The effect of framing incentives as either losses or gains with contingency management for smoking cessation

Abstract

Cumulative prospect theory predicts that losses motivate behavior more than equal gains. Contingency management procedures effectively reduce drug use by placing incentives in direct competition with the drug taking behavior. Therefore, framing incentives as losses, rather than gains should decrease drug use to a greater extent, given equivalent incentives. We examined whether contingent vouchers described as either losses or gains differentially affected smoking abstinence rates. Over 5 consecutive days, participants could either gain $75 per day for verified abstinence or lose $75 per day (initial endowment=$375) for continuing to smoke. As a result, loss-framed participants were more likely to achieve at least one day of abstinence. There was a trend towards loss-framed participants reducing the amount smoked more than gain-framed participants. However, participants in the gain-framed group were more likely to maintain abstinence, once initiated. The results partially support cumulative prospect theory and suggest additional ways to initiate behavior change using incentives, outside of using larger magnitude incentives in contingency management procedures.

Figure 1

Event records for loss- and gain-framed participants, where participants could either gain or lose $75 per day. Visit number is shown on the abscissa. The black areas represent visits with a breath CO < 3 ppm, grey areas represent visits with a breath CO ≥3 ppm, and white areas represent missed visits. Each number represents the obtained breath CO sample in ppm across the intake, baseline, incentive and follow-up phases of the experiment.

Figure 2

Mean incentive phase breath CO samples as a function of the percent of the lowest baseline breath CO sample for each participant. Participants are ranked based on their ability to decrease breath CO values from baseline to the incentive phase of the experiment. Thus, lower ranks are indicative of a lower mean incentive phase breath CO level, relative to their baseline breath CO level. Each point represents one incentive phase breath CO sample. Open circles and open triangles represent loss- and gain-framed participants, respectively.