Premise surveys in LA County also indicated that unlicensed dispensaries were more likely to be found located near sensitive areas where dispensary regulations prohibit them . Thus, although there is a smaller number of unlicensed dispensaries operating in a city compared to the total number of dispensaries, unlicensed outlets may have greater impact of the likelihood of lifetime marijuana use among the city’s high school students. In contrast, licensed storefront dispensaries have considerable incentive to comply with State and city regulations to maintain their licensure. Many have received their licenses through a lottery process where dozens of prospective business owners have applied for the handful of licenses that are ultimately granted . Others have endured long waits before obtaining licenses as cities and states process a backlog of applicants. Finally, licensed dispensaries stand to lose out on immense profits if they lose their license. For example, in an interview with Forbes Magazine, one California dispensary owner reported that on a busy day her shop could serve as many as 1,000 customers, while on a slow day they could serve closer to 500-600. Since the dispensary’s customers spent on average $50 per person, a busy sales day could bring in $50,000 . With Research Question 5, I hypothesized that the number of dispensaries located within 2,000 feet of a school would be an important mediator of the effect of dispensary bans on student marijuana use if dispensary bans were more effective than city policies that allow dispensaries at limiting the number of unlicensed outlets near schools .
H5.1 was supported, as significantly less dispensaries were found within 2,000 feet of the study participants’ schools . However,4 x 8 grow tray the number of dispensaries within 2,000 feet only had a significant positive association for lifetime marijuana use , and not for recent use . Adding the number of dispensaries within 2,000 feet to the regression equation did not substantially strengthen the association between dispensaries and student marijuana use and the association remained non-significant for both lifetime and recent use. However, investigating the number of dispensaries within 2,000 feet has identified that unlicensed dispensaries have significant effects on students’ lifetime marijuana use behaviors at more than triple the distance the State of California requires dispensaries to be located away from schools. This finding indicates a need to rethink how dispensaries are zoned and regulated. Their number did not have an effect on a city level, but their distance did and having them near schools does impact some measures of marijuana use. Perhaps instead of siting dispensaries in mainstream business districts that are close to residential areas, they should be located in industrial zones, as in the City of Huntington Park. Current regulations limit how close they can be to each other, but perhaps a better approach may be to cluster dispensaries in the areas of a city where they are the furthest away from sensitive areas like schools. Given the economic potential of the marijuana market, this could even present a revitalizing presence for industrial zones where manufacturing presence has been declining. This could also keep different functions of the marijuana industry in close proximity, where labs and growing operations could be near storefront outlets.
There are certain limitations to this dissertation that must be acknowledged. A key limitation of this study is that it could not directly measure constructs that were not measured in the CHKS or city policy data, such as the resources and approaches different cities used to enforce dispensary regulations. Enforcement in particular may be important. It is one reason why I included analyses using the actual number of dispensaries in a city or near the schools, because so many cities had dispensaries located within their borders despite dispensary bans, which could be expected to confound results. Enforcement is a factor that should be studied much more extensively in the future using qualitative as well as quantitative approaches to determine why there was so much variation in the effectiveness of dispensary bans between cities. Another important limitation of this study is that I was only tangentially able to account for dispensaries being located nearby a city but outside of its’ borders by using the continuous measure of distance between the school and the nearest dispensary. Many of the unlicensed dispensaries in LA County in this study were found in the unincorporated areas bordering incorporated cities that did not allow dispensaries. This could partially explain the lack of a significant effect for dispensary bans and for the number of dispensaries in city, as residents of those cities can easily cross city borders to buy marijuana from adjacent areas. Similarly, the City of Los Angeles’ is colossal size and irregular borders mean that many of the incorporated cities in LA County share a border with it, and resident of bordering cities could easily obtain marijuana from Los Angeles dispensaries.
Additionally, the ubiquity of marijuana delivery services operating in Los Angeles County may confound these analyses of the spatial influence of marijuana dispensaries and would be expected to weaken spatial associations by representing a source for marijuana that may circumvent city bans on dispensaries to the extent that delivery-only dispensaries can deliver marijuana to cities that ban storefront MMDs. Furthermore, enforcement of dispensary regulations among the cities of LA County has been uneven at best, resulting in considerable variation of the effectiveness of dispensary bans by city. Medical marijuana delivery and mail order services may also hamper the effectiveness of dispensary bans in limiting the availability of marijuana in a city. Additional limitations are based in the data I used. The CHKS survey is a well-validated behavioral survey, but it is not a population-based random sample and was not designed to compare student substance use behavior between cities, but rather over time within districts. It does not use a complex sampling strategy and therefore there are not weights than can be used to approximate populations. The CHKS sample is therefore more of a convenience sample when used at the city level and results from these analyses can’t be construed to be representative of all the students in a city, even if due to the large sample size most of the city’s students participated in the survey. I also needed to pool two school years of data to include more of the schools in LA County because the survey is administered by schools every other year and on a staggered basis, so on any given year some schools will be between surveys. This meant that although the data collection for the dispensary locations and city policies fell in the middle of the range used,vertical racking the time of data collection and the measurement of student marijuana use did not coincide as perfectly as if I had used data only from the 2016/2017 school year. The data for dispensary locations does not come from an official source but rather a commercial listing of dispensaries. I did what I could to verify that outlets were in active operation and excluded those I could not verify, but I suspect I may have excluded some unlicensed dispensaries that were in operation but that screened phone calls. Recent studies should like the Los Angeles County Cannabis Dispensary Premise Survey were able to use a more rigorous method of verification by driving out to each dispensary and reported similar numbers of licensed and unlicensed dispensaries in the County in 2019 and 2019, although they also found that the number of unlicensed outlets is declining in cities that allowed dispensaries so the 3:1 ratio of unlicensed to licensed outlets that I observed in 2016 is likely to change. There are a number of strengths to this dissertation that should be highlighted. It is the first to investigate the influences of city dispensary policies, the actual number of dispensaries in a city on high school students’ marijuana use at a city level, and the proximity of dispensaries on high school students’ marijuana use. It also integrates data of different types and for different sources: administrative data like the city policy database I collected from municipal codes and school addresses and attributes from the California Schools Directory, location data from online dispensary listings, and school-based survey data from the California Healthy Kids Survey. The CHKS study provides detailed information not only about students’ substance behaviors, but also their attitudes about the risks of harm from substance use and how easy they perceived it for their peers to access substances, which allowed for investigation not only of marijuana use but attitudes that research has shown is correlated with marijuana use.
Including these variables resulted in important findings about not only students who already use marijuana, but variables associated with use, such as perceived risk. This information is key from a prevention perspective and it is hoped that the data presented in this dissertation can be applied to improve substance abuse prevention among high school students in LA County. To date, much of the research on adolescent substance use has been confined to individual-level variables when studying influences on substance use behaviors. These proximal influences on substance use behavior are undoubtedly powerful, but are less amenable to change using cost-effective primary prevention approaches like policies and regulations that discourage use. While there have been studies that have looked at neighborhood and community-level measures and their influence on marijuana use , they have not focused on the individual-environment interaction that can be studied using hierarchical generalized linear modeling. Local leaders want to know the best use for their limited resources when it comes to preventing marijuana-related harms like youth use. Accounting for the influence of city dispensary policies on an individual level was essential to capturing the differences in dispensary regulations among the cities of LA County and determining whether these different approaches had an effect on youth use in the population at greatest risk of harm from marijuana use: adolescents. The focus on adolescents presents an additional strength. This study represents the first comparative analysis of the effects of city dispensary policies on youth use. As demonstrated from the results of this dissertation, the influence of dispensaries is very localized and further study on a neighborhood level is needed. Hopefully the findings from this study will justify the need the need for population-based research at a more local level going forward. Finally, by using a continuous measure of distance between schools and dispensaries this study has demonstrated that their influence extends much further than expected. I had expected to compare the associations of the distance participants’’ schools and marijuana use starting with the longest distance, the distance to the nearest unlicensed dispensary within the County, then to the nearest dispensary within a mile, and then within a mile, getting progressively closer to the school before finding a significant effect. Instead, the effect of the distance to the nearest dispensary in LA County was statistically significant, as it also was at a mile and for lifetime use, within 2,000 feet. Some of the findings in this dissertation were rather unexpected and merit further investigation. The most important unanswered question is how LA County youth are obtaining marijuana and from who. The results presented here suggest that they are obtaining it one way or another from unlicensed dispensaries, but recent research indicating that compliance with ID checks is high even at unlicensed dispensaries suggests that youth are not obtaining it from dispensaries directly. More study is needed about compliance with other regulations that might be associated with diversion to the illicit market and updated, local research about where or who adolescents obtain marijuana from. That the number of MMDS located in a city was not influential on high school students’ marijuana use was truly surprising and merits further study. I also tested the raw number and normalizing the number of dispensaries by the area in square miles of the city, with similar results. That this variable was not statistically significant in the multivariate regression models is important but should not be taken as meaningless. For example, although the number of dispensaries did not seem to influence the students’ behavior unless they were located near an area they frequented, in this analysis the number of dispensaries in a community was linked to the number that were found near schools.