Noncompliance with laws to prevent polysubstance misuse: Recreational cannabis sales to apparently alcohol-intoxicated customers.

Noncompliance with laws to prevent polysubstance misuse: Recreational cannabis sales to apparently alcohol-intoxicated customers - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Official websites use .gov A .gov website belongs to an official government organization in the United States. Secure .gov websites use HTTPS A lock ( Lock Locked padlock icon ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites. Search Log in Dashboard Publications Account settings Log out Search… Search NCBI Primary site navigation Search Logged in as: Dashboard Publications Account settings Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide PERMALINK Copy As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice Alcohol Clin Exp Res (Hoboken) . Author manuscript; available in PMC: 2026 Apr 15. Published in final edited form as: Alcohol Clin Exp Res (Hoboken). 2026 Jan;50(1):e70204. doi: 10.1111/acer.70204 Search in PMC Search in PubMed View in NLM Catalog Add to search Noncompliance with laws to prevent polysubstance misuse: Recreational cannabis sales to apparently alcohol-intoxicated customers W Gill Woodall W Gill Woodall 1 Klein Buendel, Inc., Golden, Colorado, USA Find articles by W Gill Woodall 1 , David B Buller David B Buller 1 Klein Buendel, Inc., Golden, Colorado, USA Find articles by David B Buller 1 , Robert Saltz Robert Saltz 2 Prevention Research Center, Berkeley, California, USA Find articles by Robert Saltz 2 , James C Fell James C Fell 3 James C. Fell Consulting, LLC, Annandale, Virginia, USA Find articles by James C Fell 3 , Lila Martinez Lila Martinez 1 Klein Buendel, Inc., Golden, Colorado, USA Find articles by Lila Martinez 1 , Amanda N Brice Amanda N Brice 1 Klein Buendel, Inc., Golden, Colorado, USA Find articles by Amanda N Brice 1 , Noah Chirico Noah Chirico 1 Klein Buendel, Inc., Golden, Colorado, USA Find articles by Noah Chirico 1 , Gary R Cutter Gary R Cutter 4 School of Public Health, University of Alabama Birmingham, Birmingham, Alabama, USA Find articles by Gary R Cutter 4 Author information Copyright and License information 1 Klein Buendel, Inc., Golden, Colorado, USA 2 Prevention Research Center, Berkeley, California, USA 3 James C. Fell Consulting, LLC, Annandale, Virginia, USA 4 School of Public Health, University of Alabama Birmingham, Birmingham, Alabama, USA ✉ Correspondence W. Gill Woodall, Klein Buendel, Inc., 1667 Cole Boulevard, Suite 220, Golden, CO 80401, USA. gwoodall@kleinbuendel.com PMC Copyright notice PMCID: PMC13078678  NIHMSID: NIHMS2144898  PMID: 41575362 The publisher's version of this article is available at Alcohol Clin Exp Res (Hoboken) Abstract Background: Effectiveness of statutes to combat hazardous polysubstance use has rarely been evaluated. Compliance with a state law prohibiting recreational cannabis sales to apparently intoxicated customers was assessed in one of the first states to legalize cannabis sales. Methods: In January to June 2024, pseudo-patrons twice visited recreational cannabis stores ( n = 189) in two large metropolitan areas and attempted to purchase cannabis while displaying alcohol intoxication behaviors. Observers recorded whether sellers were willing to sell the product along with characteristics of the stores (busyness, cleanliness, and signage) and cannabis sellers (sex, race and ethnicity, and age). Neighborhood characteristics by US census tract were obtained (income, race, and ethnicity; population density). Sex and race and ethnicity of the pseudo-patrons, and extent and type of intoxication cues, were recorded. Descriptive statistics and logistic regression were used to describe the sales rate and predictors of the sales rate. Results: Assessments were completed at 173 stores. Sellers were willing to sell cannabis to pseudo-intoxicated buyers at 255 of 346 visits (73.7%; 26.3% refused). Sellers refused buyers at both visits in 6.9% of stores but in 54.3% were willing to sell at both visits. Sellers refused cannabis sales at higher rates in stores with signs saying, “no sales to intoxicated customers” (34.3%, p = 0.04), particularly when buyers displayed more obvious signs of intoxication (39.8%, p = 0.049). Conclusions: Low compliance with the state regulation possibly occurred because sellers were unaware of the law, perceived little deterrence, or lacked the skills to recognize and refuse intoxicated customers. Policy Implications: Noncompliance with the law on selling cannabis to apparently alcohol-intoxicated customers increases the risks of polysubstance impairment and harm. Keywords: alcohol and cannabis, driving while intoxicated, point of sale, policy, prevention INTRODUCTION In 2024, 24 US states and three US territories had legal recreational cannabis sales ( Schaeffer, 2024 ). Recreational cannabis sales laws in many states are similar to alcohol laws and are often overseen by alcohol control agencies ( Barry & Glantz, 2018 ). A few states prohibit sales to customers who show signs of intoxication with the aim of reducing morbidity and mortality from impaired driving, impulsive behaviors, injuries, and violence ( Foxcroft, 2014 ). Emerging evidence has shown that cannabis use can impair driving skills ( Compton, 2017 ; Sevigny, 2021 ) and increase motor vehicle crash risk ( Brubacher et al., 2019 ; Elvik, 2013 ; Farmer et al., 2022 ; Washington Traffic Safety Commission, 2015 ) and other harms ( Marinello & Powell, 2023 ). Cannabis sales were associated with increased motor vehicle crashes and possibly more so in states that first legalized sales ( Adhikari et al., 2023 ; Aydelotte et al., 2019 ; Calvert & Erickson, 2020 ; Couper & Peterson, 2014 ; Farmer et al., 2022 ; Kamer et al., 2020 ; Reed, 2018 ; Santaella-Tenorio et al., 2020 ; Tefft & Arnold, 2020 ). There are some limitations to this research, primarily based on a lack of a standardized and reliable method for measuring cannabis impairment in drivers, but across different kinds of studies (driver simulation experiments, traffic safety studies linking cannabis, and auto crash rates), the evidence suggests a link between cannabis consumption, impairment during driving, and increased risk for car crashes and resultant morbidity and mortality. Cannabis and alcohol used together can result in high impairment ( Hartman et al., 2015 ; Marillier & Verstraete, 2019 ; Yurasek et al., 2017 ) including impaired driving ( Dubois et al., 2015 ), even at low doses of alcohol (0.04 Blood Alcohol Content— BAC), and tetrahydrocannabinol (THC) ( Ramaekers et al., 2000 ). Recreational cannabis sales may increase couse with alcohol as users seek a more pleasurable high, consume more alcohol when cannabis-impaired, or find more places for couse ( Guttmannova et al., 2016 ). Polysubstance-impaired driving appeared to increase after legalization of recreational cannabis in some studies ( Couper & Peterson, 2014 ; Reed, 2018 ), especially alcohol and cannabis ( Couper & Peterson, 2014 ; University of Montana and Washington Traffic Safety Commission, 2018 ). In the United States, fatal traffic crashes involving both cannabis and alcohol doubled from 9% in 2000 to 22% in 2018 ( Lira et al., 2021 ). Co-use of cannabis and alcohol was connected to risky, impulsive behavior, injury, and violence (including intimate partner violence) ( Asbridge et al., 2012 ; Cherpitel et al., 2017 ; Jackson et al., 2020 ; Romero-Martínez et al., 2019 ; Wardell et al., 2020 ). In 2023, a review of responsible cannabis sales was published by Lange et al. (2023) , which explored responsible cannabis sales training and policies. The study concluded that recreational cannabis outlet staff training, including management policies, could improve safety for consumers and the community ( Lange et al., 2023 ). A common policy approach has focused on direct deterrence of polysubstance users driving by visible impaired-driving enforcement ( Fell et al., 2013 ). Current DUI laws do not adequately address polysubstance intoxicated driving because (1) there is no reliable established THC blood standard that signifies intoxication, (2) unlike alcohol, there is no current field technology similar to alcohol breathalyzers that assesses cannabis intoxication, and (3) evaluations by law enforcement officers trained in drug recognition can be poor ( Marcotte & Fitzgerald, 2025 ). One alternative to prevent polysubstance intoxicated driving and other harms is state policies prescribing responsible sales behavior by cannabis sellers to prevent access to cannabis by persons already intoxicated by alcohol, which may be effective by relying on decision making by the cannabis seller and not by the impaired buyer. This policy intervention is common in alcoholic beverage markets ( Mosher et al., 2009 ; National Highway Traffic Safety Administration, 2017 ) but not in recreational cannabis markets. In Oregon, violation of the prohibition against selling to apparently intoxicated customers is a Category III Oregon Liquor and Cannabis Commission violation with a 10-day suspension of a cannabis seller’s permit and/or monetary fine of $25/day for each day of the suspension ( Oregon Liquor and Cannabis Commission, 2025 ; Oregon Revised Statutes, 2023 ). The success of this marketplace regulatory policy intervention (which is concerned with the allowable sale of recreational cannabis based on age and apparent intoxication of customers) depends on compliance by recreational cannabis retailers. Assessment in three states in 2016 and 2018, one with a law directly prohibiting sales to intoxicated customers, found that refusal of recreational cannabis sales to pseudo-patrons displaying signs of alcohol intoxication was very low (11%) ( Buller et al., 2020 ). In this paper, we report the assessment of compliance with a state law banning sales of recreational cannabis to customers who appeared to be intoxicated by alcohol in 2024 in Oregon. This assessment occurred 6 years after the study in 2016 and 2018 in a larger sample of cannabis stores, in a more mature market (9 years after legalization), after the COVID-19 pandemic when sales revenue from cannabis increased ( Dilley et al., 2023 ) and cannabis and alcohol use by seriously or fatally injured drivers increased ( Thomas et al., 2020 ). In an analysis through 2022, Oregon’s recreational cannabis market was experiencing little growth (in number of licensees, revenue, and average revenue per retailer) but had the largest density of licensees (16.8 per 100,000 state residents) ( Dilley et al., 2023 ), which may have created pressure to sell. We employed a methodology where trained actors enacted behaviors indicative of alcohol intoxication while attempting to purchase cannabis, which has been used effectively to assess alcohol overservice in licensed premises, sporting events, and music festivals ( Buller et al., 2024 ; Buvik & Rossow, 2015 ; Elgán et al., 2018 ; Feltmann et al., 2020 ; Gosselt et al., 2013 ; Gripenberg et al., 2007 , 2011 ; Hughes, 2014 ; Quigg et al., 2022 ; Saltz et al., 2024 ; Toomey et al., 2008 , 2016 ; Woodall et al., 2018 ). The data collected here were a baseline assessment for a trial evaluating an intervention to improve seller compliance and as such it was designed to test the research question of what was the current rate of compliance with the prohibition of sales to customers who appeared intoxicated by alcohol ( LoParco et al., 2024 ). MATERIALS AND METHODS Sample of state-licensed recreational cannabis stores Initially, 384 stores licensed by the OLCC for the sale of recreational cannabis located in the greater Portland and Salem, Oregon metropolitan areas, defined as Clackamas (population: 415,084), Multnomah (809,869), Marion (61,817), and Washington (595,761) Counties (see combined population density, income, and race and ethnicity of the four counties in Table 1 ) were identified from a publicly available list of approved licensees from OLCC as of October 30, 2023. All independently owned stores were included in the sample ( n = 169 stores), while for multistore chains, only one store was selected at random from each chain ( n = 51 stores) to avoid contamination and the need to adjust statistical analysis for clustering in the subsequent research trial. After an initial screening of the stores by visiting their websites and social media accounts or contacting them by telephone, we removed 12 stores that were closed, 16 stores whose existence or operating status could not be confirmed, two stores that sold cannabis through home delivery only, and two stores that were located in an adjacent county. One newly licensed independent store was also identified and added to the sample. We attempted to visit and assess all selected stores ( n = 189) that were in business and open to the public between January and July 2024. TABLE 1. Profile of cannabis store and seller samples and PiP buyer visits ( n = 346 assessment visits at n = 173 cannabis stores). n (%) Store characteristics How busy was store ( N = 339 visits) Quiet 210 (61.9%) Average 82 (24.2%) Busy 38 (11.2%) Extremely busy 9 (2.7%) Day of visit ( N = 346 visits) Thursday 118 (34.1%) Friday 117 (33.8%) Saturday 102 (29.5%) Sunday 1 (0.3%) Monday 0 (0%) Tuesday 1 (0.3%) Wednesday 7 (2.0%) Time of visit ( N = 346 visits) 4:00–5:59pm 96 (27.7%) 6:00–6:59 pm 107 (30.9%) 7:00–7:59 pm 94 (27.2%) 8:00–9:59pm 49 (14.2%) Was ID requested ( N = 346 visits) Overall–No 49 (14.2%) Yes 297 (85.8%) For entry–Yes 158 (53.2%) After entry and before purchase attempt a –Yes 141 (47.5%) For purchase attempt b –Yes 160 (53.9%) “No sale to apparently intoxicated customers” signage observed ( N = 339 visits) No 243 (71.7%) Yes 96 (28.3%) Neighborhood characteristics ( N = 173 stores) Population density of store location-persons per square mile <5000 64 (37.0%) 5000–7500 68 (39.3%) >7500 41 (23.7%) Residents’ income at store location Lower to lower-middle class (<$58,020) 40 (23.1%) Middle class ($58,021–$94,000) 105 (60.7%) Upper-middle class (>$94,001) 28 (16.2%) NA 3 (1.7%) Race and ethnic composition of residents at store location–mean % (SD) Alaskan Native/Native American 0.5% (0.8) Asian 5.5% (5.0) Black 3.9% (4.5) Hawaiian/Pacific Islander 0.5% (1.6) White 70.2% (12.7) More than one race 4.9% (2.7) Hispanic 14.2% (11.5) Cannabis seller characteristics ( N = 346 visits) Type of primary store staff person Clerk 261 (75.4%) Manager (including owner) 1 (0.3%) Bouncer/security 28 (8.1%) Other (host or cashier) 56 (16.2%) Sex of primary store staff person Female 127 (36.7%) Male 215 (62.1%) Don’t know 4 (1.2%) Age of primary store staff person Mean (std dev) 31.6 (8.1) Range 20–60 Race and ethnicity of primary store staff person Not Hispanic 316 (91.3%) Hispanic 26 (7.5%) Don’t know 4 (1.2%) American Indian/Alaska Native 1 (0.3%) Asian 7 (2.0%) Black/African American 37 (10.7%) Native Hawaiian/Pacific Islander 4 (1.2%) White 272 (68.5%) Other 21 (6.1%) Don’t know 4 (1.2%) Type of other store staff person ( n = 55) Clerk 43 (78.2%) Manager 8 (14.5%) Bouncer/security 0 (0%) Other 4 (7.3%) Sex of other store staff person ( n = 55) Female 23 (41.8%) Male 32 (58.2%) Age of other store staff person ( n = 55) Mean (std dev) 31.7 (8.6) Range 20–70 Race and ethnicity of other store staff person ( n = 55) Not Hispanic 52 (94.5%) Hispanic 2 (3.6%) Don’t know 1 (1.8%) American Indian/Alaska Native 0 (0%) Asian 2 (3.6%) Black/African American 4 (7.3%) Native Hawaiian/Pacific Islander 0 (0%) White 45 (81.2%) Other 3 (5.5%) Don’t know 1 (1.8%) PiP buyer characteristics ( N = 346 visits) Sex Female 151 (43.6%) Male 195 (56.4%) Race and ethnicity Not Hispanic 334 (96.5%) Hispanic 12 (3.5%) Prefer not to answer 0 (0%) American Indian/Alaska Native 3 (0.9%) Asian 0 (0%) Black/African American 33 (9.5%) Native Hawaiian/Pacific Islander 0 (0%) White 275 (79.5%) More than one race 35 (10.1%) Don’t know 0 (0%) Signs of intoxication displayed (multiple responses allowed) Slurred speech 161 (46.5%) Stumbling 145 (41.9%) Swaying 155 (44.8%) Lurching 122 (35.3%) Dropped ID 54 (15.6%) Forgetting things 45 (13.0%) Repeating things 60 (17.3%) Falling asleep 17 (4.9%) Staring/nonresponsive 64 (18.5%) Eye movement 37 (10.7%) Talking too loudly 46 (13.3%) PiP buy intoxication display Obviousness of intoxication display ( N = 339 visits) 1-Very subtle 4 (1.1%) 2 6 (1.8%) 3 39 (11.5%) 4 106 (31.3%) 5-Very obvious 184 (54.3%) Open in a new tab a Visits when buyer was allowed entry to store ( N = 314 visits). b Visits when store staff offered to sell cannabis to buyer ( N = 273 visits). Pseudo-intoxicated patron protocol Compliance with the state law prohibiting the sale of recreational cannabis to apparently intoxicated customers was assessed with a pseudo-intoxicated patron (PiP) protocol used in a previous study of recreational cannabis stores ( Buller et al., 2020 ). Actors were identified by a local casting agency and chosen for prior acting experience from theater, commercial projects, improvisation groups, and standardized simulated patient programs to serve as PiPs. They were female ( n = 15, 48.4%) and male ( n = 16, 51.6%), aged 22–67 years (age-eligible to purchase recreational cannabis) and diverse in race and ethnic composition (3 Hispanic [9.7%], 23 non-Hispanic White [74.2%], 5 African American [16.1%], 1 American Indian/Alaska Native [3.2%], and 1 Another Race [3.2%]). The diversity of the recruited PiP staff was representative of the demographic characteristics of the geographic area where PiP assessments took place in terms of sex, age, and race and ethnicity. Using established training ( Buller et al., 2020 , 2024 ; Saltz et al., 2024 ), the actors were instructed in performing the PiP buyer and observer methods, completing the observer and buyer forms (see below), and responding to store staff, including how to answer whether their legitimacy was questioned and manage complaints from store staff (a letter from the PI explaining the study was provided). Actors could choose not to visit a store that appeared unsafe due to store conditions (e.g., location, lighting, or cleanliness). They were trained to feign signs of a high level of alcohol intoxication, such as swaying, slurred speech, stumbling, dropping an ID, or staring/being unresponsive ( Droste et al., 2018 ; McKnight et al., 1997 ). To obtain realistic displays, actors were trained to display cues at levels that were obvious (displaying at least three intoxication cues) but not so obvious as to be disruptive. They practiced procedures in a group of trainees and had intoxication displays filmed, receiving corrective feedback. The actors self-reported their sex, age, race, and Hispanic ethnicity. Working in two-person teams, PiPs visited recreational cannabis stores twice between January and June 2024 and assessed whether store personnel were willing to sell a cannabis product to a PiP buyer feigning intoxication. The two visits to each store were scheduled at least 6 weeks apart, assigning different actors to prevent store staff from recognizing them. PiP teams were scheduled to visit stores on Thursday, Friday, and Saturday between 4 and 10 pm because stores were most likely to be open on these days and times. At each visit, one trained actor was the PiP buyer and the other was the observer. They entered the store together, with the buyer exhibiting the cues of alcohol intoxication. If they were asked for ID, they provided a valid driver’s license showing that they were 21 or older. If they were refused entrance, the buyer and observer exited the store, and this was noted by the observer. In the store, the buyer continued to show intoxication cues and attempted to purchase an inexpensive prerolled joint. However, the product was not actually purchased by either the PiP buyer or observer in keeping with federal prohibition. Instead, they declined to purchase if the store personnel indicated willingness to sell to them. The PiP buyers were trained to indicate that they had forgotten to bring cash for the purchase and would return later, and observers said they would not purchase the cannabis product for the PiP buyers, if asked by the seller. The observer examined the store environment and noted the buyer’s intoxication cues, the interaction between the buyer and store personnel, and whether the store personnel agreed to sell the cannabis product to the buyer. After the purchase attempt, the buyer and observer exited the store and completed buyer and observer data forms independently, which were checked by supervisors. If an observer rated a PiP buyer’s intoxication display as 1 or 2 on the obviousness scale (rated as very subtle to somewhat subtle intoxication cues), the supervisor met with the PiP buyer and retrained them to achieve ratings of 3 or higher. The buyer and observer forms were programmed in the Qualtrics survey software and completed by buyers and observers on tablet computers, which resulted in very little missing data. However, for a very few initial visits, observers inadvertently completed a buyer form, so a very small amount of data was missing. Responses were uploaded to secure project servers when connected to Wi-Fi. Not being willing to sell to the PiP buyer and failure to gain entry were coded as “refusal to sell.” The protocol and measures were approved by the Western Copernicus Group (WCG) Institutional Review Board. The PiP assessment presented the sellers with the most overt situation in which recreational cannabis sales should be refused (i.e., when a customer is showing clear signs of intoxication), and modeled behavior of customers at risk for polysubstance impairment. Store personnel were blind to PiP assessments. In our experience, PiP buyer behavior is not out of the ordinary or very memorable. The PiP protocol has been used in thousands of visits to alcohol premises ( Buller et al., 2024 ; McKnight, 1993 ; Saltz et al., 2024 ) and cannabis stores ( Buller et al., 2020 ). Measures After each visit, the buyer and observer completed buyer and observer data forms with outcome measures ( Buller et al., 2020 , 2024 ; Saltz et al., 2024 ). The primary outcome measure was willingness to sell recreational cannabis to the PiP buyer, recorded by PiP buyers. Buyers also recorded if ID was checked (which could occur at entry, on the floor, or at the register) and entry was permitted and if store personnel was willing to sell cannabis (yes v. no). A measure of the consistency of willingness to sell was created by combining sales outcome measures of the two visits into a three-level variable—two refusals (consistent refusal), one refusal and one willing to sell (inconsistent refusal), two willing to sell (consistent sale). Beyond willingness to sell, observers also recorded store personnel comments during the visit including when selling: (a) said nothing, (b) commented on buyer’s behavior (joking or seriously), (c) mentioned buyer’s intoxication, (d) enlisted help from other personnel/manager in deciding to sell, (e) tried to arrange transportation for buyer (e.g., call friend, rideshare, or taxi), (f) tried to ignore purchase request, (g) got help from manager/other personnel, (h) cited buyer’s intoxication, (i) cited policy banning sales to intoxicated customers (e.g., state law or house policy), (j) cited another reason, and/or (k) provided no reason (yes v. no). The observer recorded whether the buyer displayed each of the signs of intoxication trained and rated the obviousness of the intoxication cues on a 1 (very subtle) to 5 (very obvious) scale based on number of intoxication cues displayed. Observers only rarely rated PiP buyers as 1 or 2 on this scale ( n = 2 visits) so ratings of 1, 2, and 3 were combined ( n = 42 visits), yielding a three-level obviousness rating, Buyers and observers recorded sex (Male, Female) and race and ethnicity (Race: American Indian/Alaska Native, Asian, Black/African American, Native Hawaiian/Pacific Islander, White, Other, Ethnicity: Hispanic/Latino, Not Hispanic/Latino) of store personnel who interacted with the buyers based on an appraisal of biological features (hair color, skin color, and facial features), language (accent and dialect), and cultural expressions, which was collapsed into 0 = race or ethnic minority versus 1 = non-Hispanic White for analysis. Observers also recorded information about the store environment including (1) how busy the store was with other customers on a 4-point scale (1 = quiet, 2 = average, 3 = busy, and 4 = very busy); (2) store signage, specifically any signs indicating that the sale of recreational cannabis to intoxicated customers is prohibited, as well as signs required by OLCC; and (3) whether the store was clean and well-maintained (yes/no). Neighborhood characteristics were obtained from the US Census, including median income (treated as a continuous variable), population density (categorized as >7500 residents per square mile, 5000–7500 residents per square mile, or <5000 residents per square mile), and race and ethnic composition (proportion of White, Black, and Hispanic residents) for the census tract in which the store was located. Analyses The primary outcome was the willingness to sell cannabis at each visit (0 = no [refusal]; 1 = yes), with the consistency of willingness to sell as a secondary outcome (0 = not willing to sell at both visits, 1 = willing to sell at one visit but refusing to sell at the other visit, 2 = willing to sell at both visits). Initially, descriptive statistics on each outcome were obtained ( Table 1 ). Next, logistic regression analyses in R version 4.4.2 were used to test the relationship between store, neighborhood, cannabis seller, and PiP buyer factors and willingness to sell during each visit, treating visit as a repeated measure within each establishment. Potential predictors were observer ratings of how busy the store appeared, a posted sign stating “no sales to intoxicated customers,” neighborhood characteristics (income, population density, and race and ethnic composition), characteristics of the PiP buyer and store personnel (sex and race and ethnicity), and obviousness of PiP buyer intoxication signs (3-level scale = 1–3, 4, 5). Store personnel characteristics were associated with the refusal of sales of alcohol in past research ( Buvik & Rossow, 2015 ; Saltz et al., 2024 ; Woodall et al., 2018 ). Finally, the interactions between the obviousness of PiP buyer intoxication displays and signage on willingness to sell to intoxicated customers and interactions of race and ethnicity and sex of the buyer and age, race and ethnicity and sex of the seller were analyzed using similar logistic regression models with visits as a repeated measure. These comparisons could not be performed on the consistency of the willingness to sell variable that combined results from the two visits, because PiP buyer and store personnel characteristics varied between the two visits. Also, we did not compare stores on whether they sold medical as well as recreational cannabis products because all stores in Oregon can sell both types of products regardless of whether they emphasize sales of one type of product over another. For all tests, the alpha coefficient was set at 0.05 (two-tailed). RESULTS Profile of sample A sample of 173 recreational cannabis stores was assessed of 189 stores that originally qualified. Of the 16 stores that were not assessed, nine were permanently closed when visited (five at first visit and four at second visit), one was selling cannabis by home delivery only, and one had been purchased by a multistore chain from which another store was already selected. PiP teams did not feel safe entering one store as it appeared to be not well-maintained physically and not well lit, and per our PiP training protocol the store was not assessed. In addition, PiP teams had to reveal the mystery-shopper protocol when questioned by personnel in four stores, so a total of five stores were dropped from the analyses. PiP teams visited each store twice and performed a total of N = 346 assessments. As Table 1 shows, most stores were in urbanized locations with population densities exceeding 5000 persons per square mile but just over one-third were located in more sparsely populated areas with less than 5000 persons per square mile. PiP visits primarily occurred on Thursdays, Fridays, and Saturdays between 4:00 and 9:59 pm. Nine visits occurred on Sunday, Tuesday, or Wednesday because the stores were not open when PiP teams visited on Thursday–Saturday. Stores had few customers in them during the majority of visits, with PiP observers rating them as quiet or average at 86.1% of visits. Store personnel checked IDs in most of the visits (85.5%), either at entry, on the floor, or for the purchase attempt ( Table 1 ). In 28.3% of the visits, a sign was observed stating that there would be no sale of cannabis products to apparently intoxicated customers. In terms of cannabis sellers, PiP buyers interacted predominately with young, male, White individuals serving as clerks. They infrequently interacted with managers, bouncers, or other staff. Of the 346 assessments, just under half were conducted by female PiP buyers and one-quarter by ethnic minority PiP buyers. Intoxication displays Observers confirmed that PiP buyers displayed obvious signs of alcohol intoxication ( Table 1 ). The most common intoxication cues enacted by PiP buyers were slurred speech, stumbling, swaying, and lurching. Less common intoxication cues were dropping ID, forgetting things, repeating things, falling asleep, staring or being nonresponsive, unusual eye movement, and loud talking. Observers rated the obviousness of buyers’ intoxication display as 3 or higher on the 5-point obviousness scale in 97.1% of the visits. Sales of recreational cannabis products to PiP buyers Cannabis sellers were willing to sell recreational cannabis products to PiP buyers at 255 (73.7%) visits, while refusing to sell at 91 (26.3%) visits. Sales were slightly lower in the second visits (sellers were willing to sell at 123 [71.1%] visits and refused at 50 [28.9%] visits) than the first visits (132 [76.3%] visits and 41 [23.7%] visits, respectively). When sales at both visits were combined for each store, cannabis sellers refused to sell to PiP buyers at both visits in only 12 (6.9%) stores but refused to sell to PiP buyers at one of the two visits in 67 (38.7%) stores. They were willing to sell cannabis products to PiP buyers at both visits in over half of the stores (54.3%). During a small number of visits (35 [10.1%] of 346 visits), the store personnel indicated that they were reluctant but still willing to sell cannabis to the PiP buyer and at eight visits (2.3%), they made a joke or similar remark about the buyer’s behavior, while indicating a willingness to sell the cannabis product. Recreational cannabis sales by store, neighborhood, cannabis seller, and PiP buyer characteristics Results of bivariate tests examining willingness to sell cannabis products by store, neighborhood, cannabis seller, and PiP buyer characteristics are presented in Table 2 . Willingness to sell cannabis products was lower in stores posting signs that there would be no sales to intoxicated customers compared to stores without such signage, but other store characteristics were unrelated to willingness to sell (busyness, day and time of visit, and number of warning signs). Neighborhood characteristics (population density, residents’ income, and residents’ ethnicity in the store’s census tract) were unrelated to willingness to sell cannabis to PiP buyers. The age, sex, and race/ethnicity of PiP buyers and cannabis sellers were not associated with willingness to sell to the PiP buyer, nor was the combination of PiP buyer and cannabis seller sex (buyer X staff interaction OR = 0.45, 95% CI [0.16, 1.26], p = 0.128) or race and ethnicity (buyer X staff interaction OR = 0.37, 95% CI [0.09, 1.59], p = 0.181). TABLE 2. Bivariate associations between cannabis store, neighborhood, cannabis seller, and pseudo-intoxicated buyer characteristics with the sale of cannabis to pseudo-intoxicated buyer. No sale (refusal) Sale Logistic regression results Store characteristics How busy store appeared Quiet 53 (27.3%) 141 (72.7%) OR = 1.23, 95% CI [0.91, 1.66], p = 0.179 Average 26 (31.7%) 56 (68.3%) Busy 8 (16.7%) 40 (83.3%) Very busy 1 (12.5%) 7 (87.5%) Time of visit ( N = 346 visits) 4:00–5:59 pm 21 (21.9) 75 (78.1) OR = 0.87, 95% CI [0.69, 1.10], p = 0.243 6:00–6:59 pm 27 (25.2) 80 (74.8) 7:00–7:59 pm 30 (31.9) 64 (68.1) 8:00–9:59pm 13 (26.5) 36 (73.5) “No sale to intoxicated customers” sign posted No 51 (22.8%) 173 (77.2%) OR = 1.77, 95% CI [1.09, 2.88], p = 0.022 * Yes 37 (34.3%) 71 (65.7%) Number of warning signs posted 3 or less 20 (27.8%) 52 (72.2%) OR = 0.98, 95% CI [0.71, 1.36], p = 0.904 4–6 26 (24.1%) 82 (75.9%) 7 or more 42 (27.6%) 110 (72.4%) Neighborhood characteristics Population density at store location (census tract)—persons per square mile <5000 28 (21.9%) 100 (78.1%) OR = 1.16, 95% CI [0.86, 1.58], p = 0.329 5000–7500 40 (29.4%) 96 (70.6%) >7500 22 (26.8%) 60 (73.2%) Residents’ income at store location (census tract) Lower to lower-middle class (<$58,020) 19 (22.6%) 65 (77.4%) OR = 0.89, 95% CI [0.61, 1.31], p = 0.555 Middle class ($58,021–$94,000) 59 (27.6%) 155 (72.4%) Upper-middle class (>$94,001) 13 (26.0%) 37 (74.0%) Cannabis seller characteristics Cannabis seller age 20–29 35 (24.6%) 107 (75.4%) OR = 0.85, 95% CI [0.62, 1.17], p = 0.319 30–39 37 (26.4%) 103 (73.6%) 40 or older 19 (31.7%) 41 (68.3%) Cannabis seller ethnicity Minority 8 (30.8%) 18 (69.2%) OR = 1.25, 95% CI [0.51, 3.07], p = 0.630 Non-Hispanic White 83 (26.3%) 233 (73.7%) Cannabis seller sex Female 29 (22.8%) 98 (77.2%) OR = 1.37, 95% CI [0.81, 2.32], p = 0.241 Male 62 (28.8%) 153 (71.2%) Pseudo-intoxicated buyer characteristics PiP buyer age 20–29 30 (38.5%) 48 (61.5%) OR = 1.08, 95% CI [0.93, 1.26], p = 0.306 30–39 6 (13.3%) 39 (86.7%) 40–49 6 (19.4%) 25 (80.6%) 50–59 20 (22.5%) 69 (77.5%) 60–69 29 (28.2%) 74 (71.8%) PiP buyer ethnicity Minority 3 (25.0%) 9 (75.0%) OR = 1.07, 95% CI [0.28, 4.05], p = 0.917 Non-Hispanic White 88 (26.3%) 246 (73.7%) PiP buyer sex Female 33 (21.9%) 118 (78.1%) OR = 0.66, 95% CI [0.39, 1.12], p = 0.122 Male 58 (29.7%) 137 (70.3%) PiP buyer intoxication display Obviousness of intoxication display Low intoxication (rating = 1–3) 7 (16.7%) 35 (83.3%) OR = 0.66, 95% CI [0.44, 0.98], p = 0.038 * Moderate intoxication (rating=4) 17 (21.3%) 63 (78.8%) High intoxication (rating = 5) 64 (30.5%) 146 (69.5%) Open in a new tab Abbreviation: OR, odds ratio. * p < 0.05. The obviousness of the PiP buyers’ intoxication displays, as rated by the PiP observers, was predictive of lower willingness to sell cannabis products (more refusals), both in the bivariate analyses ( Table 2 ) and in the logistic regression models testing the interaction with signage and sex of cannabis seller ( Table 3 ). Sellers were less likely to be willing to sell to PiP buyers displaying highly obvious signs of intoxication in stores with posted signs saying no sales to intoxicated customers compared to stores without such signs where willingness to sell appeared unaffected by obviousness of intoxication cues ( Table 3 ). There was also a statistically significant interaction of obviousness of intoxication displays and the seller’s sex such that female sellers were less likely to be willing to sell to PiP buyers only when they displayed more obvious signs of intoxication. By contrast, male sellers refused PiP buyers who displayed more obvious signs of intoxication at similar rates to female sellers but refused more PiP buyers who displayed less obvious signs of intoxication more than female sellers ( Table 3 ). TABLE 3. Logistic regression testing effect of obviousness of PiP buyer intoxication display by cannabis store, cannabis seller, and PiP buyer characteristics on sale of cannabis to pseudo-intoxicated buyer. Multivariate logistic regression results Factors modeled No sale (refusal) Sale OR p Store characteristics “No sale to intoxicated customers” sign posted No Low obviousness of intoxication 7 (19.4%) 29 (80.6%) No sale signage 2.48 0.247 Moderate obviousness of intoxication 15 (22.7%) 51 (77.3%) Obviousness of intoxication High obviousness of intoxication 29 (23.8%) 93 (76.2%) Low/moderate vs. high 0.88 0.695 Yes Signage × intoxication 0.19 0.049 * Low obviousness of intoxication 0 (0.0%) 6 (100.0%) Moderate obviousness of intoxication 2 (14.3%) 12 (85.7%) High obviousness of intoxication 35 (39.8%) 53 (60.2%) Cannabis seller characteristics Cannabis seller age 20–29 Seller age Low obviousness of intoxication 2 (10.5%) 17 (89.5%) 30–39 vs. 21–29 1.42 0.492 Moderate obviousness of intoxication 8 (29.6%) 19 (70.4%) 40 or older vs. 21–29 0.83 0.792 High obviousness of intoxication 23 (24.7%) 70 (75.3%) Obviousness of intoxication 30–39 Low/moderate vs. high 0.79 0.578 Low obviousness of intoxication 2 (10.5%) 17 (89.5%) Seller age × intoxication Moderate obviousness of intoxication 7 (19.4%) 29 (80.6%) 30–39 vs. 21–29 0.50 0.260 High obviousness of intoxication 27 (35.1%) 50 (64.9%) 40 or older vs. 21–29 0.78 0.745 40 or older Low obviousness of intoxication 3 (75.0%) 1 (25.0%) Moderate obviousness of intoxication 2 (12.5%) 14 (87.5%) High obviousness of intoxication 14 (37.8%) 23 (62.2%) Cannabis seller ethnicity Minority Low obviousness of intoxication 1 (33.3%) 2 (66.7%) Seller ethnicity Moderate obviousness of intoxication 2 (50.0%) 2 (50.0%) Non-Hispanic White vs. Minority 1.37 0.559 High obviousness of intoxication 5 (31.3%) 11 (68.8%) Obviousness of intoxication Non-Hispanic White Low/moderate vs. high 0.70 0.529 Low obviousness of intoxication 6 (15.4%) 33 (84.6%) Seller ethnicity × intoxication 0.76 0.680 Moderate obviousness of intoxication 15 (20.0%) 60 (80.0%) High obviousness of intoxication 59 (30.9%) 132 (69.1%) Cannabis seller sex Female Low obviousness of intoxication 0 (0%) 12 (100%) Seller sex Moderate obviousness of intoxication 2 (8.7%) 21 (91.3%) Female vs. male 5.67 0.024 * High obviousness of intoxication 26 (28.9%) 64 (71.1%) Obviousness of intoxication Male Low/moderate vs. high 0.76 0.390 Low obviousness of intoxication 7 (23.3%) 23 (76.7%) Seller sex × intoxication 0.19 0.044 * Moderate obviousness of intoxication 15 (26.8%) 41 (73.2%) High obviousness of intoxication 38 (32.5%) 79 (67.5%) PiP buyer characteristics PiP buyer ethnicity Minority Low obviousness of intoxication 0 (0%) 1 (100%) Buyer ethnicity Moderate obviousness of intoxication 0 (0%) 2 (100%) Non-Hispanic White vs. Minority 1.71 0.326 High obviousness of intoxication 3 (33.3%) 6 (66.7%) Obviousness of intoxication Non-Hispanic White Low/moderate vs. high 1.16 0.803 Low obviousness of intoxication 7 (17.1%) 34 (82.9%) Buyer ethnicity × intoxication 0.41 0.181 Moderate obviousness of intoxication 17 (21.8%) 61 (78.2%) High obviousness of intoxication 61 (30.3%) 140 (69.7%) PiP buyer sex Female Low obviousness of intoxication 1 (6.3%) 15 (93.8%) Buyer sex Moderate obviousness of intoxication 8 (17.0%) 39 (83.0%) Female vs. male 2.04 0.126 High obviousness of intoxication 22 (28.2%) 56 (71.8%) Obviousness of intoxication Male Low/moderate vs. high 0.74 0.385 Low obviousness of intoxication 6 (23.1%) 20 (76.9%) Buyer sex × intoxication 0.60 0.363 Moderate obviousness of intoxication 9 (27.3%) 24 (72.7%) High obviousness of intoxication 42 (31.8%) 90 (68.2%) Open in a new tab Note : Each multivariate regression model forecasted the sale of cannabis to pseudo-intoxicated buyers at each obviousness of intoxication cues level moderated by the predictor variable. Each model included the main effect of obviousness and the predictor variable and their interaction term. Abbreviation: OR, odds ratio. * p < 0.05. DISCUSSION Marketplace regulatory policies that prohibit sales of cannabis and alcohol to apparently intoxicated customers aim to reduce problems associated with intoxication, including impaired driving ( Foxcroft, 2014 ) and especially when alcohol and cannabis are combined. This study examined compliance with a statewide prohibition in Oregon’s recreational cannabis market, one of the first US states to legalize sales. Compliance in 2024 (i.e., 26.3% refused sales) had improved somewhat over compliance observed in 2016 and 2018 (15.2% refused sales) ( Buller et al., 2020 ). However, cannabis sellers were still willing to sell to apparently alcohol-intoxicated PiP buyers in nearly three-quarters of the purchase attempts and over half of both visits by PiP buyers to a store (only 12 stores [6.9%] refused at both visits). Observers confirmed that PiP buyers displayed alcohol intoxication cues at moderate-to-high obviousness in nearly all visits, using behaviors identified by the Oregon regulators as signs of intoxication in their mandatory cannabis seller manual that all cannabis sellers are tested on for recreational sales certification ( Oregon Liquor and Cannabis Commission, 2025 ). Thus, sellers should have recognized that PiP buyers appeared intoxicated and refused to sell. In some instances, sellers seemed to detect the PiP intoxication cues by expressing reluctance, joking, or commenting but were still willing to sell. It is possible that cannabis sellers in Oregon may perceive limited deterrence from state regulations. In a 2022 assessment, stores in Portland were less likely to check identifications for age verification than stores in four other states with legalized recreational cannabis sales ( Berg et al., 2023 ). Another analysis found that licensed retail stores in Oregon had variable compliance with restrictions on marketing messages in their social media feeds (e.g., not modeling overconsumption or using youth-focused messages and including warnings on age restrictions and avoiding impaired driving) ( Moreno et al., 2022 ). Whether the lack of deterrence is due to insufficient communication and enforcement from regulators or pressure to sell due to flat and possibly declining revenue for cannabis stores in Oregon ( Dilley et al., 2023 ) is not clear. The increased refusals of sales in recreational cannabis stores that had posted signs indicating that there would be no sales to intoxicated customers may remind cannabis sellers to look for intoxication and refuse apparently intoxicated customers. However, sellers in those stores were less willing to sell mostly to PiP buyers displaying highly obvious intoxication cues. Sellers need to refuse moderately obvious intoxicated customers, too, because consuming both cannabis and alcohol can produce high levels of intoxication, even at low BAC and THC levels ( Ramaekers et al., 2000 ). State regulations did not require that stores post this sign, so store management appeared to choose to post it and make their commitment to this responsible sales practice public. Management support has motivated alcohol servers to practice responsible beverage service ( Ker & Chinnock, 2008 ) and may have this same effect in cannabis markets. Requiring stores to post signage on refusing sales to intoxicated customers may be a low-cost intervention to increase compliance with this state law. Without communication and support from management and systematic training in ways to detect intoxication and refuse sales, many refusals to PiP buyers were likely individual decisions by the cannabis seller. Female sellers may hesitate to refuse customers unless they appear very obviously intoxicated, perhaps because they feel less confident justifying their refusal when cues are less certain and dealing with potential conflict with customers than male sellers. Management of cannabis stores could support responsible sales practices by female sellers, repeatedly stressing the importance of refusing any customer who appears intoxicated, and training staff in responsible sales practices. This speculation should be followed up in future research with cannabis sellers. The low refusal rate may explain the lack of associations with other characteristics of cannabis sellers or PiP buyers (i.e., a basement effect). Comprehensive training requirements and certification of cannabis retail sellers with required retraining periods ( National Academies of Sciences, Engineering, and Medicine, et al., 2024 ), similar to requirements for alcohol servers and sellers in many states ( Alcohol Policy Information System, 2025 ) could provide an important policy method to prevent polysubstance intoxicated driving, its consequences, and other harms. Recent syntheses of the literature indicated that responsible beverage service training is effective in the alcohol market ( Fell et al., 2024 ). The low refusal rate by cannabis sellers was similar to the low rate seen in a US state that had only recently required sellers to complete responsible beverage service training (which was considerably lower than in two other states with long-standing training requirements) ( Buller et al., 2024 ) and low rates in music festivals and sporting arenas in Sweden, which were not required to train alcohol servers in responsible sales and service training (which was lower than at premises outside these events) ( Feltmann et al., 2020 ). This suggests that recreational cannabis outlet staff training, along with management policies, could improve responsible sales and safety for consumers and the community ( Lange et al., 2023 ). As of 2023, only seven of 20 states with legal recreational cannabis sales had some form of training requirements for cannabis sellers, including Oregon. However, Oregon’s training requirement involves just reading an OLCC booklet and passing an examination, rather than multi-hour trainings, some with video and other simulations, which may be more effective at teaching skills to recognize and refuse intoxicated customers. Limitations This study had several strengths. The observational PiP assessment methods were well-established, used twice at each store, and superior to other measures (e.g., prevalence of intoxicated customers is too unpredictable for on-site observation; self-reports by sellers or buyers can be biased). The large sample of stores and blinding of stores and sellers to assessment were strengths. However, generalizability was limited by the study being conducted in only one US state and two metropolitan areas. It may not apply to other states or cities or in highly rural regions (although population density was unrelated to willingness to sell in this study). The omission of most stores in multistore chains did not account for their consistent management practices. We do not believe we missed stores. We were able to confirm that the stores were open using the OLCC license number and address, even if they had moved. A newly licensed store was identified for the sampling during this confirmation process. PiP teams were able to access all stores that were in business. The restricted days and times of assessment may have resulted in assessments occurring during busier periods in the store but ratings of how busy they were during the visits were unrelated to sales behavior. Variation in obviousness of intoxication displays was a limitation. It did appear that the training succeeded in obtaining displays of at least moderate obviousness of alcohol intoxication that should have been recognizable. We also retrained a few PiPs not to enact extreme intoxication. The ratings of obviousness may have rater bias, but we did train observers to look for specific behaviors associated with intoxication when making their ratings. We were unable to assess inter-rater reliability of observers’ obviousness ratings as these ratings were conducted in two-person field teams. Finally, the analysis did not account for any information on the enforcement actions by state or local entities that could affect store compliance rates. Public health implications While the market has expanded rapidly in the past decade, problems associated with legal recreational cannabis sales are emerging, such as polysubstance impairment. Marketplace regulatory interventions that regulate the way that cannabis is sold show promise, but only if licensees and their staff have the perceived deterrence and skills to comply effectively with them. The prohibition of sales to apparently intoxicated customers is long-standing in Oregon’s recreational cannabis market, existing from when it debuted. Thus, sellers should be aware of it. Low enforcement of overservice has been observed in alcohol markets ( Lenk et al., 2023 ) and cannabis regulators may focus more on preventing underage sales rather than on cannabis or alcohol impairment ( Wiens et al., 2018 ). Training could improve cannabis sellers’ ability to recognize signs of intoxication and effectively refuse sales, as research on responsible beverage service training has confirmed ( Gripenberg et al., 2007 , 2011 ; Woodall et al., 2018 ). FUNDING INFORMATION This research was supported by a grant from the National Institutes of Health/National Institute on Alcohol Abuse and Alcoholism, 1 R01 AA031591-01A1. Dr. Woodall, Dr. Buller, Ms. Martinez, Ms. Brice, and Mr. Chirico receive a salary from Klein Buendel, Inc. Dr. Buller’s spouse is an owner of Klein Buendel, Inc. Dr. Cutter is employed by the University of Alabama at Birmingham and reports personal fees from Pythagoras Board membership, advisory board fees from Alexion, Antisense Therapeutics, Avotres, Biogen, Clene Nanomedicine, Clinical Trial Solutions LLC, Entelexo Biotherapeutics, Inc., Genzyme, Genentech, GW Pharmaceuticals, Hoya Corporation, Immunic, Immunosis Pty Ltd, Klein-Buendel Incorporated, Linical, Merck/Serono, Novartis, Perception Neurosciences, Protalix Biotherapeutics, Regeneron, Roche, SAB Biotherapeutics, and data and safety monitoring board fees from Applied Therapeutics, AI therapeutics, AMO Pharma, Astra-Zeneca, Avexis Pharmaceuticals, Bristol Meyers Squibb/Celgene, CSL Behring, Horizon Pharmaceuticals, Immunic, Karuna Therapeutics, Kezar Life Sciences, Mapi Pharmaceuticals LTD, Merck, Mitsubishi Tanabe Pharma Holdings, Opko Biologics, Prothena Biosciences, Novartis, Regeneron, Sanofi-Aventis, Reata Pharmaceuticals, Teva Pharmaceuticals, NHLBI (Protocol Review Committee), University of Texas Southwestern, University of Pennsylvania, Visioneering Technologies, Inc. outside the submitted work. Footnotes CONFLICT OF INTEREST STATEMENT Dr. Saltz and Mr. Fell have no conflicts. REFERENCES Adhikari K, Maas A. & Trujillo-Barrera A. (2023) Revisiting the effect of recreational marijuana on traffic fatalities. International Journal of Drug Policy, 115, 104000. Available from: 10.1016/j.drugpo.2023.104000 [ DOI ] [ PubMed ] [ Google Scholar ] Alcohol Policy Information System. (2025) Beverage service training and related practices. Available from: https://alcoholpolicy.niaaa.nih.gov/apis-policy-topics/beverage-service-training-and-related-practices/26 [Accessed 6th August 2025]. Asbridge M, Hayden JA & Cartwright JL (2012) Acute cannabis consumption and motor vehicle collision risk: systematic review of observational studies and meta-analysis. BMJ, 344, e536. Available from: 10.1136/bmj.e536 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Aydelotte JD, Mardock AL, Mancheski CA, Quamar SM, Teixeira PG, Brown CVR et al. (2019) Fatal crashes in the 5 years after recreational marijuana legalization in Colorado and Washington. Accident Analysis & Prevention, 132, 105284. Available from: 10.1016/j.aap.2019.105284 [ DOI ] [ PubMed ] [ Google Scholar ] Barry RA & Glantz SA (2018) Marijuana regulatory frameworks in four US states: an analysis against a public health standard. American Journal of Public Health, 108(7), 914–923. Available from: 10.2105/AJPH.2018.304401 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Berg CJ, Romm KF, Pannell A, Sridharan P, Sapra T, Rajamahanty A. et al. (2023) Cannabis retailer marketing strategies and regulatory compliance: a surveillance study of retailers in 5 US cities. Addictive Behaviors, 143, 107696. Available from: 10.1016/j.addbeh.2023.107696 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Brubacher JR, Chan H, Erdelyi S, Macdonald S, Asbridge M, Mann RE et al. (2019) Cannabis use as a risk factor for causing motor vehicle crashes: a prospective study. Addiction, 114(9), 1616–1626. Available from: 10.1111/add.14663 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Buller DB, Woodall WG, Saltz R, Grayson A, Svendsen S. & Cutter GR (2020) Sales to apparently alcohol-intoxicated customers and online responsible vendor training in recreational cannabis stores in a randomized trial. International Journal of Drug Policy, 83, 102860. Available from: 10.1012/j.drugpo.2020.102860 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Buller DB, Woodall WG, Saltz R, Martinez L, Small A, Chirico N. et al. (2024) Sales to apparently intoxicated customers in three states with different histories of responsible beverage service training. Journal of Studies on Alcohol and Drugs, 85, 312–321. Available from: 10.15288/jsad.23-00258 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Buvik K. & Rossow I. (2015) Factors associated with over-serving at drinking establishments. Addiction, 110(4), 602–609. Available from: 10.1111/add.12843 [ DOI ] [ PubMed ] [ Google Scholar ] Calvert C. & Erickson D. (2020) An examination of relationships between cannabis legalization and fatal motor vehicle and pedestrian-involved crashes. Traffic Injury Prevention, 21, 1–6. Available from: 10.1080/15389588.2020.1810246 [ DOI ] [ PubMed ] [ Google Scholar ] Cherpitel CJ, Ye Y, Andreuccetti G, Stockwell T, Vallance K, Chow C. et al. (2017) Risk of injury from alcohol, marijuana and other drug use among emergency department patients. Drug and Alcohol Dependence, 174, 121–127. Available from: 10.1016/j.drugalcdep.2017.01.019 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Compton RP (2017) Marijuana-impaired driving – a report to congress. Report No. DOT HS 812 440. Washington, DC: National Highway Traffic Safety Adminstration. Available from: https://www.nhtsa.gov/sites/nhtsa.gov/files/documents/812440-marijuana-impaired-driving-report-to-congress.pdf [Accessed 6th August 2025]. [ Google Scholar ] Couper FJ & Peterson BL (2014) The prevalence of marijuana in suspected impaired driving cases in Washington state. Journal of Analytical Toxicology, 38(8), 569–574. Available from: 10.1093/jat/bku090 [ DOI ] [ PubMed ] [ Google Scholar ] Dilley JA, Johnson JK, Colby AM, Sheehy TJ, Muse EJ, Filley JR et al. (2023) Cannabis retail market indicators in five legal states in the United States: a public health perspective. Clinical Therapeutics, 45(8), 778–786. Available from: 10.1016/j.clinthera.2023.06.015 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Droste N, Miller P, Kaestle CE, Curtis A, Hyder S, Coomber K. et al. (2018) Comparing levels of blood alcohol concentration and indicators of impairment in nightlife patrons. Drug and Alcohol Review, 37(Suppl 1), S348–S356. Available from: 10.1111/dar.12639 [ DOI ] [ PubMed ] [ Google Scholar ] Dubois S, Mullen N, Weaver B. & Bedard M. (2015) The combined effects of alcohol and cannabis on driving: impact on crash risk. Forensic Science International, 248, 94–100. Available from: 10.1016/j.forsciint.2014.12.018 [ DOI ] [ PubMed ] [ Google Scholar ] Elgán TH, Durbeej N, Holder HD & Gripenberg J. (2018) Overserving and allowed entry of obviously alcohol-intoxicated spectators at sporting events. Alcoholism, Clinical and Experimental Research, 42(2), 444–452. Available from: 10.1111/acer.13563 [ DOI ] [ PubMed ] [ Google Scholar ] Elvik R. (2013) Risk of road accident associated with the use of drugs: a systematic review and meta-analysis of evidence from epidemiological studies. Accident Analysis and Prevention, 60, 254–267. Available from: 10.1016/j.aap.2012.06.017 [ DOI ] [ PubMed ] [ Google Scholar ] Farmer CM, Monfort SS & Woods AN (2022) Changes in traffic crash rates after legalization of marijuana: results by crash severity. Journal of Studies on Alcohol and Drugs, 83(4), 494–501. Available from: 10.15288/jsad.2022.83.494 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Fell JC, McKnight AS & Auld-Owens A. (2013) Increasing impaired-driving enforcement visibility: six case studies. Report No. DOT HS 811 716. Washington, DC: National Highway Traffic Safety Administration. Available from: https://www.nhtsa.gov/sites/nhtsa.gov/files/811716.pdf [Accessed 6th August 2025]. [ Google Scholar ] Fell JC, Scolese J. & Achoki T. (2024) The effectiveness of responsible beverage service training programs: a literature review and synthesis. Journal of Alcoholism, Drug Abuse and Substance Dependence, 10(1), 036. Available from: 10.24966/ADSD-9594/100036 [ DOI ] [ Google Scholar ] Feltmann K, Gripenberg J. & Elgán TH (2020) Compliance to the alcohol law: overserving to obviously intoxicated visitors at music festivals. International Journal of Environmental Research and Public Health, 17(22), 8699. Available from: 10.3390/ijerph17228699 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Foxcroft DR (2014) Can prevention classification be improved by considering the function of prevention? Prevention Science, 15(6), 818–822. Available from: 10.1007/s11121-013-0435-1 [ DOI ] [ PubMed ] [ Google Scholar ] Gosselt JF, Van Hoof JJ, Goverde MM & De Jong MD (2013) One more beer? Serving alcohol to pseudo-intoxicated guests in bars. Alcoholism, Clinical and Experimental Research, 37(7), 1213–1219. Available from: 10.1111/acer.12074 [ DOI ] [ PubMed ] [ Google Scholar ] Gripenberg J, Abdon J, Wallin E. & Andreasson S. (2011) Long-term effects of a community-based intervention: 5-year follow-up of ‘clubs against drugs’. Addiction, 106(11), 1997–2004. Available from: 10.1111/j.1360-0443.2011.03573.x [ DOI ] [ PubMed ] [ Google Scholar ] Gripenberg J, Wallin E. & Andreasson S. (2007) Effects of a community-based drug use prevention program targeting licensed premises. Substance Use & Misuse, 42(12–13), 1883–1898. Available from: 10.1080/10826080701532916 [ DOI ] [ PubMed ] [ Google Scholar ] Guttmannova K, Lee CM, Kilmer JR, Fleming CB, Rhew IC, Kosterman R. et al. (2016) Impacts of changing marijuana policies on alcohol use in the United States. Alcoholism, Clinical and Experimental Research, 40(1), 33–46. Available from: 10.1111/acer.12942 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Hartman RL, Brown TL, Milavetz G, Spurgin A, Gorelick DA, Gaffney G. et al. (2015) Controlled cannabis vaporizer administration: blood and plasma cannabinoids with and without alcohol. Clinical Chemistry, 61(6), 850–869. Available from: 10.1373/clinchem.2015.238287 [ DOI ] [ PubMed ] [ Google Scholar ] Hughes T. (2014) Colo. kids getting into their parents’ pot-laced goodies. USA Today, 2 April. Available from: http://www.usatoday.com/story/news/nation/2014/04/02/marijuana-pot-edibles-colorado/7154651/ [Accessed 6th August 2025]. [ Google Scholar ] Jackson KM, Sokolovsky AW, Gunn RL & White HR (2020) Consequences of alcohol and marijuana use among college students: prevalence rates and attributions to substance-specific versus simultaneous use. Psychology of Addictive Behaviors, 34(2), 370–381. Available from: 10.1037/adb0000545 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Kamer RS, Warshafsky S. & Kamer GC (2020) Change in traffic fatality rates in the first 4 states to legalize recreational marijuana. JAMA Internal Medicine, 180(8), 1119–1120. Available from: 10.1001/jamainternmed.2020.1769 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Ker K. & Chinnock P. (2008) Interventions in the alcohol server setting for preventing injuries (review). Cochrane Database of Systematic Reviews, 16(3), CD005244. Available from: 10.1002/14651858.CD005244.pub2 [ DOI ] [ Google Scholar ] Lange JE, Moyers SA & Croff JM (2023) Responsible cannabis sales: a narrative review considering interventions for dispensary staff to address cannabis harms. Public Administration Quarterly, 47(3), 374–391. Available from: 10.37808/paq.47.3.7 [ DOI ] [ Google Scholar ] Lenk KM, Scholz N, Erickson DJ, Joshi S, Toomey TL, Jones-Webb R. et al. (2023) Alcohol enforcement in the United States from 2010 to 2019. Journal of Studies on Alcohol and Drugs, 84(3), 416–423. Available from: 10.15288/jsad.22-00096 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Lira MC, Heeren TC, Buczek M, Blanchette JG, Smart R, Pacula RL et al. (2021) Trends in alcohol and cannabis involvement in motor vehicle crash fatalities in the United States from 2000–2018: a repeated cross-sectional study. American Journal of Public Health, 111(11), 1976–1985. Available from: 10.2105/AJPH.2021.306466 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] LoParco CR, Cui Msph Y, McCready Ms D, Romm Phd KF, Yang Sc DLMYT, Rossheim Ph DMME et al. (2024) State requirements for non-medical US cannabis retail personnel. Journal of Public Health Management and Practice, 30(4), 517–525. Available from: 10.1097/phh.0000000000001892 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Marcotte TD & Fitzgerald RL (2025) Validation of methods for detecting driving under the influence of cannabis: paths forward. Journal of Studies on Alcohol and Drugs, 86(4), 510–514. Available from: 10.15288/jsad.25-00110 [ DOI ] [ PubMed ] [ Google Scholar ] Marillier M. & Verstraete AG (2019) Driving under the influence of drugs. Wiley Interdisciplinary Reviews: Forensic Science, 1(3), e1326. Available from: 10.1002/wfs2.1326 [ DOI ] [ Google Scholar ] Marinello S. & Powell LM (2023) The impact of recreational cannabis markets on motor vehicle accident, suicide, and opioid overdose fatalities. Social Science & Medicine, 320, 115680. Available from: 10.1016/j.socscimed.2023.115680 [ DOI ] [ PubMed ] [ Google Scholar ] McKnight AJ (1993) Server intervention: accomplishments and needs. Alcohol Health and Research World, 17(1), 76–83. [ Google Scholar ] McKnight AJ, Langston EA, Marques PR & Tippetts AS (1997) Estimating blood alcohol level from observable signs. Accident Analysis and Prevention, 29(2), 247–255. Available from: 10.1016/S0001-4575(96)00079-6 [ DOI ] [ PubMed ] [ Google Scholar ] Moreno MA, Jenkins M, Binger K, Kelly L, Trangenstein PJ, Whitehill JM et al. (2022) A content analysis of cannabis company adherence to marketing requirements in four states. Journal of Studies on Alcohol and Drugs, 83(1), 27–36. Available from: 10.15288/jsad.2022.83.27 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Mosher JAH, Carmona M, Treffers R, Reitz D, Curtis RR, Moore A. et al. (2009) Legal research report: laws prohibiting alcohol sales to intoxicated persons. Washington, DC: National Highway Traffic Safety Administration. Available from: https://www.nhtsa.gov/sites/nhtsa.gov/files/811142.pdf [Accessed 6th August 2025]. [ Google Scholar ] National Academies of Sciences, Engineering, and Medicine, Health and Medicine Division, Board on Population Health and Public Health Practice & Committee on the Public Health Consequences of Changes in the Cannabis Policy Landscape. (2024) Cannabis policy impacts public health and health equity. Washington, DC: National Academies Press (US). Available from: 10.17226/27766 [ DOI ] [ Google Scholar ] National Highway Traffic Safety Administration. (2017) Digest of impaired driving and selected beverage control laws, 30th edition, December 31, 2015. Report No. DOT HS 812 394. Available from: https://www.nhtsa.gov/sites/nhtsa.gov/files/documents/812394-digest-of-impaired-driving-and-selected-beverage-control-laws.pdf [Accessed 6th August 2025]. Oregon Liquor and Cannabis Commission. (2025) Recreational marijuana laws and rules. Available from: https://www.oregon.gov/olcc/marijuana/Pages/Recreational-Marijuana-Laws-and-Rules.aspx#Oregon_Revised_Statutes_-_Cannabis_Regulation [Accessed 8th August 2025]. Oregon Revised Statutes. (2023) ORS 475C.329 – Prohibition regarding person who is visibly intoxicated; penalty, in Oregon Revised Statutes, Chapter 475C: Cannabis Regulation. Available from: https://oregon.public.law/statutes/ors_475c.329 [Accessed 6th August 2025]. Quigg Z, Butler N, Hughes K. & Bellis MA (2022) Effects of multicomponent programmes in preventing sales of alcohol to intoxicated patrons in nightlife settings in the United Kingdom. Addictive Behaviors Reports, 15, 100422. Available from: 10.1016/j.abrep.2022.100422 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Ramaekers JG, Robbe HW & O’Hanlon JF (2000) Marijuana, alcohol and actual driving performance. Human Psychopharmacology, 15(7), 551–558. Available from: 10.1002/1099-1077(200010)15:7<551::aid-hup236>3.0.co;2-p [ DOI ] [ PubMed ] [ Google Scholar ] Reed JK (2018) Impacts of marijuana legalization in Colorado: a report pursuant to Senate Bill 13–283. Denver, CO: Colorado Department of Public Safety, Office of Research and Statistics. Available from: http://cdpsdocs.state.co.us/ors/docs/reports/2018-SB13-283_Rpt.pdf [Accessed 6th August 2025]. [ Google Scholar ] Romero-Martínez Á, Lila M. & Moya-Albiol L. (2019) Long-term drug misuse increases the risk of cognitive dysfunctions in intimate partner violence perpetrators: key intervention targets for reducing dropout and reoffending. International Journal of Environmental Research and Public Health, 16(20), 3792. Available from: 10.3390/ijerph16203792 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Saltz R, Paschall MJ, O’Hara S, Buller DB, Woodall WG & Martinez L. (2024) Serving alcohol to an “obviously intoxicated” patron. Journal of Studies on Alcohol and Drugs, 85(2), 168–174. Available from: 10.15288/jsad.23-00253 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Santaella-Tenorio J, Wheeler-Martin K, DiMaggio CJ, Castillo-Carniglia A, Keyes KM, Hasin D. et al. (2020) Association of recreational cannabis laws in Colorado and Washington state with changes in traffic fatalities, 2005–2017. JAMA Internal Medicine, 180(8), 1061–1068. Available from: 10.1001/jamainternmed.2020.1757 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Schaeffer K. (2024) 9 facts about Americans and marijuana. Available from: https://www.pewresearch.org/short-reads/2024/04/10/facts-about-marijuana/ [Accessed 6th August 2025]. [ Google Scholar ] Sevigny EL (2021) Cannabis and driving ability. Current Opinion in Psychology, 38, 75–79. Available from: 10.1016/j.copsyc.2021.03.003 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Tefft BC & Arnold LS (2020) Cannabis use among drivers in fatal crashes in Washington state before and after legalization. Washington, DC: AAA Foundation for Traffic Safety. Available from: https://aaafoundation.org/cannabis-use-among-drivers-in-fatal-crashes-in-washington-state-before-and-after-legalization/ [Accessed 6th August 2025]. [ Google Scholar ] Thomas FD, Berning A, Darrah J, Graham L, Blomberg R, Griggs C. et al. (2020) Drug and alcohol prevalence in seriously and fatally injured road users before and during the COVID-19 public health emergency. Report No. DOT HS 813 018. Washington, DC: National Highway Traffic Safety Administration. Available from: https://rosap.ntl.bts.gov/view/dot/50941 [Accessed 6th August 2025]. [ Google Scholar ] Toomey TL, Erickson DJ, Lenk KM & Kilian GR (2008) Likelihood of illegal alcohol sales at professional sport stadiums. Alcoholism, Clinical and Experimental Research, 32(11), 1859–1864. Available from: 10.1111/j.1530-0277.2008.00770.x [ DOI ] [ PubMed ] [ Google Scholar ] Toomey TL, Lenk KM, Nederhoff DM, Nelson TF, Ecklund AM, Horvath KJ et al. (2016) Can obviously intoxicated patrons still easily buy alcohol at on-premise establishments? Alcoholism, Clinical and Experimental Research, 40(3), 616–622. Available from: 10.1111/acer.12985 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] University of Montana and Washington Traffic Safety Commission. (2018) Driving under the influence of cannabis and alcohol: key findings report from 2018 Survey. Available from: http://wtsc.wa.gov/wp-content/uploads/2018/12/Driving-Under-the-Influence-of-Cannabis-and-Alcohol-Key-Findings-Report-from-2018-Survey.pdf [Accessed 6th August 2025]. [ Google Scholar ] Wardell JD, Egerton GA & Read JP (2020) Does cannabis use predict more severe types of alcohol consequences? Longitudinal associations in a 3-year study of college students. Alcoholism, Clinical and Experimental Research, 44(5), 1141–1150. Available from: 10.1111/acer.14320 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Washington Traffic Safety Commission. (2015) Marijuana increased in 2014 as a factor in deadly crashes. Olympia, WA: Washington Traffic Safety Commission. Available from: https://media.npr.org/assets/news/2015/08/crashes.pdf [Accessed 6th August 2025]. [ Google Scholar ] Wiens T, Lenk KM, Fabian LEA & Erickson DJ (2018) Law enforcement practices in the first two states in U.S. to legalize recreational marijuana. International Journal of Drug Policy, 61, 38–43. Available from: 10.1016/j.drugpo.2018.08.018 [ DOI ] [ PubMed ] [ Google Scholar ] Woodall WG, Starling R, Saltz RF, Buller DB & Stanghetta P. (2018) Results of a randomized trial of web-based retail onsite responsible beverage service training: WayToServe. Journal of Studies on Alcohol and Drugs, 79(5), 672–679. Available from: 10.15288/jsad.2018.79.672 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Yurasek AM, Aston ER & Metrik J. (2017) Co-use of alcohol and cannabis: a review. Current Addiction Reports, 4(2), 184–193. Available from: 10.1007/s40429-017-0149-8 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] ACTIONS View on publisher site PDF (168.1 KB) Cite Collections Permalink PERMALINK Copy RESOURCES Similar articles Cited by other articles Links to NCBI Databases Cite Copy Download .nbib .nbib Format: AMA APA MLA NLM Add to Collections Create a new collection Add to an existing collection Name your collection * Choose a collection Unable to load your collection due to an error Please try again Add Cancel Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top

Related documents

• The Effect of Medical Marijuana Laws on Marijuana, Alcohol, and Hard Drug Use#21142
• The Business Ethics of Recreational Marijuana#777465
• Ethical tensions in legally mandated substance use treatment: A scoping review of participant perspectives.#367898
• Present Conditions of Marijuana Regulation in USA: Medical and Recreational Use#22436
• Gun Rights and Noncompliance: Two Problems of Prohibition#741097