Employers in safety-sensitive industries, such as transportation, have a legitimate and compelling interest in safety. They must protect the public, their employees, and their business from the potentially catastrophic consequences of impaired work performance.
Against that backdrop, many employers assume that random drug and alcohol testing is a lawful and effective safety measure. Recent decisions suggest otherwise.
In McCarthy v. Bison Transport Inc., 2026 ONSC 3729 [Bison], released in June 2026, the Ontario Superior Court of Justice upheld the termination for cause of a long-haul truck driver who failed a random drug test for a second time. The decision appears to confirm that employers in safety-sensitive industries may enforce robust drug and alcohol policies, including random testing programs, provided the policy is reasonable, clearly communicated, and consistently enforced.
Less than a month later however, an Ontario labour arbitrator reached the opposite conclusion in Amalgamated Transit Union, Local 113 v. Toronto Transit Commission, 2026 CanLII 72132 [TTC]. After an extensive review of evidence, the arbitrator found the TTC's random drug and alcohol testing policy unreasonable, ordered the TTC to cease random testing, and reinstated employees who had been terminated following positive test results.
Both cases involved transportation employers with safety-sensitive jobs. Yet the results were entirely different.
The lesson for employers is simple: random drug and alcohol testing remains a legally complex and fact-specific area of workplace law.
Bison Transport: Random Testing Upheld
In Bison, the employee worked as a long-haul truck driver and regularly crossed the Canada-U.S. border as part of his duties. Approximately 22% of his mileage involved travel into the United States.
The employee:
- Tested positive for cannabis in a random drug test in 2014.
- Signed a written warning acknowledging that future violations could result in immediate termination.
- Tested positive again in 2017 and admitted to marijuana use.
- Was terminated for just cause as a result.
The Court upheld the termination for cause. In doing so, it emphasized several factors:
- The employee occupied a safety-sensitive position.
- The drug and alcohol policy was clearly communicated to and understood by the employee.
- The policy was consistently enforced.
- The employee had been specifically warned that a second violation could result in immediate termination.
- The employee did not suffer from a substance-dependency disability that would trigger accommodation obligations under human rights legislation.
Importantly, the Court also recognized the practical realities of cross-border trucking. U.S. regulatory requirements contemplate random testing for certain commercial drivers. A failed random test can prevent a driver from performing cross-border work, affecting the employer's ability to service customers and comply with regulatory obligations. As a result, the risks associated with a positive test extended beyond workplace safety and included legitimate business and operational concerns.
For employers, Bison initially appears to be a reassuring decision. It confirms that properly designed and enforced drug and alcohol policies can support a dismissal for cause in appropriate circumstances.
TTC: Random Testing Struck Down
The TTC decision demonstrates why the analysis is far more complicated than it may appear at first.
In TTC, the Arbitrator concluded that the TTC's random testing policy was an unreasonable exercise of management rights and violated employees' privacy interests.
The arbitrator accepted that public safety is a core interest of the TTC. However, she explained that safety concerns alone do not automatically justify random testing.
The key issue was whether the TTC could demonstrate a general workplace problem involving drugs and alcohol that justified the intrusion of privacy by random testing. After reviewing years of evidence, including accidents, incidents, testing results, and workplace reports, the arbitrator found that it could not. While some employees had attended work under the influence of drugs or alcohol, the evidence reflected only a small number of individuals within a workforce consisting of thousands of employees.
In the arbitrator's words, the evidence represented "a matter of individuals and not a general problem" to justify subjecting approximately 87% of the TTC's workforce to random testing.
The arbitrator also expressed concerns regarding the reliability of oral fluid testing as a measure of actual impairment and fitness for duty. She found that the testing method could identify past use without reliably establishing whether an employee was impaired at work.
Because the policy itself was found to be unreasonable, the arbitrator excluded the testing evidence and ordered the reinstatement of affected employees, while directing the TTC to cease random testing of bargaining-unit employees.
Why Were the Outcomes So Different?
Three distinctions help explain these dramatically different results.
1. Unionized Versus Non-Unionized Workplaces
TTC arose in a unionized workplace and was decided under labour arbitration principles which require a careful balancing of safety interests against employee privacy rights.
Bison, by contrast, was a wrongful dismissal action in a non-unionized workplace. The Court's primary focus was whether the policy was well-known to the employee, consistently enforced, and whether the for-cause termination was proportionate to the breach.
2. Cross-Border Regulatory Requirements
The Bison employee regularly drove into the United States, where random testing is legally required for safety sensitive functions. The Court recognized that a failed test could affect not only safety but also Bison's ability to conduct its business and service customers.
No comparable regulatory context existed in TTC.
3. Evidence of a Workplace Substance-Abuse Problem
The TTC arbitrator placed significant weight on the absence of evidence of a general workplace drug and alcohol problem, as the existence of a safety-sensitive workplace is generally not, on its own, sufficient to justify random testing. Employers must usually show enhanced safety risks, such as a demonstrable workplace substance-abuse problem.
That issue was simply not considered in the Court's analysis in Bison.
Human Rights Risks Remain
One common theme in both cases is the need to consider human rights obligations.
Drug and alcohol dependency may constitute a disability under human rights legislation. Where an employee has an addiction or is perceived to suffer from an addiction, employers may have a duty to accommodate to the point of undue hardship.
Accordingly, a positive drug or alcohol test should never automatically lead to termination without first considering whether accommodation obligations may be engaged.
What Employers Should Take Away
The juxtaposition of Bison and TTC demonstrates that random testing remains far from settled, even within the transportation industry.
An employer may successfully rely on a positive test and a carefully drafted policy to support a termination for cause, as occurred in Bison. Yet another employer in the same industry may find its entire testing program declared unlawful, as occurred in TTC.
Employers must also appreciate the significant consequences if they get it wrong. An invalid policy can lead to wrongful dismissal claims, labour grievances, human rights complaints, reinstatement orders, and substantial back-pay liability.
At the same time, employers cannot ignore the risks of workplace impairment. A serious accident involving drugs or alcohol may expose the organization to significant safety, regulatory, reputational, and economic consequences.
Employers in safety-sensitive industries must balance this dangerous dilemma.
Final Thoughts
The real lesson from Bison and TTC is not that random testing is permissible or impermissible. The answer depends heavily on the particular workplace, the governing legal framework, the available evidence, the testing technology being used, and the employer's human rights obligations.
Employers should carefully assess their circumstances and obtain expert legal advice before implementing, revising, or enforcing a random drug and alcohol testing policy.
In short, random testing is much trickier than it looks. A policy that works for one employer may expose another to significant legal liability. The most effective approach is a carefully tailored policy that maximizes safety while minimizing exposure to wrongful dismissal, privacy, labour relations, and human rights claims.
If you require assistance with the above, please do not hesitate to contact us for expert legal advice and representation.
