Effects of E-cigarette Liquid Flavors and Modified Risk Messages on Measures of Abuse Liability

Objective: While adult smoking prevalence stagnates in the United States, electronic cigarettes (ECs) are gaining popularity. ECs are currently unregulated at the federal level. In April 2014, the US Food and Drug Administration’s (FDA) Center on Tobacco Products proposed that ECs be subject to FDA regulation. The FDA’s proposal, among other things, highlighted interest in evidence to inform the regulation of EC liquid flavors and requiring approval for modified risk descriptors suggesting lower harm potential. These two studies examine how abuse liability of ECs, a measure of consumer demand used in the psychology and behavioral economics literature, varies by EC liquid flavors and the presence of modified risk messages as well as compared to own brand combustible tobacco cigarette (CTC) abuse liability.

Study design: Data are gathered from a sample of regular CTC smokers, naïve to ECs.   Using two within-subjects designs we assess abuse liability across EC liquid flavors (tobacco vs. menthol; unflavored vs. cherry) and message (reduced harm vs. no message; reduced exposure to carcinogens vs. no message) conditions. At baseline, own brand CTC abuse liability was assessed, and participants sampled EC flavors. Abuse liability measures included a revealed preference method, the multiple-choice procedure (MCP) that randomly reinforces participants’ choices, and a stated preference method, the cigarette purchasing task (CPT).  In each condition, participants completed five MCP trials and the CPT. Linear regressions compared the crossover points at which participants chose money over EC or own brand puffs after adjusting for own brand CTC flavor preference (menthol, tobacco). Standard errors were adjusted for repeated measures.

Sample: We have accrued 28 of the 88 subjects we plan to enroll across both studies. Our sample is predominantly African American (91%), smokes menthol own brand CTC (82%), female (55%), single (82%) and, on average, is 46 years old.

Results: After adjustment, in the first study (tobacco vs. menthol liquid flavor; reduced harm vs. no message), the crossover point for own brand CTC was $1.89. For all EC conditions, crossover points were significantly lower than own brand CTC: these ranged from -$0.71 (95% CI -$1.31, -$0.10; tobacco flavor/reduced harm message) to -$0.81 (95% CI -$1.48, -$0.13; menthol flavor/no message). There was a trend towards higher crossover points for conditions in which participants were told the EC was less harmful than CTCs. In the second study (unflavored vs. cherry liquid flavor; reduced exposure to carcinogens vs. no message), similar significant patterns of effects in crossover points were observed, although the magnitudes were larger.  Results across the two measures of abuse liability, the MCP and CPT, are broadly consistent in the first but not the second study.

Conclusions: This study provides preliminary evidence that flavor availability and modified risk messages associated with ECs may affect abuse liability. Importantly, modified risk messages appear positively associated with abuse liability. If replicated in larger samples, these results suggest how FDA regulations concerning EC flavors and risk messages will likely impact CTC smokers’ uptake and use of EC products.