Preferences for Electronic Nicotine Delivery Systems among Young Adults: Results from an Online Discrete Choice Experiment

Objective: To examine the impact of flavor, device type, and health warning messages on young adult preference for electronic nicotine delivery systems (ENDS), and to provide evidence and data to inform the FDA’s potential regulatory actions on ENDS.

Design: An online discrete choice experiment (DCE) was conducted among a convenience sample of 400 young adults aged 18-24 in May 2016. Each study participant was given 9 choice sets. For each choice set, the participant was asked to choose from two hypothetical ENDS products and not using any ENDS. If the participant had ever used ENDS, s/he was given an additional option which is their most-used ENDS product. In young adults DCE, prices of devices and refill, varying at different levels, were also provided to the participants along with three attributes (flavor, device type, and health warning message) to measure willingness-to-pay (WTP). Participants were also surveyed about their tobacco use history and related behaviors following the experiment. The impact of the attributes on the probability of choosing ENDS was analyzed using conditional logit regressions, controlling for individual socio-demographic characteristics and current smoking status.

Setting and Participants: 400 young adults aged 18-24 were recruited to complete the experiment and a short survey using an online panel. Among the participants, 200 had used ENDS and 200 had never used ENDS at the time of the experiment.

Main Outcomes and Measures: The main outcome of this study is the probability of a participant choosing ENDS. We hypothesized that young adults will be more likely to use ENDS if they have menthol or fruit/sweets/beverage flavors; if they are more-advanced ENDS device types such as E-go/Mods/APVs; and/or if they do not contain health warning messages or strong messages.

Results:

Among young adult ENDS ever users, fruit/sweets/beverage and menthol flavors significantly increase the probability of choosing ENDS (p<0.001). Higher prices of refills and devices decrease the probability of choosing ENDS (p<0.05 for refill prices and p<0.001 for device prices), and the effect of refill prices is stronger than that of device prices. These results also suggest a marginal WTP of fruit/sweets/beverage flavors of $13 and a marginal WTP of menthol flavor of $6.42. Device type and warning messages did not significantly influence ENDS choices among this age group. Results suggest that none of the attributes matter to young adult ENDS never users.

Conclusions and Relevance: Higher prices of devices and refills reduce the probability of choosing ENDS among young adults who have tried ENDS. Young adult ever users also prefer ENDS with flavors (i.e. menthol and fruit/sweets/beverage flavors), particularly fruit/sweets/beverage flavors. Restricting flavors in ENDS may reduce the probability of choosing ENDS among young adult ever-users.