It has been approximately 12 years since recreational marijuana was legalized in Colorado. There were a handful of early adopters that followed, but recreational marijuana remained the exception. Until now.
From 2020 to 2023, fifteen states legalized recreational marijuana. Many others have legalized medicinal marijuana. Currently, thirteen states have recreational marijuana bills at various stages of the legislative process. Legalization has expanded far beyond the usual suspects and become accepted in states with reputations for being far more conservative, and far less permissive. Recreational marijuana has been legalized in South Dakota and this author’s home state, Missouri. There is legislation under consideration in states such as Iowa, Kansas, Nebraska, West Virginia, and Mississippi. According to a Pew Research Center study completed earlier this year, a whopping 88% of U.S. adults at least support medicinal use of marijuana, and 57% support legalized recreational marijuana.
For those like me residing in a conservative mid-western state, I have largely ignored legalized marijuana, believing it would never come to Missouri. Now that it has, and seeing the pace of legislation, it is readily apparent that I, much like the residents of similar states, cannot ignore legalized marijuana any longer. If it is not legal in your state today, just wait.
This new reality for many clearly indicates a need to revisit the impacts of legalized marijuana. We can, and should, turn to ALFA members from our early adopter states for lessons learned, and to identify key impacts of legalized marijuana on our industry. And there are many.
Employment Matters
Should employers require employees to be THC free?
A 2023 study from the National Institute of Health found a 12.9% increase in work-place injuries associated with the adoption of recreational marijuana. At the same time, it appears our industries employ the highest percentage of recreational marijuana users. A 2018 Colorado study found restaurant and hotel employees were the highest users, checking in at 30.8%, and the next highest employer segment was arts, recreation, and entertainment, at 28%. At the same time, the U.S. is experiencing a labor shortage, and a study from the U.S. Chamber of Commerce indicates the December 2023 quit rate hit the leisure and hospitality industries the hardest (4.3%), followed by the wholesale/retail industries (2.5%).
Faced with the growing ubiquity of legalized marijuana, the use rate of our industry’s employees, and the shrinking labor market, can hospitality and retail afford to insist on THC free employees?
One concern is potential liability for adverse employment decisions if a drug-free policy is adopted. Patients who use prescription medications often have recourse under the ADA or state disability acts if they are discriminated against for using their medicine. Employees or potential employees who are using medical marijuana may similarly assert a claim for discrimination based on a disability. At the federal level, the argument is a complete nonstarter. Courts have universally found that the ADA does not apply to medical marijuana since its use remains illegal under the Controlled Substances Act (21 U.S.C. 812). Noffsinger v. SSC Niantic Operating Co. LLC, 273 F. Supp. 3d 326, 338 (D. Conn. 2017); James v. City of Costa Mesa, 700 F.3d 394, 403 (9th Cir. 2012); Steele v. Stallion Rockies, Ltd., 106 F. Supp. 3d 1205, 1212 (D. Colo. 2015); Forest City Residential Mgmt. v. Beasley, 71 F. Supp. 3d 715, 731 (E.D. Mich. 2014).
The outcome is more mixed at the state level. In at least 17 states, either the legalizing statute provides explicit employment protections or courts have ruled in favor of employee protection. See e.g., Chance v. Kraft Heinz Foods Co., 2018 Del.Super. Lexis 1773 (Dec. 17, 2018)(interpreting Del.Code Ann. tit. 16, 4901A(g) as affording protection from employment discrimination based on an employee’s status as a medical marijuana patient); Me.Rev.Stat.Ann. tit. 22, 2423-E(1); Barbuto v. Advantace Sales and Marketing, LLC, 78 N.E.2d 37 (Mass. 2017); 35 P.S. 10231.2103(b)(1)(stating: “no employer may discharge, threaten, refuse to hire or otherwise discriminate or retaliate against an employee regarding an employee’s compensation, terms, conditions, location or privileges solely on the basis of such employee’s status as an individual who is certified to use medical marijuana.”) Some states may provide protection for medical marijuana patients but allow employers to make decisions in the context of safety sensitive positions. Okla. Stat. tit. 63, Sec. 420.
Other states have found accommodations are not required under fair employment acts. See e.g., Ross v. RagingWire Telecommunications, Inc., 42 Cal.4th 920 (2008)(finding the California Compassionate Use Act of 1996 only afforded protection from criminal prosecution, it did not require employer accommodations); Coats v. Dish Network, LLC, 303 P.3d 147 (Colo. 2015)(ruling against a paralyzed patient who sued after being terminated for off-hours medical marijuana use); Lambdin v. Marriott Resorts Hosp. Corp., 2017 U.S.Dist.Lexis 149570 (D.Hawaii 2019); Emerald Steel Fabricators, Inc. v. BOLI, 238 P.3d 518 (Ore. 2010).
As one might anticipate, however, even in those states where accommodations might be required, employers are protected from adverse employment decisions that arise out of an employee’s possession or impairment at work.
Negligent Supervision/Negligent Entrustment
Even where use of marijuana is legal, injuries by an employee intoxicated by marijuana is no different than injuries caused by employees intoxicated by alcohol. Case law abounds finding employers liable under a theory of negligent supervision of an employee who causes injuries while intoxicated. Negligent supervision is generally deemed distinct from negligent service or Dram Shop liability. See e.g., Widdows v. Crown St. Bar Ltd. P’Ship, 2008 Conn.Super.LEXIS 153 (January 10, 2008)(allowing a claim for negligent supervision to proceed when an intoxicated employee struck a patron with a beer bottle); Hays v. Royer, 384 S.W.3d. 330 (Mo.App. 2012)(holding an employer liable for entrusting a vehicle to a person known to be habitually intoxicated).
This begs the question, what must employers do to ensure its employees are not intoxicated by marijuana? One answer might be training supervisory employees in identifying signs of workplace impairment. Trainings are available to assist in identify workplace impairment.
But while there is overlap between symptoms of alcohol intoxication and marijuana intoxication, research reflects identifying marijuana intoxication is far more difficult. Adding to the challenge is that the signature odor associated with marijuana use is not as prominent as it once was. If there is any odor at all, many THC products are largely odorless (THC infused soft drinks, teas, alcoholic drinks, candies, edibles, and vape pens). Police forces often employe Drug Recognition Experts (DREs) who are employed to recognize impairment in drivers under the influence of drugs. While DREs are consistently reliable in identifying alcohol-impaired drivers, DREs are more variable in their ability to correctly identify THC-impaired individuals. Celeste M. A judicial perspective on expert testimony in marijuana driving cases. J Med Toxicol. (2017) 13:117–23. 10.1007/s13181-016-0579-z. Tests are being developed, but level of impairment after smoking weed is still hard to measure. Chicago Tribune. (2019); Stelter RL, Kupersmidt JB, Brodar K, Eisensmith S. The prevention of drugged driving: needs, barriers, and self-efficacy of prevention professionals. J Prim Prev. (2019) 40:13. 10.1007/s10935-019-00555-2. Although DRE testimony is generally deemed admissible, Courts have limited the scope of the opinion the officer may offer. Specifically, the officer may only opine that their evaluation is “consistent with” drug usage, not that the subject was intoxicated. State v. Olenowski, 304 A.3d 598 (N.J. 2023).
Toxicology
Unlike alcohol impairment, which can be measured through a blood test and by law enforcement at the roadside with a Breathalyzer machine, there is currently no effective way to measure impairment from THC. A lab test can show the level of THC in a person’s blood, but impairment is not directly correlated to THC levels.
The primary difference between alcohol and THC is that alcohol dissolves in water, whereas THC is fat soluble, and dissolves in fat. When one drinks, alcohol spreads through saliva and breath. It evenly saturates lungs and blood. Thus, measuring the volume of alcohol in one part of the body can predictably indicate how much is in any other part of your body. This has allowed for reliable research to study and help formulate the 0.08 BAC in alcohol.
In contrast, the height of intoxication is not the moment when blood THC levels peak, and the high does not rise and fall uniformly based on how much THC leaves and enters bodily fluids. For example, THC rapidly clears out of the blood in occasional users within a couple of hours. While still high, a trickle of THC leaches out of their brains and other fatty tissue back into the blood until it is all gone. Thus, lab tests only find trace amounts of THC in the blood of occasional smokers after a few hours. This means an occasional smoker could still be intoxicated, despite having a low level of THC in their blood.
Daily users are different. They build up so much THC in their body fat that it continues leaching over weeks after last use, which means they have a constant, moderate level of blood THC, even when not high. In short, the elimination rate of THC from one user to another is variable and, therefore, tying a specific THC level to intoxication is not reliable .
Also, the THC delivery mechanism can dramatically affect THC in the blood. If weed is eaten instead of being smoked, the user’s blood never carries that much THC.
Tolerance is also a major factor. THC intoxication does not necessarily produce readily reliable symptoms and behavior patterns, particularly for heavy users.
Another key factor is the THC concentration of the product can vary dramatically. As such, the mere fact of marijuana use, proximate in time, may provide little guidance as to its intoxicating effect. Not all THC products have uniform percentages of THC.
These differences from alcohol make it extraordinarily hard to develop an analog to BAC, i.e., a biological measurement for marijuana intoxication.
Despite the absence of an accurate biological measurement, many states have adopted per-se limits for intoxication. Illinois, Montana, Nevada, Ohio, Pennsylvania, Washington, and West Virginia have cutoffs that range from 1 to 5 ng/ml. Are Court’s allowing evidence of THC at levels in excess of the per se levels into evidence without more from experts? Alternatively, what additional foundation are Courts requiring to introduce a THC level into evidence to establish intoxication?
The unreliability of the data creates a major challenge to the admissibility of evidence of marijuana use at trial. For example, in Secrist v. Treadstone, LLC, 356 S.W.3d 276 (Mo.App. 2011), the Missouri Court of Appeals found the trial court erred in admitting a worker’s THC blood levels of 50ng/ml to show his comparative fault. To be admissible, the Court held the jury must be given an indication as to: 1) what effect that level of marijuana would reasonably have on the subject, 2) whether the subject’s behavior on the day of the accident was consistent with marijuana impairment, and 3) the temporal proximity between ingestion of the drug and the accident.
Meeting the Missouri standard is far more challenging than it might appear. As set forth above, the research does not support a level of THC that produces reasonably predictable effects and, moreover, intoxicated persons do not reliably demonstrate behavior or outward manifestations of their intoxication.
Thus, in Schultz v. Great Plains Trucking, Inc., 2024 Mo.App.Lexis 195 (March 26, 2024), a trial court’s exclusion of evidence of a party’s blood draw that reflected the presence of THC at 3.5ng/ml was upheld. In that case, the doctor admitted one “cannot draw a relationship between the degree of impairment and blood THC concentration.” The doctor’s opinion on impairment was based in part on the party’s testimony that she smoked two hits of marijuana resin approximately 1 hour before the accident. But the doctor admitted she could not speak to the THC concentration of the resin, or that the THC found in the party’s blood was from the resin, as opposed to historic ingestion of THC.
Impacts on Dram Shop Liability
The impact of marijuana use on dram shop liability may vary from state to state, depending on the language of the state’s Dram Shop Act. For example, in Alaska, civil liability may arise from providing alcohol “to a drunken person.” Alaska Stat. 04.21.020. But a person who is high is not “drunken.” Does this mean a Dram Shop may serve, without repercussions, an apparently THC intoxicated person?
Other states premise liability based on a “visibly intoxicated” or “clearly intoxicated” standard, which does not distinguish between alcohol or drug intoxication. Ark. Code 16-126-103; Mo. Rev. Stat. 537.053. Because of the variability of the effect of THC on persons, a recent study in the Journal of the American Medical Association found that trained law enforcement did not reliably identify intoxicated subjects using field sobriety tests. Marcotte, Thomas, et al., Evaluation of Field Sobriety Tests for Identifying Drivers Under the Influence of Cannabis, JAMA Psychiatry, 2023 Sep; 80(9): 914-923. This being true, how can the average server be expected to identify signs of intoxication?
And then there is Illinois, where a dram shop may be liable if it causes the intoxication of the person who causes injury to a third-person. 235 Illinois Comp.Stat 5/6-21. The Illinois standard poses several interesting issues. Liability is entirely unrelated to whether the server could appreciate the patron’s intoxication. As we have discussed, a person intoxicated by marijuana may not demonstrate outward manifestations of the same. Therefore, a dram shop could be, without knowing it, serving an intoxicated person. Moreover, the standard requires “causation” of intoxication. This raises an important issue, can an Illinois dram shop defend by arguing it was not the single beer it served the patron which “caused” the intoxication, it was the marijuana use that “caused” the intoxication ?
But then there are the practical implications. Hindsight is always an issue in dram shop cases, but it may be compounded by the introduction of legalized marijuana. Stated differently, are juries more (or less) inclined to find visible intoxication when evidence of marijuana use is a part of the case ?
For example, assume a patron presents himself at a dram shop after consuming edibles. It is generally known that edibles take much longer before peak intoxication is reached than smoking. Thus, one can envision a scenario where at the time of service, a patron may not have visible signs of intoxication, or even be intoxicated. But after service, the pre-service use of marijuana may cause the patron to become intoxicated. Can we effectively defend based on the absence of visible signs of intoxication at the time of service in such cases?
Similarly, as we have already discussed, outward manifestations of THC intoxication are more subtle than alcohol. Accepting that, what happens in the case of a single encounter? For example, a patron that is already intoxicated might be served one drink at the dram shop. Thus, the server, has a single encounter and from that, is expected to detect signs of visible intoxication. How will juries to react to such cases?
The importance of answering these questions is highlighted by the fact that a study by the Center for Disease Control and Prevention found that from 2015-2019, one-third of adults in Colorado who reported binge drinking also reported using marijuana. Crawford, Kacy, Current Marijuana Use and Alcohol Consumption Among Adults Following the Legalization of Non-medical Retail Marijuana Sales – Colorado, 2015-2019, MMWR Morb Mortal Wkly Rep., v. 70 (43), October 29, 2021.
Other Interesting Issues.
The beverage industry, including the alcoholic beverage industry, is now producing THC infused drinks and, in some states, the sale of such drinks are permissible in traditional dram shops. Minnesota adopted a statute that prohibited bar servers from selling a patron alcohol and THC beverages in the same five-hour period. The statute gave rise to several practical challenges for the dram shop. Keeping track of who consumed what beverage over the course of time is practically impossible. Part of the concern was the differing effects of only drinking alcohol, or only using a THC product. Sometimes called cross-fading, the mixing of two intoxicants can intensify the effects of both.
Many places across the U.S. only permit cannabis use in private residences. But there are currently 10 state that have legalized social cannabis consumption at some level, including consumption lounges. Much like the “coffee shops” in Amsterdam, these venues provide a place where cannabis users can consume cannabis, instead of coffee. Strictly speaking, patrons can only buy marijuana at licensed dispensaries. However, in many states, licenses for cannabis lounges are being first offered to dispensary operators, and many of them have opened lounges adjacent to their dispensaries.
Moreover, whereas liability insurance is often available for dispensaries, and liquor liability policies generally protect the dram shop that serves alcohol to a THC-intoxicated person, there is a limited insurance market for cannabis lounges.
There has been much discussion, in light of the aforementioned, and even for traditional dispensaries, regarding the potential for the adoption of Gram Shop Statutes, which would operate similarly to Dram Shop Statutes. But even these are not as simple. For example, what happens when the intoxicated person purchases a THC product while intoxicated and is later in an accident, but the intoxicated person has not consumed the product they purchased from the Gram Shop?
In some states, even the local convenience store is permitted to sell THC-infused soft drinks. Thus, the universe of businesses who may find potential liability for the sale of THC products is extending beyond what we might have first imagined.
Finally, some jurisdictions have passed ordinances that allow a bar or restaurant owner to permit patrons to consume marijuana on their premises. In 2023, the City of Raytown, Missouri amended its municipal code to allow bar and restaurant owners to permit onsite consumption of marijuana, provided a sign is posted at the entrance saying “WARNING! Smoking of Marijuana is not regulated in this establishment.” To date, no establishment has adopted such a policy.