Pre-employment drug and alcohol testing

The law of drug and alcohol testing in Canada is in a state of evolution. While the Supreme Court of Canada’s decision in Communications, Energy and Paperworkers Union of Canada, Local 30 v Irving Pulp & Paper Ltd., 2013 SCC 34, provided important guidance on the strict standard that employers must meet in order to subject employees to random testing, it raised many questions regarding how those principles would be applied to other forms of testing.

In the two years since the decision was released, the trend in the case law suggests that the Court’s analysis in Irving Pulp & Paper is not limited to random testing. For example, in Re Mechanical Contractors Association Sarnia v UA Local 663, 2014 ONSC 6909, the Ontario Superior Court of Justice upheld an arbitration award which applied Irving Pulp & Paper to conclude that the employer did not have the right to require pre-employment testing.

In Sarnia, the arbitrator found that it was a bona fide occupational requirement that employees not be under the influence of drugs or alcohol at work. However, he disagreed with the employer’s position that pre-employment testing was reasonably necessary to ensure that employees would not be impaired at work. Importantly, he emphasized that a positive test before hiring is not a valid predictor that the candidate will be impaired at work in the future.

The arbitrator extended the principles from Irving Pulp & Paper to pre-employment testing and concluded that an employer does not have the right to impose pre-employment testing unless there is evidence of an actual substance abuse problem at the specific worksite where the person is hired which creates a real risk of significant negative workplace health and safety events. The arbitrator rejected the employer’s attempt to establish a safety risk by relying on statistical evidence regarding drug and alcohol use in the region, and regarding a correlation between pre-employment testing and a reduction in positive post-incident tests. The arbitrator held that this “broad-based statistical inferential reasoning” was not sufficient to prove an existing problem at the employer’s worksite which would justify pre-employment testing.

If this case is followed, it will mean that an employer must meet the high threshold of proving an actual significant drug or alcohol problem at its particular worksite before it can implement pre-employment testing. Even if an employer can meet this strict requirement, it will likely also still need to show that the testing is reasonably balanced against the candidate’s right to privacy. This may be difficult given the arbitrator’s comments in this case that a positive test is not a valid predictor of future impairment at work.