Pre-Access Testing Rejected in Ontario

Drug and alcohol testing of employees in the construction industry continues to be an area of concern and constantly developing law for employers. The issue of random testing was recently dealt with by the Supreme Court of Canada in the Irving Pulp and Paper Ltd. decision of June, 2013, wherein the Court held that random alcohol testing was impermissible without extraordinary circumstances showing a pervasive problem in the workplace.

This month, an arbitration award out of Ontario (Re Mechanical Contractors Association Sarnia et. al.) held that pre-access testing was an unreasonable exercise of management rights, and that it was in violation of the applicable human rights legislation.

The general contractor had insisted that each of its contractors have a drug and alcohol policy which met or exceeded the standards of its own policy or that of the 2005 Canadian Model for Providing a Safe Workplace. The contractors subsequently imposed pre-access testing (which included breath, saliva, oral fluid and urine testing) in accordance with this arrangement.

In coming to the conclusion that the testing was unreasonable, the arbitrator relied heavily on the Irving Pulp and Paper Ltd. ruling that there was an obligation upon employers to justify the invasion upon employee privacy rights occasioned by drug and alcohol testing. As with random testing (which was dealt with in that SCC decision), the arbitrator held that it was insufficient for an employer to cite general concerns about safety as a justification for the imposition of pre-access drug testing, but that an employer must show evidence of an actual, existing substance abuse problem at the specific worksite(s) “which creates a real potential for significant negative workplace health and safety events” in order to satisfy the pre-condition for drug testing without cause. That was determined to be a precondition before the arbitrator would consider whether the policy itself is reasonably balanced against privacy interests.

In support of its argument that there were pre-existing substance abuse problems, the employer relied on evidence of general statistics surrounding drug and alcohol use in Southwestern Ontario, statistics suggesting reduced positive pre-access tests at that location, and statistics which suggested reductions in positive post-incident tests for other employers who had implemented pre-access testing.

The arbitrator concluded that this form of evidence which he labelled “broad-based statistical inferential reasoning” was not sufficient to prove an existing problem on a balance of probabilities. In particular, the arbitrator rejected the evidence given by a number of experts called by the employer since their conclusions were based upon broad statistical averages, even though a number of them were qualified to opine on statistics.

The award also held that the pre-access testing was a violation of human rights legislation in that pre-access testing does not demonstrate impairment at work (since the testing occurs prior to access) and there was no evidence that a positive test would be an accurate predictor of future use. The discrimination occurred here because the result of the positive test was a ban from Suncor worksites and an effective termination of employment. Moreover, the policies failed to include an individual consideration or accommodation of employees.

If this award is followed, it means that an employer must show a pre-existing drug and/or alcohol problem at its worksite before it can implement either random testing, or pre-access testing. Moreover, it appears that an employer will not be able to rely on general statistical averages regarding substance-use in the region, or on effectiveness of similar programs for other employers to show a problem. At a minimum, an employer implementing drug testing will need evidence that those statistical averages relate to the worksite in question; and more likely, will require actual evidence from its own specific workplace before it can justify not-for-cause testing. Once it clears this hurdle, employers will still be faced with showing that the testing is reasonably balanced against privacy rights by showing that the policy is effective and impairs employee rights to the minimal extent necessary.

This award is further evidence of the contrasting attitudes of Eastern and Western Canadian arbitrators. Western Canadian arbitrators have valued safety over privacy. However, in Western Canada there is also more supporting data, particularly in the construction industry, which is important, if not vital, in justifying testing. The requirement to provide evidence of an existing substance abuse problem limits an employer’s ability to take proactive steps to protect the health and safety of workers, the public and property and moves employers into a reactive position wherein it must wait for problems, and potentially injuries/fatalities, before it can take steps to limit those risks. Which should be more important? Safety or privacy?