Neighborhood Sociodemographic Correlates of Cannabis Dispensary Availability in States with Legalized Adult Recreational Use, United States, 2021.

Abstract

Introduction: The availability of recreational cannabis in the United States has increased dramatically. Neighborhood cannabis dispensary availability may increase problematic use. State and local studies suggest that cannabis dispensary availability may be correlated with neighborhood sociodemographic characteristics. We provide a national-level examination of census tract (CT) sociodemographic characteristic correlates with the availability of cannabis dispensaries in 18 U.S. states that have legalized adult recreational cannabis use. Methods: We downloaded the locations of cannabis dispensaries (N = 3,167) from Weedmaps in November 2021. We downloaded ethnic, racial, and socioeconomic estimates from the U.S. Census and categorized CTs into quintiles (Q), where Q1 represents the CTs with the lowest percentage of a variable and Q5 represents CTs with the highest. We fit unadjusted generalized linear mixed models to examine associations between quintiles of each sociodemographic characteristic and the presence of at least one dispensary (vs. none). Results: CTs with a greater percentage of Black residents had a higher odds of having at least one cannabis dispensary (vs. none) across all quintiles. For example, compared to CTs with the lowest percentage of Black residents (Q1), CTs with the greatest percentage of Black residents (Q5) had 2.07 (95% CI: 1.70-2.52) times the odds of having at least one dispensary versus none. We observed a similar pattern of a greater likelihood of a CT having a dispensary (vs. none) as the percentage of Hispanic/Latine residents and percentage of individuals living below the federal poverty line increased across all quintiles. In contrast, as the percentage of homeowner-occupied housing increased across all quintiles, there were lower odds of having at least one dispensary (vs. none). For example, CTs with the highest percentage of homeowner-occupied housing (vs. lowest) had 0.21 (95% CI: 0.17-0.26) times the odds of having at least one dispensary vs. none. Discussion/Conclusions: Differences in cannabis dispensary availability exist by neighborhood sociodemographic composition. The growing acceptance of cannabis, concern about public health and safety, and the proliferation of adult recreational use laws offer local and state policymakers an opportunity to engage local communities in shared decision-making about the location of dispensaries.