County- and Individual-Level Cannabis Norms and Adolescent Cannabis Use: A Cross-Sectional, Multilevel Analysis of Oregon Youth

Introduction: Since 2012, four states and the District of Columbia have legalized cannabis for non-medical purposes for those over the age of 21. Liberalization of cannabis policies within multiple states has raised public health concerns about potential impacts on adolescent cannabis use. These impacts are hypothesized to come from supply side mechanisms such as increased access to cannabis or demand side mechanisms such as loosening social norms and decreasing perception of harm related to adolescent cannabis use. County-level cannabis norms, as represented by voting patterns on cannabis-related ballot initiatives, may help contextualize well-known associations among individual-level norms and perceptions about cannabis and adolescent cannabis use. Prior to passing a ballot initiative to legalize cannabis in 2014 and increasing access to cannabis for adults, Oregonians voted on and narrowly failed to approve similar legislation (Measure 80 [M80]) in 2012. This presentation will examine the associations between county-level voting on M80, individual-level norms and perceptions about cannabis and adolescent cannabis use.

Methods: Data were collected from over 20,000 8^th and 11^th graders from 32 of 36 counties by the 2013 Oregon Healthy Teen survey. Outcomes included adolescent self-reports of past month cannabis use, perceived harm from occasional cannabis use and both peer and parent norms about cannabis use. County-level voting results for M80 in 2012 were obtained from the office of the Secretary of the State. County-level demographic and socioeconomic characteristics were obtained from the American Community Survey. All models controlled for individual- and county-level demographic and SES covariates.

Results: Results from a series of multi-level logistic regression models indicated that higher levels of county-level endorsement for M80 were associated with less restrictive parent norms, reduced perception of risk and a higher likelihood of past month cannabis use for 8^th graders. Among 11^th graders, higher levels of endorsement for M80 were associated with less restrictive parental norms alone. As expected, students reporting more restrictive parent and peer cannabis norms and greater perception of harm had a lower likelihood of any past month cannabis use. Further analyses will examine higher thresholds for cannabis use (3+ times and 10+ times in the past month).

Conclusions: Results suggest the importance of both proximal and more distal contextual cannabis-related norms for understanding adolescent cannabis use; and, highlight typically unexamined associations between county-level cannabis norms and individual-level norms and use. Limitations, implications for prevention programs and suggestions for evaluating changing cannabis laws will be discussed.