IOL: Positive dagga test doesn't prove impairment at work

The core finding: a test shows exposure, not impairment

A positive urine screen for cannabis detects metabolites from past use, and it cannot show whether a worker was impaired during a shift.

That's the thrust of the IOL piece, published November 12, 2026 and surfaced through a Google News query on cannabis policy. Toxicologists have long known the distinction. Disciplinary hearings still routinely ignore it.

The gap is structural. THC is fat-soluble, and its metabolites linger for days in occasional users and for weeks in heavy ones, so a person can test positive long after any psychoactive effect has ended. Alcohol is the opposite case, because blood and breath levels track impairment closely enough that employers and courts accept them as proxies.

One caveat on sourcing. The signal available to CannIntel is the headline and publication details of the IOL article, not its full text. This report therefore doesn't attribute specific quotes, named experts or case figures to IOL, because none were supplied. What follows is the legal and operational context the headline sits inside.

The legal backdrop: private use is legal, workplace rules are not settled

South Africa decriminalized private adult cannabis use through the Constitutional Court's 2018 Prince judgment, but that ruling did not rewrite workplace law.

In Minister of Justice and Constitutional Development v Prince (2018), the Constitutional Court held that criminalizing the use, possession and cultivation of cannabis by an adult in private for personal consumption was unconstitutional. The ruling protected the private sphere. It said nothing that obliges an employer to tolerate use or to ignore a positive result.

Parliament later moved to codify the position through the Cannabis for Private Purposes Act, signed into law in 2024. The statute reinforces private adult use and cultivation. Employment obligations still flow primarily from the Labour Relations Act, the Occupational Health and Safety Act and an employer's own disciplinary code.

So the legal picture is split. Off-duty private use is lawful. Being impaired at work, particularly in safety-sensitive roles, remains a legitimate ground for discipline. Everything turns on proving which of those two situations a positive test reflects.

Why the testing gap creates exposure for employers

An employer that dismisses on a bare positive result, with no evidence of impairment, risks losing at the CCMA on procedural and substantive fairness grounds.

The practical problem is evidentiary. Under South African labour law, a dismissal must be both substantively and procedurally fair, and if the stated offense is working while impaired, the employer carries the burden of showing impairment, not mere prior consumption. A metabolite reading, standing alone, struggles to carry that weight when the worker can credibly say the use occurred off-site and off-shift.

That creates a real fault line between two kinds of policy:

• Policies that prohibit being under the influence at work, which require evidence of impairment.
• Zero-tolerance policies that treat any positive result as a violation, which are easier to administer but harder to defend after Prince.

Safety-sensitive sectors get more latitude. Mining, transport, heavy machinery and construction can credibly argue that a lower threshold is justified by risk. Even there, a well-drafted policy ties discipline to impairment indicators or to fitness-for-duty assessments, not to a lab number alone.

The IOL framing, that a positive test does not equal being stoned at work, is a blunt version of an argument employment lawyers have been making since 2018. It's hard to dispute on the science.

What a defensible workplace policy looks like now

The strongest policies pair testing with documented observation of impairment and clear, pre-announced rules for safety-critical roles.

Employers reviewing their codes have a short checklist of fixes. Define the prohibited conduct as impairment or being under the influence, not as the presence of a substance. Identify which roles are safety-sensitive and say why. Require a trained supervisor to record observable signs of impairment before a test is ordered where practicable. Treat a positive screen as the start of an inquiry. It isn't the end of one.

Medical cannabis adds another wrinkle. A worker with a valid prescription or authorization who tests positive raises accommodation questions under anti-discrimination principles, and a blanket-ban response is more exposed to challenge. Expect these disputes to grow as lawful private and medical use normalizes.

For the testing industry, the pressure points toward impairment-focused methods and away from legacy urine panels. Oral-fluid tests detect more recent use and are gaining ground, though they still don't measure impairment directly. No validated roadside or workplace test yet does for cannabis what a breathalyzer does for alcohol.

Workers and unions have a clear lever. Any dismissal resting only on a positive result invites a challenge, and the science favors the employee on that narrow point.

For broader context on how this issue is developing, see the CannIntel topic hub on South Africa cannabis workplace drug testing.

The next signal to watch is whether the CCMA or the Labour Court issues a ruling squarely testing a dismissal built on a bare positive result in the post-Prince, post-2024 Act environment. Such a decision would turn a widely shared scientific point into binding employment precedent, and employers would rewrite their codes within weeks.