Further validation of a marijuana purchase task

Abstract

Background: A valid measure of the relative economic value of marijuana is needed to characterize individual variation in the drug's reinforcing value and inform evolving national marijuana policy. Relative drug value (demand) can be measured via purchase tasks, and demand for alcohol and cigarettes has been associated with craving, dependence, and treatment response. This study examined marijuana demand with a marijuana purchase task (MPT).

Methods: The 22-item self-report MPT was administered to 99 frequent marijuana users (37.4% female, 71.5% marijuana use days, 15.2% cannabis dependent).

Results: Pearson correlations indicated a negative relationship between intensity (free consumption) and age of initiation of regular use (r=-0.34, p<0.001), and positive associations with use days (r=0.26, p<0.05) and subjective craving (r=0.43, p<0.001). Omax (maximum expenditure) was positively associated with use days (r=0.29, p<0.01) and subjective craving (r=0.27, p<0.01). Income was not associated with demand. An exponential demand model provided an excellent fit to the data across users (R(2)=0.99). Group comparisons based on presence or absence of DSM-IV cannabis dependence symptoms revealed that users with any dependence symptoms showed significantly higher intensity of demand and more inelastic demand, reflecting greater insensitivity to price increases.

Conclusions: These results provide support for construct validity of the MPT, indicating its sensitivity to marijuana demand as a function of increasing cost, and its ability to differentiate between users with and without dependence symptoms. The MPT may denote abuse liability and is a valuable addition to the behavioral economic literature. Potential applications to marijuana pricing and tax policy are discussed.

Keywords: Behavioral economics; Cannabis dependence; Demand curve; Marijuana; Purchase task.

Figure 1

Demand curve for consumption of marijuana hits, with the x-axis providing price in dollars ($) and the y-axis providing self-reported consumption in marijuana hits. Conventional log coordinates are used on the x-axis to accommodate large inter-price intervals, with zero values replaced by trivial nonzero values (0.1).

Figure 2

Expenditure curve for purchase of marijuana hits, with the x-axis providing price in dollars ($) and the y-axis providing expenditure in dollars ($). Values are presented in actual units rather than conventional logarithmic units for interpretational clarity.

Figure 3

Demand curves for No Dependence Symptoms (0 DSM-IV symptoms) and Any Dependence Symptoms (1+ DSM-IV symptoms) Groups. Conventional log coordinates are used on the x-axis to accommodate large inter-price intervals, with zero values replaced by trivial nonzero values (0.1).