New debate over drug, alcohol testing

Supreme Court of Canada determined employers must have proof of problem, but new case asks what qualifies as a problem?

By Jeffrey R. Smith

Last year, the Supreme Court of Canada weighed in on the merits of random drug and alcohol testing for employees in its decision in Communications, Energy and Paperworkers Union of Canada, Local 30 v. Irving Pulp & Paper Ltd.

At the time, I discussed how the debate over such testing had gone on for many years, with various courts and arbitrators going back and forth over whether the advantages of such testing outweighed the negative effects on the employees being tested. The top court ruled random testing using methods such as urinalysis and oral swabs infringed on employee privacy and dignity, and to justify it an employer must not prove just that it had a dangerous workplace with safety sensitive positions — as Irving Pulp and Paper argued — but also that there was a serious problem with drug or alcohol use by employees. It was thought this decision would perhaps end the ongoing debate, but maybe it just changed it a little bit.

An Alberta arbitration board recently reached a decision on a grievance over a drug and alcohol testing policy by Suncor Energy for employees in its oil sands operation. The policy instituted random testing of employees at its operations near Fort McMurray. The testing involved mostly breathalyzers and urinalysis. The union said the testing violated employee privacy and caused unnecessary stress for those being tested. Suncor argued it had proof that use of drugs and alcohol by its workers was a problem that needed to be addressed, and its oil sand operation was an extremely dangerous environment in which any instances of impaired employees was unacceptable.

Upon examination, the arbitration board found Suncor had little concrete evidence to support its claims of a problem. The company had reports of an increase in drug and alcohol-related accidents and arrests in the area, but the board noted with the increase in employees over time, the incidents were actually decreasing and the percentage of employees involved was very small. In addition, Suncor’s figures didn’t distinguish between its unionized employees — which only made up about one-third of the workers — and employees of contractors. In fact, many of the incidents to which Suncor referred happened in work camps, which were primarily occupied by contractor workers. With it likely a significant portion of the drug and alcohol problem Suncor presented was from non-employees, it would be unfair to paint all its employees with the same brush and force them to submit to the unpleasant experience of random testing, said the board.

All through its decision, the arbitration board kept the Supreme Court’s decision in mind, so it seems the standard for allowing drug and alcohol testing has been set — the employer must be able to prove a significant, ongoing problem with alcohol and drugs that threatens the safety of employees and/or the public. But as the Alberta case above shows, there may be a new debate beginning over what will be considered proof of a problem. If a workplace is dangerous and employees are in safety-sensitive positions, should there be no tolerance for any incidents, as Suncor argued?

If not, what should be considered the level before random testing is allowed?

Latest stories