posted 01/05/2020
publication Show more https://doi.org/10.1016/j.drugpo.2020.102740
Jonathan P. Caulkinsa Bryce Pardob Beau Kilmerb
a
Carngie Mellon University, Heinz College, 5000 Forbes Ave., Pittsburgh, PA 15213, USA
b
RAND Drug Policy Research Center, 1776 Main St., Santa Monica, CA 90401, USA
https://doi.org/10.1016/j.drugpo.2020.102740Get rights and content
Abstract
Background
Drug use is often measured in terms of prevalence, meaning the number of people who used any amount in the last month or year, but measuring the quantity consumed is critical for making informed regulatory decisions and estimating the effects of policy changes. Quantity is the product of frequency (e.g., number of use days in the last month) and intensity (amount consumed per use day). Presently, there is imperfect understanding of the extent to which more frequent users also consume more intensively.
Methods and data
We examine cannabis flower consumption reported in three similar online surveys fielded in times and places where cannabis was and was not legal. These convenience samples returned enough valid responses (n = 2,618) to examine consumption across different frequencies of use via analyses of measures of central tendency, data visualizations, and multivariate regressions. Additional calculations incorporate data from the National Survey on Drug Use and Health.
Findings
Respondents who reported using daily (i.e., 30 days in the past month) consumed almost twice as much per day of use on average as did those reporting less than daily. We find only modest increases in intensity among those using less than daily, but then a substantial increase (p< 0.001) for those who use daily. Most respondents report that on heavy or light use days their consumption differs from a typical day of use by a factor of 2 or more, but only about 25% of days were described as heavy or light. We estimate those using cannabis 21+ days a month account for 80% of consumption vs. 71% of the days of use.
Discussion
Daily cannabis users consume more intensively than others, including near-daily users. When possible, survey questions should move beyond the presence or absence of use and number of days used.