Study: Dispensary Distance Doesn't Drive Teen Cannabis Use

October 14, 2026

A new peer-reviewed study tracking nearly 420,000 Canadian teenagers over five years has produced a finding that cuts against decades of zoning policy: how close a kid lives to a cannabis dispensary doesn't meaningfully predict whether they'll use the stuff. The research, drawn from a massive repeat cross-sectional dataset collected between 2018 and 2023, lands squarely on one of the most durable assumptions in cannabis regulation -- the idea that keeping stores a few hundred feet further from a school building meaningfully protects students.

That assumption underwrites buffer-zone laws across North America. Legislators in dozens of U.S. states and Canadian provinces wrote setback requirements into their adult-use frameworks almost reflexively, treating physical distance as a proxy for exposure and exposure as a proxy for risk. Canada's legal market, now mature enough to have generated six years of post-legalization data, finally gave researchers a long enough window to actually test that chain of logic rather than just assume it. The results suggest the chain breaks down well before it reaches teenage behavior -- which raises an uncomfortable question for policymakers who built zoning rules around it: if dispensary proximity isn't the driver, what is?

Inside the COMPASS Study Data

Photo by https://kaboompics.com/ via Pexels.

The study, published November 1, 2026 in Drug and Alcohol Dependence (Volume 288, Article 113368), comes out of the HEAL research group at Western University under lead author Jamie Seabrook. It leans on the COMPASS study, a long-running repeat cross-sectional survey that has been collecting health behavior data from Canadian secondary school students for years. For this analysis, the team pulled responses from 419,887 students across Alberta, British Columbia, Ontario and Quebec between 2018 and 2023 -- a sample size large enough to detect even modest effects if they existed.

What makes the design notable is the geospatial linkage. Rather than relying on students' self-reported sense of how far away a dispensary was, researchers mapped each school's actual location against licensed retailer data and calculated three separate exposure measures: straight-line distance to the nearest retailer, raw retailer density in the surrounding area, and an inverse distance weighted density score within a 2,000-meter radius of each school. That last measure matters because it accounts for clusters of multiple stores sitting at varying distances, rather than just asking whether one store happens to be nearby.

On the behavioral side, the team ran multilevel logistic regression models against four separate use outcomes: ever having tried cannabis, use in the past 12 months, use in the past 30 days, and daily or almost-daily use. For students who were already using, they layered on ordinal regression to look at frequency patterns. That multi-angle approach is what lets the authors speak with some confidence -- they weren't just checking one outcome against one exposure measure, they were stress-testing the proximity hypothesis from several directions at once.

What the Numbers Actually Show

Despite the spread of legal cannabis dispensaries across Canada, past 12-month adolescent cannabis use steadily declined from 22.2% in 2018 to 10.0% in 2023, showing no evidence of a youth usage spike linked to legalization.

The headline trend in the data has nothing to do with retail geography at all: past-12-month adolescent cannabis use fell from 22.2% in 2018 to 10.0% in 2023, cutting by more than half during the exact window when Canada's legal retail footprint expanded dramatically. If physical proximity to stores were a meaningful driver of youth use, that curve should have bent the other way, or at least flattened. Instead, teen use kept dropping as storefronts kept multiplying.

When the researchers isolated the retail access variables, the associations with any level of youth use came back small and inconsistent -- they shifted direction depending on the province and the year, which is usually a sign that whatever relationship exists is weak enough to be noise rather than signal. There was one exception worth flagging: in the earliest post-legalization period, 2018 to 2019, retail proximity did show a measurable link to frequent use among kids who were already using cannabis. But that effect vanished entirely in later years as the market settled and normalized, suggesting it was a transitional artifact of a brand-new retail system rather than a stable pattern.

Far more predictive than any physical distance measure was perceived access -- essentially, how easy a teenager believed it would be to get cannabis if they wanted it. That subjective sense of availability tracked with actual use far more consistently than any map-based metric did. The authors' explanation is straightforward: legal age restrictions keep minors out of licensed stores almost entirely, so the dispensary on the corner isn't really a supply option for a 15-year-old. Social sources -- older friends, siblings, parents' stashes -- remain the actual pipeline, and those pathways don't care how many meters away the nearest storefront sits.

Why Buffer-Zone Laws May Miss the Point

Photo by Erik Mclean via Unsplash.

This matters because buffer-zone requirements are everywhere. U.S. states and Canadian provinces routinely mandate that dispensaries sit 500 to 1,000 feet or more from schools, parks, or daycares, almost always justified with some version of the claim that proximity invites youth exposure and use. The Western University team's conclusion is blunt on this point: restricting retailer proximity to schools appears to have limited real-world impact on adolescent behavior, at least based on what six years of Canadian data show.

This isn't an isolated finding. It echoes a 2018 study out of UC San Diego, published in the Journal of Adolescent Health by Shi, Cummins, and Zhu, which looked at California data and found no link between school-to-dispensary distance and either past-month use or future susceptibility to using. Two studies, different countries, different legal frameworks, different years -- landing on the same basic conclusion is the kind of convergence that's hard to wave away as a fluke.

NORML Deputy Director Paul Armentano has pointed to the structure of licensed markets as part of the explanation. Regulated dispensaries check ID at the door and frequently employ dedicated security staff, which makes them a fundamentally different kind of access point than, say, a dealer operating without any age verification at all. A legal storefront being physically close to a school doesn't translate into easier access for a minor the way an unregulated source would.

Taken together, the research points toward social, behavioral, and household-level factors as the stronger levers for reducing adolescent use -- not the geography of where stores are allowed to sit.

What This Means for Regulators and Operators

Having two major studies from different countries and different regulatory frameworks reach the same conclusion strengthens the case that distance-based zoning deserves a second look. This isn't one outlier dataset -- it's a pattern holding up across jurisdictions and years, which is exactly the kind of evidence that should inform policy rather than get dismissed as a one-off.

For operators, this research has practical teeth. Dispensary applicants who run into arbitrary buffer-distance disputes with city councils or zoning boards now have a citable, peer-reviewed study to bring into the room when a proposed setback gets challenged as protective rather than symbolic. That doesn't mean every zoning board will budge, but it shifts the burden of proof back toward the side defending the buffer requirement.

For regulators and public health officials, the more interesting implication is about where prevention money goes. If physical proximity isn't moving the needle but perceived access and social sourcing are, then dollars spent widening retail setbacks may be dollars misallocated. Education programs, family communication initiatives, and efforts that directly address teens' perception of how easy cannabis is to get all look like better investments based on this data than another few hundred feet of distance between a storefront and a school parking lot.

None of this changes the legal reality on the ground, which still varies enormously by state and country. Zoning ordinances, licensing rules, and buffer requirements are set locally and change often, so operators, advocates, and anyone citing this research in a policy debate should confirm current local law before assuming it applies to their situation.

None of this means dispensary location is irrelevant to every policy question a city might reasonably ask. Traffic, odor complaints, neighborhood character, and a dozen other legitimate zoning concerns still exist independent of youth use rates. What this data actually undercuts is narrower and more specific: the claim that distance from a school is doing meaningful protective work against teenage cannabis use. Across 419,887 students and six years of a maturing legal market, it just wasn't.

If perceived availability and social sourcing are the real drivers, the policy conversation needs to follow the evidence there instead of staying parked on retail maps. That means prevention education aimed at correcting kids' assumptions about how easy cannabis is to get, and it means supporting the kind of family-level conversations that actually interrupt the friend-to-friend and sibling-to-sibling pathways researchers keep identifying as the real supply chain for underage use.

Expect this study to show up again, probably in a zoning board meeting near you. As more legal markets age past their chaotic first few years and start generating the kind of multi-year datasets Canada now has, the proximity assumption is going to keep getting tested -- and based on what's come back so far, it's going to keep failing to hold up.