chevron-down Created with Sketch Beta.
April 01, 2017 Features

State Medical Marijuana Legalization and OSHA Anti-Retaliation Rules: Post-Accident Drug Testing Considerations for Employers

By Peter J. Gillespie

In recent years, workplace drug tests showing positive test results have increased steadily to a 10-year high.1 Employers concerned by these findings face an uphill battle given the recent loosening of restrictions on marijuana use at the state level. Moreover, despite the fact that marijuana remains illegal under federal law, the Occupational Safety and Health Administration (OSHA) has somewhat protected drug users from work-related discipline stemming from marijuana use. Specifically, based on findings from the National Highway Traffic Safety Administration (NHTSA),2 OSHA recently contended that mandatory drug testing of employees following a reported injury is a form of retaliation that violates section 11(c) of the Occupational Safety and Health Act of 1970, 29 U.S.C. section 660(c).3 Moreover, OSHA has further advised employers that the agency does not view current drug testing methods as an effective way of determining whether a drug-related impairment caused or contributed to a workplace accident: “For substances other than alcohol, currently available tests are generally unable to establish a relationship between impairment and drug use. Employers should be aware that post-incident drug testing will not necessarily indicate whether drug use played a direct role in the incident.”4

The conflicts raised by state legalization laws and OSHA’s stance on drug testing present a conundrum to employers over when and how to conduct drug testing, especially post-accident testing.

Are Employers Legally Required to Permit Employee State-Legal Medicinal Marijuana Use?

Before launching a deep dive into OSHA’s position on drug testing, it is important to consider whether state or federal laws prevent employers from prohibiting employee marijuana use. In particular, if an employee uses marijuana for medical reasons, the employee may be considered disabled for purposes of state and federal laws. Also, under some state laws, the employee may be protected from discipline based on the lawful use of marijuana outside of the workplace.

From a practical perspective, there are sound reasons why employers should be permitted to exclude medical marijuana users from the workplace. Marijuana intoxication can cause distorted perception, loss of coordination, and problems with memory, learning, and problem-solving. An employee who may use marijuana for medical reasons may be less safe than his or her coworkers. By law, employers are required to provide a workplace free from recognized hazards.5

To date, courts have found that disability laws and state laws permitting the use of medical marijuana do not override employer policies prohibiting the use of illegal drugs. For example, in Coats v. Dish Network, LLC, the Colorado Supreme Court held that because marijuana use remains illegal under federal law, an employee could not assert that his use of medicinal marijuana constituted protected, lawful activity under state law.6 Similarly, federal courts have held that the Americans with Disabilities Act does not protect marijuana users from being disciplined because marijuana remains illegal at the federal level.7 As state laws proliferate, a future challenge to an employer policy may come out the other way.

Why Does OSHA Distinguish Between Post-Injury Alcohol Testing and Testing for Other Drugs?

Because, as noted above, OSHA has rejected the legitimacy of drug testing (but not alcohol testing) in response to a workplace injury, employers should understand the differences between alcohol testing and marijuana testing. The detection window for alcohol largely depends on how high a user’s blood alcohol concentration (BAC) reached. A moderately intoxicated individual (i.e., 0.08 percent BAC) could have detectable levels of alcohol for about eight hours. In addition, alcohol levels in the blood and the brain are relatively consistent at any given point in time. As a result, alcohol impairment levels are fairly closely tied to blood alcohol levels and create consistent symptoms among users.

By contrast, marijuana smokers experience a high that lasts approximately two hours. Most behavioral and physiological effects end within three to five hours, but users may suffer residual effects for up to 24 hours, such as difficulty with complex, divided attention tasks.8 As tetrahydrocannabinol (THC) metabolizes, inactive carboxy-THC (THC-COOH) appears in the user’s urine. However, THC-COOH may remain detectable in a user’s urine for days. For example, in the case of a heavy user of marijuana, urine tests for THC-COOH can show positive results for weeks or months. In other words, a urine test mainly flags the use of marijuana but not current impairment, because the user may no longer be impaired when THC-COOH is detected.9

Oral fluid (saliva) is a newer method of testing that is beginning to be used in place of urinalysis.10 This method has the shortest detection window and, unlike a urinalysis test, detects the presence of active THC. For these reasons, saliva testing may eventually provide a way of establishing injury-related marijuana impairment that will be accepted by OSHA. Nevertheless, questions currently remain about this testing method. For the time being, the safest course to avoid getting into OSHA’s crosshairs may be to limit injury-related testing to alcohol only.

Why Doesn’t OSHA Believe a Positive Marijuana Test Correlates with Current Impairment?

Alcohol testing generally produces results that are readily understood and generally accepted indicators of an individual’s impairment level. By contrast, it is less clear whether the presence of THC or THC-COOH at a particular concentration is indicative of a current impairment. In particular, the NHTSA currently advises on testing for marijuana:

It is inadvisable to try and predict effects based on blood THC concentrations alone, and currently impossible to predict specific effects based on THC-COOH concentrations. It is possible for a person to be affected by marijuana use with concentrations of THC in their blood below the limit of detection of the method.11

In other words, the NHTSA has not identified a THC test result that can be relied on to establish impairment.12 Because many states have used a THC blood level of 5 ng/ml to define impairment under DUI laws, some experts have recommended that employers adopt this level as well.13 However, others have described the 5 ng/ml THC blood concentration level as a purely arbitrary figure.14 Indeed, in 2015, the NHTSA published a report finding no correlation between marijuana impairment and driving accidents based on a study of more than 9,000 drivers conducted over a 20-month period.15

Based on these findings, employers should recognize that there currently is no generally accepted “magic number” test result for THC that employers can rely on to establish that an employee was impaired at the time of a workplace injury. For these reasons, the better course for employers that intend to prohibit marijuana use is to maintain a blanket prohibition on illegal drug use, rather than attempt to tailor a drug testing policy around establishing that the employee was impaired at the time of an injury or an accident.

What to Do Now?

Employers that maintain drug testing policies should assess these policies in light of recent changes to state marijuana laws and OSHA’s new regulations. Employers can maintain policies that prohibit working while impaired from marijuana use. Because there is still significant uncertainty as to whether employers can effectively test for impairment, the safer position remains simply to prohibit the use of marijuana as an illegal substance, unless such a policy violates applicable state or local laws. Additionally, unless employers have other lawful reasons for relying on mandatory injury-related testing for drugs,16 applicable policies should be tailored not to require mandatory drug tests triggered by reported injuries. Because there have been significant changes in the law in this area, employers should keep abreast of developments on these issues to reduce the risk of inadvertent noncompliance with applicable requirements. ◆


1. Press Release, Quest Diagnostics, Drug Positivity in U.S. Workforce Rises to Nearly Highest Level in a Decade, Quest Diagnostics Analysis Finds (Sept. 15, 2016), /employer_solutions/dti-2016/Quest-Diagnostics-Drug-Testing-Index-press-release-9-15-16/Quest%20Diagnostics%20Drug%20Testing%20Index%20press%20releas%2009-16-2016.pdf.

2. Fiona J. Couper & Barry K. Logan, Nat’l Highway Traffic Safety Admin., U.S. Dep’t of Transp., DOT HS 809 725, Drugs and Human Performance Fact Sheets (2014), available at; see also Richard P. Compton & Amy Berning, Nat’l Highway Traffic Safety Admin., U.S. Dep’t of Transp., DOT HS 812 117, Drug and Alcohol Crash Risk 4 (2015), available at (“Drug tests do not necessarily indicate current impairment.”).

3. Improve Tracking of Workplace Injuries and Illnesses, 81 Fed. Reg. 29,624, 29,627 (May 12, 2016).

4. Injury Tracking and Use of Disciplinary, Incentive or Drug Testing Programs, OSHA, (follow “Drug Testing Programs” hyperlink) (last visited Apr. 19, 2017).

5. 29 U.S.C. § 654(a).

6. 350 P.3d 849, 850–53 (Colo. 2015); see also Garcia v. Tractor Supply Co., 154 F. Supp. 3d 1225, 1230 (D.N.M. 2016) (recognizing that affirmatively requiring an employer to accommodate marijuana use under state law would compel the employer “to permit the very conduct the [Controlled Substances Act] proscribes”).

7. See, e.g., James v. City of Costa Mesa, 700 F.3d 394, 397 (9th Cir. 2012).

8. Strains of marijuana with higher potency may have greater and longer-lasting impairment effects. Robert S. Goldsmith et al., Medical Marijuana in the Workplace, 57 J. Occupational & Envtl. Med. 518, 522 (2015). Also, the effects of marijuana consumed as an edible are felt at a slower rate than when inhaled. Id.

9. Couper & Logan, supra note 2, at 9 (“Detection of total THC metabolites in urine, primarily THC-COOH-glucuronide, only indicates prior THC exposure. Detection time is well past the window of intoxication and impairment.”).

10. Olaf H. Drummer, Drug Testing in Oral Fluid, 27 Clinical Biochemist Revs. 147, 150 (2006).

11. Couper & Logan, supra note 2, at 9.

12. Amy Berning & Dereece D. Smither, Nat’l Highway Traffic Safety Admin., U.S. Dep’t of Transp., DOT HS 812 072, Understanding the Limitations of Drug Test Information, Reporting, and Testing Practices in Fatal Crashes 1 (2014), available at (“Every State has enacted a law defining drivers who are at or above .08 grams per deciliter BAC as ‘legally impaired,’ but there are no similar, commonly accepted impairment levels for other drugs.”).

13. Jennan A. Phillips et al., Marijuana in the Workplace: Guidance for Occupational Health Professionals and Employers, 57 J. Occupational & Envtl. Med. 459, 462 (2015).

14. Robert L. DuPont, Inst. for Behavior & Health, Inc., Workplace Drug Testing in the Era of Legal Marijuana (2015), available at; see also Goldsmith et al., supra note 8, at 523 (“[I]t is difficult, if not impossible, to establish a relationship between a person’s THC blood or plasma concentration and performance impairing effects.”).

15. Compton & Berning, supra note 2.

16. For example, OSHA has acknowledged that injury-related drug testing required by Department of Transportation regulations, workers’ compensation drug-free workplace policies, or other similar state or federal legal requirements are not prohibited by OSHA’s anti-retaliation rules. Injury Tracking and Use of Disciplinary, Incentive or Drug Testing Programs, supra note 4.

Peter J. Gillespie

Peter J. Gillespie ([email protected]) is of counsel at Laner Muchin, Ltd. in Chicago, where he represents and counsels management on a wide array of employment law issues, including workplace safety and health, wage and hour laws, covenants not to compete, discrimination and harassment, wrongful discharge, whistleblower claims, class actions, hiring, discipline, promotion and dismissal decision making, workplace privacy, and statutory compliance. He is the Chair of SciTech’s Membership and Diversity Committee.