sectional surveys is to monitor perceptions and in-market use-behaviours to evaluate the benefits and risks of potentially reduced-risk products on the health of the population as a whole.
Post-market surveillance studies follow the principles of Good Epidemiological Practice and the study protocol is reviewed and approved by an Independent Ethics Committee.
The above study aims to find out about the consumption of tobacco and nicotine products in the general Japanese population, including initiation of use of the tobacco product among never users and former users, rates that current tobacco users switch to the tobacco product, and patterns of use of the tobacco product by current tobacco users. In addition, we want to compare the risk perceptions of different tobacco and no tobacco products and assess perceived quality of life and health status.
Assessing risk perception, quality of life and health status takes advantage of the non-smoking cohort of participants to gain insight into the overall impact of our products in a market. For instance, comprehension and the perceived risk of our new potentially reduced-risk products are measured and collected by all consumer types, including never users. Using epidemiology, the impact of the introduction of our potentially reduced-risk products on the health of the population can be assessed (as opposed to assessing the inherent risk to an individual).
The data could be used to inform a System Dynamics Population Model, which would allow us to estimate the public health benefit of introducing a THP to the Japanese population.
Pauly et al. (2009); Shepperd et al. (2006); St. Charles et al. (2009); Watson et al. (2004).
Cappelleri, J., Bushmakin, A., Baker, C., Merikle, E., Olufade, A., & Gilbert, D. (2007).
Confirmatory factor analyses and reliability of the modified cigarette evaluation questionnaire. Addictive Behaviors, 32(5), 912-923).