Switching completely from combusted cigarettes to smoke-free tobacco products, such as HTPs, presents a public health opportunity to reduce premature mortality and disease burden associated with continued smoking among AS unwilling or unable to stop consuming tobacco products (5, 17). Flavor is known to influence uptake of smoke-free products and subsequent switching behavior among AS (21, 31) and is also a topic of regulatory concern in the US (32). The lack of data—overall and considering the role of flavor—on HTP uptake and switching behavior among AS in the US is not only a gap in the literature but leaves policy makers without empirical evidence on which to base their decisions. To address this gap, we conducted a six-week actual use study to characterize switching behavior and changes in smoking behavior among AS when provided with free choice of menthol and non-menthol HTPs for ad libitum use in their natural environment. To our knowledge, this is the first study to report differences in HTP switching behavior by AS’ preference for menthol versus non-menthol combusted cigarettes and HTP flavor variety.
Overall, AS acclimated to and used HTPs over the six-week study. Using an ITT approach, 21.1% of participants completely switched from combusted cigarettes to the research product at Week 6, lower than values reported by Caponnetto et al. (25).The relatively lower observed switching among the ITT sample in the current study may be due to methodological differences between studies, such as randomized trial versus observational actual use study designs, Caponnetto et al.’s use of motivational counseling, or conducting a study in Italy where the sale of menthol tobacco products is banned versus in the US where the sale of such products is widely possible.
Among our analytic sample overall, 40.2% switched at Week 6. Here, switching behavior varied between Menthol Cigarette Preference Groups, with adults who preferred smoking menthol cigarettes significantly more likely to switch than those who preferred non-menthol cigarettes. This finding is consistent with a recent ENDS switching study reporting greater switching among AS who preferred smoking menthol versus non-menthol cigarettes (31), extending these earlier findings to HTP.
Interestingly, the probability of switching increased with greater proportional use of menthol versus non-menthol HTPs. This outcome is consistent with recent cross-sectional analysis of the Tobacco Use Supplement to the Current Population Survey reporting greater switching among AS when using menthol versus non-menthol ENDS (33). We also observed that non-menthol cigarette preferring participants—but not menthol preferring—were more likely to switch when they used proportionally more menthol than non-menthol HTPs. Similar results have been found among AS in the US switching to ENDS over a one-year period (31), yet another point of convergence in switching behavior across smoke-free products. As discussed in greater detail below, those who preferred smoking non-menthol cigarettes used menthol HTPs (either exclusively or interchangeably with non-menthol HTPs), but AS who preferred menthol cigarettes tended to use the HTP flavor congruent with their cigarette preference. The overall higher switching among participants who preferred smoking menthol cigarettes combined with low use of non-menthol HTP within this group likely explains why the probability of switching is relatively consistent, no matter how much menthol HTP was used. Our results suggest that the combined availability of both menthol and non-menthol HTPs may offer an increased tobacco harm reduction opportunity over solely non-menthol HTPs by facilitating switching away from combusted tobacco among AS in the US regardless of their preference for smoking menthol or non-menthol cigarettes.
Use of menthol HTP sticks was not exclusive to participants who preferred smoking menthol cigarettes. Menthol HTP varieties had greater reach than non-menthol, demonstrated by their uptake and use by AS who prefer smoking menthol or non-menthol cigarettes. This finding contrasts with switching studies, subjective sensory response research, and behavioral economic research of other inhaled tobacco products (i.e., combusted cigarettes and electronic cigarettes) where preference for menthol cigarettes is typically congruent with menthol preference for other tobacco products (31, 34, 35).
Only a small proportion of participants who preferred menthol cigarettes chose to exclusively use non-menthol HTP sticks, suggesting limited substitutability of non-menthol HTPs among this subgroup of AS. In the case of a possible combustible menthol cigarette ban in the US, the availability of reduced harm smoke-free products in flavors appealing to AS who prefer menthol cigarettes—and are not able or willing to stop consuming tobacco products—would likely reduce the possibility of these individuals continuing to use combusted cigarettes that are still commercially available (i.e., non-menthol) or seeking out combustible menthol cigarettes via illicit markets (36, 37). Given that the majority of combusted menthol cigarettes are consumed by racial, ethnic, and sexual preference minorities (22, 38, 39), availability of reduced-harm nicotine and tobacco products that appeal to these individuals presents an opportunity for equitable tobacco harm reduction for those not able or willing to stop consuming nicotine.
Among those who did not switch and continued smoking cigarettes (n = 368), the majority (57.9%) reduced their cigarette consumption by 50–99%, with relatively smaller proportions either continuing to smoke the same amount (3.0%) or increasing consumption (8.4%) while using HTP. Those who increased their cigarette consumption did so, on average, by 3.8 CPD. Despite this small proportion who increased their cigarette consumption, participants who continued smoking had a meaningful reduction in cigarette consumption overall—at just under half a pack per day—of 9.6 (± 13.4) CPD, a reduction of 47.1% on average. Reducing cigarette consumption consequentially reduces HPHC exposure and has been associated with reductions in tobacco-related negative health outcomes, but not at the same magnitude as stopping use of all tobacco products (40–43). Additionally, dual use incorporates a wide range of use behaviors, with varying proportions of use across the two or more tobacco products used by an individual (44). Dual use behavior has often been shown to be a transitional period when switching from combusted cigarettes to ENDS and HTP (25, 45–48). It is possible that some of the dual use observed in the current study may have occurred during such a transitional period among those needing longer than six weeks to completely switch away from combusted cigarettes.
Results discussed above need to be considered in context of the study limitations. Data were collected within an actual use study design, without randomization or a control condition. Our results are therefore considered observational, not causal. This study was intended to assess how adults who smoke would use the research product under near real-world conditions and when they can freely choose the products they wish to use. Use of randomized controlled trial methodology to assign participants to flavor varieties or continued cigarette smoking would not have been within scope of the project. Further, our study design enabled observations that may not have been contemplated a priori in a randomized controlled trial, such as participants’ use of both menthol and non-menthol HTPs and preference for menthol HTPs among AS who preferred smoking non-menthol cigarettes. Participants were provided with free research product for their ad libitum use during the study, which may have inflated switching or smoking reduction relative to a scenario in which they had to pay for the product (i.e., free product effect). Finally, switching and smoking reduction outcomes relied on self-report and did not include biochemical verification. Participants were not incentivized to switch or reduce cigarette consumption, reducing potential for bias in these self-reported outcomes. Further, there were no a priori reasons to expect self-reported tobacco product use behaviors to vary across the independent variables of interest in this study (i.e., preference for menthol or non-menthol cigarettes and use of menthol or non-menthol HTP). Although remote biochemical verification of smoking status has become logistically possible in recent years (e.g., 49, 50–52), its use in the current actual use study would have further reduced ecological validity and placed unnecessary burden upon participants that may outweigh any benefit of such measurement (53, 54).