On December 18, 2025, President Trump signed an executive order directing the Department of Justice and the Drug Enforcement Administration to expedite rulemaking to reclassify marijuana from Schedule I to Schedule III of the Controlled Substances Act.
The following morning, HR professionals across the country started fielding the same question from employees: does this mean the drug testing policy has changed?
On December 19, 2025, the Department of Transportation published a compliance notice clarifying that until rescheduling is formally completed through the rulemaking process, DOT drug testing requirements remain unchanged. Safety-sensitive transportation employees will continue to be tested for marijuana.
The policy had not changed. The confusion, however, was entirely predictable — and it reveals something important about where workplace cannabis policy now sits. Employers are managing a workforce that is actively tracking federal and state cannabis developments and drawing their own conclusions about what that means for workplace rules. The gap between what employees believe the policy is and what the policy actually requires has never been wider. And in several major employment markets, the legal requirements around cannabis testing have genuinely changed in ways most employers have not fully absorbed.
What Rescheduling Means — and Doesn’t Mean
The distinction between Schedule I and Schedule III matters for several reasons, but for most employers, it does not change the immediate operational picture.
Schedule I drugs are defined as having no currently accepted medical use and a high potential for abuse. Schedule III drugs are recognised as having an accepted medical use, with moderate to low potential for physical and psychological dependence. Reclassifying cannabis to Schedule III would not make it federally legal. Possession and use would still be regulated under federal law. But the penalties for certain offences would be reduced, and — importantly — individuals might gain access to ADA reasonable accommodation claims for medical cannabis use that have historically been unavailable because Schedule I substances are not recognised as having medical legitimacy.
If rescheduling is completed, employees who have a disability may be able to obtain a reasonable accommodation under the Americans with Disabilities Act for consuming cannabis, because Schedule III drugs are recognised as having a currently accepted medical use. This would be a significant development — medical cannabis users who have had limited success asserting ADA accommodation claims at the federal level may find that avenue opening.
The rulemaking process is not instantaneous. Federal agency rulemaking typically takes months. Until the process concludes, the current Schedule I designation remains in effect and employers are not required to change anything. But HR and legal teams who are not tracking this actively risk being caught flat-footed when the formal reclassification occurs.
The practical guidance for now: maintain existing drug testing policies for cannabis. Communicate clearly to your workforce that the executive order does not change workplace rules. And begin building the policy review into your 2026 compliance calendar so you are not responding reactively when reclassification is finalised.
The State Law Patchwork: Where Cannabis Testing Is Now Legally Restricted
While federal rescheduling remains in progress, state law has moved considerably further and faster. The compliance landscape for cannabis employment testing is now a genuine multi-jurisdictional patchwork, and for employers operating across multiple states, applying a single uniform testing policy has become legally risky.
Forty states and Washington DC have legalised marijuana for medical purposes. Twenty-four states and Washington DC allow marijuana for recreational purposes. Many states prohibit employers from disciplining or firing an employee for off-duty, legal medical and/or recreational marijuana use.
The specific employment protections vary significantly. At one end of the spectrum are states like Colorado and Texas, which have legalised cannabis in some form but provide relatively limited employment protections — employers in those states retain broad discretion to maintain drug-free workplace policies and act on positive cannabis tests. At the other end are states with specific anti-discrimination protections tied to cannabis use.
At least nine recreational cannabis states — California, Connecticut, Nevada, New Jersey, New York, Minnesota, Montana, Rhode Island, and Washington — have some form of employment protections for recreational cannabis use. The specific protections differ, but in each of these states, adverse action based solely on off-duty cannabis use — or on a drug test that detects past use — creates legal exposure.
California has gone furthest. California AB 2188, effective January 2024, amended the Fair Employment and Housing Act to make it unlawful for an employer to discriminate against an applicant or employee for the use of marijuana off the job and away from the workplace, or for an employer-required drug screening test that finds the person to have non-psychoactive cannabis metabolites in their system. The California law reflects a scientific reality that is driving similar legislation elsewhere: the standard urine test detects metabolites that remain in the body for up to four weeks after last use, bearing no relationship to whether an employee was impaired at the time of testing or on the day they work.
New York’s Marijuana Regulation and Taxation Act prohibits most employers from testing candidates for cannabis pre-employment at all, with exceptions for safety-sensitive roles, federal contractors, and certain other categories. Minnesota’s law prevents employers from withdrawing a job offer based solely on an initial positive cannabis screening. New Jersey’s regulations restrict adverse action based on positive cannabis tests without additional evidence of impairment.
For an employer operating across California, New York, New Jersey, and Minnesota — covering a significant portion of the US workforce in major employment centres — the pre-employment cannabis testing programme that was standard practice five years ago now creates liability in every one of those markets.
The Fundamental Problem: Drug Tests Don’t Measure Impairment
The deepest challenge in workplace cannabis policy is one that legislation is beginning to reflect but that most employers have not yet fully internalised: no currently available, commercially deployed drug test can reliably determine whether someone is currently impaired by cannabis.
The standard urine test detects THC metabolites. Those metabolites are the by-products of THC processing in the body and can persist for days or weeks depending on frequency of use, body composition, and other factors. Their presence confirms past cannabis use. They do not indicate that the person is currently impaired — or that they were impaired at any particular time. Drug tests can detect THC in urine for up to four weeks after use, meaning occasional cannabis users are much more likely to test positive than people who use alcohol or other substances that leave the system faster.
This creates the policy paradox that California’s AB 2188 addresses directly: a test that measures past use rather than current impairment is a poor tool for making employment decisions in a safety context, and it disproportionately affects people who use cannabis legally and responsibly — including medical cannabis users treating chronic conditions — rather than catching the employees who are actually impaired at work.
The alternative being developed is oral fluid testing. Oral fluid tests offer a shorter detection window, typically detecting usage over the past 24 to 48 hours, and may better indicate recent use. THC is often detectable in blood tests until about a week after last use. Saliva testing aligns better with the question employers actually care about in a safety context: has this person used cannabis recently enough that impairment is plausible right now?
As of January 6, 2026, the Federal Register confirmed there are still no HHS-certified labs for oral fluid DOT testing. For DOT-regulated employers, oral fluid testing remains unavailable until certification is issued. For private employers, oral fluid testing is available and increasingly represents the appropriate methodology in cannabis-legal states where metabolite-based adverse action is restricted.
The transition to impairment-focused testing rather than metabolite detection is where workplace cannabis policy is heading. Employers who understand this now are better positioned to build a programme that is both legally defensible and genuinely effective at its stated purpose.
Safety-Sensitive Roles: Where the Rules Are Different
The legal protections for cannabis use described above contain a consistent exception: safety-sensitive roles. Almost universally, state cannabis employment protection laws preserve the employer’s right to maintain drug-free policies for positions where impairment creates significant risk to the worker, co-workers, or the public.
Defining safety-sensitive is not always straightforward — and getting it wrong in either direction creates exposure. Over-designating roles as safety-sensitive to justify expansive testing may not withstand legal scrutiny in jurisdictions with strict cannabis employment protection laws. Under-designating them and applying relaxed testing standards to roles where impairment genuinely matters creates negligent hiring and retention exposure.
DOT-regulated employers have the clearest framework: the DOT defines the safety-sensitive positions subject to mandatory drug testing in each regulated industry, and cannabis testing for those positions remains fully in effect regardless of state legalisation. DOT safety-sensitive rules and prohibitions remain unchanged today. Keep current policies, testing panels, enforcement procedures, and consequences in place.
For non-DOT employers, the safety-sensitive designation requires a genuine analysis of role-specific risk. A forklift operator in a distribution centre has a different risk profile from a remote software developer. A home health aide working with vulnerable adults has a different profile from a back-office accounting role. The framework for this analysis should be documented, applied consistently, and reviewed when roles change.
Importantly, safety-sensitive designation does not simply mean “we can test more” — it means the testing programme for those roles can be designed around genuine safety objectives rather than the employment protection constraints that apply to other positions. That distinction, clearly documented and consistently applied, is the foundation of a defensible safety-sensitive testing programme.
Building the Policy for 2026: The Practical Framework
Given the complexity of the current landscape — federal rescheduling in progress, state-by-state protections, fentanyl testing gaps, methodology shifts — what does a drug testing policy appropriate for 2026 actually look like?
Audit your current panel against actual risk. The standard five-panel test was designed for a different drug landscape. If your workforce operates in industries where fentanyl is a genuine presence in the broader population, a panel that cannot detect it is not fit for purpose. Add fentanyl and norfentanyl to the panel for safety-sensitive roles. The HHS rule applying to federal employees sets the standard; the risk data that drove that rule applies to all employers.
Map your state-by-state obligations. For every state where you hire, establish which specific cannabis employment protections apply. The answer varies dramatically. An employer with facilities in California, New York, and Texas is operating under three significantly different legal frameworks for cannabis testing. A single national policy is likely non-compliant in at least one of those markets. The map needs to be current — this area is moving quickly, with additional states likely to strengthen protections in 2026.
Separate pre-employment from ongoing testing design. The Quest data makes clear that pre-employment testing alone is insufficient for safety-sensitive roles. Random testing, post-accident testing, and reasonable-suspicion protocols are not optional add-ons to a serious safety programme — they are the mechanism by which ongoing use is detected. Each has its own legal requirements that vary by state, including states that restrict random testing to safety-sensitive roles or require specific triggering conditions for post-accident testing.
Align methodology with purpose. In cannabis-legal states with metabolite testing restrictions, transition to oral fluid testing for pre-employment cannabis screening where operationally feasible. In states without those restrictions, urine metabolite testing remains permissible but should be applied consistently and with full awareness of what it does and doesn’t indicate.
Communicate the policy clearly and specifically. In the wake of the December 2025 rescheduling order, employees are actively forming views about what workplace drug policy requires. An organisation that has not proactively communicated that its policy remains unchanged — and why — is allowing that confusion to continue. The real risk right now is inconsistent messaging to your workforce. Employees are hearing news about cannabis policy changes and making assumptions about what that means for your workplace rules. The communication investment is small relative to the cost of an employee making a workplace decision based on an incorrect assumption about what the policy permits.
Build a review cadence into the programme. Drug testing policy is no longer a set-and-forget exercise. The regulatory environment in 2026 — DOT fentanyl rules, state cannabis protections, oral fluid testing certification, cannabis rescheduling — is producing significant changes on a rolling basis. A compliance review at least annually, and ideally semi-annually in the current environment, is the appropriate maintenance posture.
The Bigger Picture: Testing as One Tool, Not the Whole Answer
There is a version of this conversation that frames drug testing as the complete solution to workplace substance use risk. It isn’t, and treating it as such creates both false confidence and policy problems.
A pre-employment screen that passes an employee who is not using substances at that specific moment tells you something limited about that person’s ongoing behaviour. A random testing programme catches post-hire use in the tested population but cannot make the entire workforce visible. Reasonable suspicion testing relies on supervisors who are trained and willing to raise concerns — and supervisors who are close enough to their teams to notice the signs.
The comprehensive approach to workplace substance use risk combines testing with supervisory training on recognising impairment, an Employee Assistance Programme that creates a pathway to help without requiring employees to self-incriminate, a culture where safety concerns can be raised without retribution, and clear, consistently applied consequences for impairment on the job regardless of the specific substance.
The data from Quest’s 2025 index — and the legislative response to it — signals that the era of minimal, compliance-check drug testing is ending for safety-sensitive employers. The employers who replace it with something genuinely sophisticated and genuinely targeted will be better positioned legally, and their workers will be safer.
AMS Inform provides background verification and workforce screening services across 160+ countries. For organisations reviewing their workforce screening and drug testing compliance frameworks, speak to our team at AMSinform.com.
