Suncor’s drug and alcohol policy given another chance by appeal court

Not all evidence considered in initial decision

The random drug and alcohol testing at Suncor has been given a second chance by an Alberta court after an arbitration board previously struck it down.

Suncor has oilsands extraction operations in two locations north of Fort McMurray, Alta. As of July 2013, there were more than 3,300 unionized workers, almost 3,000 non-represented workers and up to 3,400 contractor employees.

Much of the activity in Suncor’s oilsands operations involved heavy equipment and dangerous activities, so safety was a concern. The company had several safety measures in place to ensure awareness and preparedness for safety hazards, such as employee and supervisor training, an employee assistance program, a treatment program for employees with addictions, drug and alcohol detection measures, and a rapid site access program for contractors’ employees.

Suncor introduced pre-employment drug and alcohol testing for all new employees in 1999 to help ensure safety and productivity wouldn’t be hindered by intoxicated workers. 

Over the next few years, the company introduced further measures to guard against workplace impairment, including alcohol and drug testing of employees via urinalysis following an incident at the workplace or where there were reasonable grounds to suspect impairment, along with those completing rehabilitation and returning to work after drug- or alcohol-related issues.

In 2012, Suncor told the union it was introducing random drug and alcohol testing for employees working in safety-sensitive positions, which described about 82 per cent of unionized employees. 

The new policy also gave supervisors discretion to allow employees to return to work if they tested positive with a 0.02 to 0.039 per cent blood alcohol concentration.

The union grieved the new policy, and an arbitration board pointed out that although Suncor claimed the danger inherent with the slightest inattention or mistake at its oilsands operations, its policy was not zero tolerance and supervisors had leeway to make a judgment call on whether employees were fit to work. However, “in a random testing regime, a positive test will be the only result on which significant discipline — or even dismissal — may rest,” said the board.

The board looked to the Supreme Court of Canada’s determination in Communications, Energy and Paperworkers Union of Canada, Local 30 v. Irving Pulp & Paper Ltd. that said random testing is only warranted if the employer establishes there is a “general problem” with drug or alcohol use in the workplace. Suncor presented reports showing issues with positive tests, evidence of drug and alcohol use by workers and the three deaths onsite, and the company also argued any instances were unacceptable in such a safety-sensitive environment.

However, the board found the evidence showed the policy and testing regime already in place — post-incident and return-to-work after rehab testing — showed a decrease in positive tests, from 8.6 per cent in 2009 to 3.5 per cent in 2012, while the workforce increased. 

Also, Suncor didn’t present evidence that showed the incidents for which the testing was done had proven to be the result of drug or alcohol use.

The board also found most of the incidents took place in the work camps, which had mostly contractor employees on site. As it turned out, only slightly more than one-half of one per cent of security incidents specifically referred to a unionized Suncor employee. 

Additionally, all three fatalities Suncor mentioned were contractor employees. Without evidence of significant use by unionized Suncor employees, it was unfair to paint them with the same brush by forcing them into random testing and the privacy violation it involved, said the board.

The majority of the arbitration board found Suncor did not prove drug and alcohol use was connected to the accident and near-miss history of its oilsands operations. Without this correlation, and the fact urinalysis does not demonstrate current impairment, the board found Suncor’s random testing policy was an unreasonable exercise of management rights.

Board didn’t consider all the evidence: Court

Suncor appealed the decision to the Alberta Court of Queen’s Bench, which found the arbitration was too strict in using the Supreme Court of Canada’s Irving decision to place limitations on establishing a workplace problem.  

While the Supreme Court allowed that random testing “might be justifiable if the employer could adduce evidence of a general problem with alcohol and drugs in the workplace,” the board said there needed to be evidence of a “significant” or “serious” problem. The court found this was “an unwarranted elevation” of this requirement.

The court also found the board incorrectly determined that a causal connection between 
alcohol and drug use and the accident evidence must be established, when no such threshold was contemplated by the Supreme Court in Irving.

In the court’s view, the board also made an error when it said it could only consider evidence demonstrating an alcohol and drug problem within the bargaining unit, not by non-unionized and contractor workers. Instead, the court found the workplace encompassed all workers — as did the Supreme Court in Irving — and they were integrated.

Given the policy only applied to two specific oilsands operations, and only workers in safety-sensitive positions, the inclusion of the whole workplace wasn’t overly broad, said the court.

“The focus on the workplace in general rather than more narrowly on members of the bargaining unit is also consistent with the obligations that employers have to ensure the safety of their entire worksite,” said the court. 

“The court is further comforted by the general workplace approach because it allows the dangerous environment to be considered in the context of the safety of everyone in that workplace.”

The court also found the arbitration board was too narrow in considering the evidence of incidents by Suncor workers. Most of the alcohol and drug-related incidents involved contractor workers or non-union employees, so the court didn’t use them as a basis for its decision. 

However, the test for determining a problem should include the entire workplace for safety reasons. This raises the number of incidents from 12 by  Suncor unionized employees to 2,276 at the Suncor worksites between 2003 and 2013.

“By focusing only on the bargaining unit, the majority (of the arbitration board) expressly excluded consideration of relevant evidence,” said the court. 

“(It) ignored evidence pertaining to some two-thirds of the individuals working in the oilsands operations. Neglecting this evidence gives rise to a reasoned belief on the part of this court that the (board) misunderstood the evidence in a manner that affected their decision.”

The appeal court quashed the arbitration board’s decision against Suncor’s alcohol and drug testing policy and remitted the case back for arbitration by a fresh panel.

For more information see:
•Suncor Energy Inc. v. Unifor, Local 707A, 2016 CarswellAlta 921 (Alta. Q.B.).
•Communications, Energy and Paperworkers Union of Canada, Local 30 v. Irving Pulp & Paper Ltd., 2013 CarswellNB 275 (S.C.C.).

Jeffrey R. Smith is the editor of Canadian Employment Law Today. For more information, visit

Latest stories