Testing Cessation Messages for Cigarette Package Inserts: Findings from a Best/Worst Discrete Choice Experiment

Abstract

This study assessed smokers' responses to different smoking cessation topics and imagery for cigarette package inserts. Adult smokers from Canada (n = 1000) participated in three discrete choice experiments (DCEs): DCE 1 assessed five cessation benefit topics and five imagery types; DCE 2 assessed five messages with tips to improve cessation success and five imagery types; DCE 3 assessed four reproductive health benefits of cessation topics and four imagery types. In each DCE, participants evaluated four or five sets of four inserts, selecting the most and least motivating (DCEs 1 & 3) or helpful (DCE 2) for quitting. Linear mixed models regressed choices on insert and smoker characteristics. For DCE 1, the most motivating messages involved novel disease topics and imagery of younger women. For DCE 2, topics of social support, stress reduction and nicotine replacement therapy were selected as most helpful, with no differences by imagery type. For DCE 3, imagery influenced choices more than topic, with imagery of a family or a mom and baby selected as most motivating. Statistically significant interactions for all three experiments indicated that the influence of imagery type on choices depended on the message topic. Messages to promote smoking cessation through cigarette pack inserts should consider specific combinations of message topic and imagery.

Keywords: health communication; health policy; smoking cessation; tobacco control.

Figure 1

Topic and imagery for cessation benefit messages in DCE 1.

Figure 2

Topic and imagery for quitting tips messages in DCE 2.

Figure 3

Topics and imagery for reproductive health benefit messages in DCE 3.

Figure 4

Example choice set for DCE 1.

Figure 5

Relative importance * of attributes on choice of insert (* Relative importance was calculated using a utility range (i.e., the difference between each characteristic’s highest and lowest estimated part-worth utility or estimated effect on choices), divided by the sum of all the characteristics’ utility ranges for a given outcome. DCE 3 was conducted only at baseline).