In November 2012, voters in Washington legalized nonmedical (retail) cannabis for people aged 21 and older. Markets opened in July 2014. The effect of this change on cannabis use among youths is of public health concern.
Cerdá et al1 analyzed data from the nationally representative Monitoring the Future survey (MTF) and used difference-in-differences methods to compare cannabis use prevalence trends among youths in Washington with use in states without legalization of recreational marijuana. Because the MTF is not designed to provide state-representative estimates, the article generated covariate-adjusted modeled prevalence estimates for each state. The article suggested complex association between legalization and cannabis use among youths: increases in prevalence among Washington 8th and 10th graders, but not among 12th graders, relative to use in states without legalization of recreational marijuana.
The authors noted that, “the sample design may lead to discrepancies between MTF results and those found in other large-scale surveillance efforts.”1(p148) The purpose of the present study was to assess whether trends in cannabis use prevalence among youths from Washington’s state-based youth survey are consistent with findings from the MTF.
The Washington Healthy Youth Survey (HYS)2 is an anonymous, school-based survey of 8th, 10th, and 12th graders and the state’s primary source of information about health behavior among youths. The HYS has been implemented in the fall of even-numbered years since 2002, using a simple random sample of public schools to generate a state-representative sample. Response rates (incorporating school and student response) in 2016 were 80% for 8th grade, 69% for 10th grade, and 49% for 12th grade. The study was approved by the Washington State Institutional Review Board, whose general policy waives informed patient consent when data are deidentified.
We generated covariate-adjusted prevalence estimates, modeling as closely as possible to Cerdá et al.1 Prevalence was based on modeled estimates (ie, SUDAAN predMARG [RTI International] postestimation command). Because the postlegalization periods are not identical, we present HYS data from both 2014 alone and 2014-2016 combined (MTF reported 2013-2015). Significance was established at P < .05 with unpaired, 2-tailed testing. Analysis was conducted using Stata, version 14.2 (StataCorp).
More schools and students are captured in the HYS than MTF (Table). The MTF included fewer low–socioeconomic status and nonwhite youth in the prelegalization vs postlegalization period.
Estimates from the MTF show statistically nonsignificant change in the prevalence of cannabis use for 8th graders (from 6.2% [95% CI, 4.4%-8.7%] to 8.2% [95% CI, 6.3%-10.7%]; P = .16), and a significant increase for 10th graders (from 16.2% [95% CI, 14.0%-18.6%] to 20.3% [95% CI, 16.9%-24.1%]; P = .02). In contrast, the HYS shows statistically significant declines in prevalence from 2010-2012 to 2014-2016 among both 8th graders (from 9.8% [95% CI, 9.1%-10.5%] to 7.3% [95% CI, 6.6%-8.0%]; P < .001) and 10th graders (from 19.8% [95% CI, 18.6%-21.0%] to 17.8% [95% CI, 16.7%-18.9%]; P = .01). Neither MTF nor HYS analysis showed changes among 12th graders (Figure). Findings from HYS comparisons to 2014 alone were of less magnitude but similar direction.
In contrast to Cerdá et al,1 Washington’s HYS data suggest that cannabis use among youths declined after legalization among 8th and 10th graders. The main difference is among 10th graders: the MTF suggests a statistically significant increase while HYS suggests a decrease.
These surveys have different purposes: the HYS provides results generalizable to youths in public schools statewide, while the MTF is designed to provide national and US regional (not state-specific) estimates. Hence, the MTF sample may be more influenced by unmeasured characteristics of Washington youths, especially if some subpopulations disproportionately captured are differently affected by legalization. For example, many Washington cities and counties have banned or restricted retail sales following state legalization,3 and differential exposure to local policy contexts may partially explain the varied patterns between the samples. Furthermore, there are differences in sampling error: the smaller MTF sample resulted in larger 95% CIs overlapping between prelegalization and postlegalization periods. In addition, the lack of an HYS comparison group limits the ability to make inferences about specific effects of cannabis legalization.
It is too soon to know the long-term influence that cannabis legalization will have on the prevalence of its use by youths. Further studies are needed with representative state samples, including subgroups; information about patterns of consumption rather than just prevalence; and attention to local implementation.
Accepted for Publication: September 27, 2018.
Corresponding Author: Julia A. Dilley, PhD, MES, Multnomah County/Oregon Public Health Division, 800 NE Oregon St, Ste 260, Portland, OR 97232 (julia.dilley@state.or.us).
Published Online: December 19, 2018. doi:10.1001/jamapediatrics.2018.4458
Author Contributions: Dr Dilley had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: All authors.
Acquisition, analysis, or interpretation of data: Dilley, Richardson, Kilmer, Pacula, Segawa.
Drafting of the manuscript: Dilley, Richardson.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Dilley, Richardson, Pacula.
Obtained funding: Dilley, Richardson.
Administrative, technical, or material support: Dilley, Richardson, Segawa.
Supervision: Dilley, Pacula, Cerdá.
Conflict of Interest Disclosures: None reported.
Funding/Support: Research reported in this article was supported by the National Institute on Drug Abuse of the National Institutes of Health (NIH) under award number 1R01DA039293.
Role of the Funder/Sponsor: The National Institute on Drug Abuse had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.
Additional Contributions: Melanie Wall, PhD (Columbia University), served as a consultant on methods used in Monitoring the Future survey analysis. No compensation was received.
2.Washington State Department of Social and Health Services, Department of Health, Office of the Superintendent of Public Instruction, and Liquor and Cannabis Board. Healthy Youth Survey 2016 Analytic Report, Olympia, WA.
http://www.AskHYS.net. Published June 2017. Accessed April 30, 2018.