Neighborhoods within 1000 meters of a cannabis retail store experienced a monthly increase of 1.30 cannabis-attributable emergency department visits per 100,000 persons compared with unexposed neighborhoods following store openings in Ontario, Canada—equivalent to a 12% relative increase from a baseline rate of approximately 11 to 13 visits per 100,000 per month—according to a quasi-experimental study published in Annals of Internal Medicine.
The population-based natural experiment examined 10,574 neighborhoods containing 6.1 million persons aged 15 years and older between April 2017 and December 2022. Cannabis retail policy in Ontario evolved through several phases: legalization in October 2018 initially permitted only online government sales, followed by a lottery system allowing up to 75 stores (April 2019–March 2020), then unrestricted commercial expansion beginning April 2020—coinciding with the COVID-19 pandemic. Researchers tracked the opening of all cannabis stores in Ontario to identify neighborhoods that became exposed (defined as having a cannabis store within 1000 meters) over time, then compared absolute and relative changes in rates of cannabis-attributable emergency departments (ED) visits between exposed neighborhoods and matched unexposed neighborhoods.
Cannabis-attributable ED visits were increasing before retail store openings. After stores opened, rates remained constant in exposed neighborhoods while decreasing in matched unexposed neighborhoods.
Dose-Response Relationship
The study revealed a dose-response relationship between store density and harms. Neighborhoods with five or more stores within 1000 meters experienced an additional 4.26 visits per 100,000 persons per month compared with unexposed neighborhoods (a measure representing how much more the rate changed in exposed vs unexposed areas). Neighborhoods with three stores within 1000 meters showed an additional 1.93 visits per 100,000 persons. Mean monthly rates of cannabis-attributable ED visits increased after store opening in exposed neighborhoods that had 3 or more stores within 1000 meters. Secondary analyses found similar effects for neighborhoods up to 3000 meters from stores.
Age and Sex Patterns
Age-stratified analyses showed varying patterns. In patients aged 15 to 18 and 19 to 24 years, cannabis-attributable ED visit rates remained stable in exposed neighborhoods but declined in unexposed neighborhoods after store opening. In contrast, for patients aged 25 to 44 and 45 to 64 years, rates increased in exposed neighborhoods while decreasing or remaining stable in unexposed neighborhoods. The magnitude of additional visits in exposed versus unexposed neighborhoods declined with age: 15 to 18 years (7.01 per 100,000 persons), 19 to 24 years (3.46), 25 to 44 years (1.91), 45 to 64 years (0.87), and 65 years and older (−0.13).
Sex-stratified analyses showed statistically significant effects for males (1.35 additional visits per 100,000) and a similar but non-significant pattern for females (1.08 additional visits), with smaller absolute effects among females.
Study Population and Methods
The cohort included 19,098 neighborhoods in Ontario containing 11.1 million persons in April 2019. By study end, 5,812 neighborhoods (30%) with a population of 3.3 million persons (29%) had a cannabis retail store open within 1000 meters. Researchers matched exposed neighborhoods to unexposed neighborhoods with similar income and rurality characteristics. The most common reason for cannabis-attributable ED visits was harmful use (50%), followed by acute intoxication (17%), withdrawal or dependence (10%), cannabis poisoning (9%), and cannabis-induced psychosis (8%).
Commercial Growth
The percentage of neighborhoods in Ontario exposed to a cannabis retail store increased from 1% in April 2019 (date stores opened) to 2% in March 2020 (month before commercialization) to 30% in December 2022 (end of study). Commercial growth began to plateau around January 2022.
Study Limitations
The study has several important limitations. The observational design cannot rule out unmeasured confounding between exposed and unexposed neighborhoods, such as differences in illicit cannabis market access. Much of the study period overlapped with the COVID-19 pandemic, which affected both cannabis consumption patterns and health care utilization, potentially limiting generalizability to post-pandemic contexts. Additionally, the findings reflect only one aspect of cannabis commercialization (retail store proximity) and do not capture effects of pricing, product potency, or marketing. Part of the observed changes in cannabis-related ED visits may also be driven by reporting and diagnostic biases—for example, individuals may be more likely to report cannabis use in ED after legalization. However, many of these biases would not be anticipated to differ between exposed and control regions.
Funding and Authorization
The study was supported by ICES, which is funded by an annual grant from the Ontario Ministry of Health and the Ministry of Long-Term Care. The research received a catalyst grant from the Canadian Institutes of Health Research, according to lead study author Daniel T. Myran, MD, of the Department of Family Medicine at the University of Ottawa, and colleagues. The study was authorized under section 45 of Ontario's Personal Health Information Protection Act and did not require research ethics board review. Disclosures can be found in the published study.
Source: Annals of Internal Medicine