The arbitration decision on the legality of Suncor's random drug and alcohol policy has been released. In a 2 to 1 decision, the Arbitration Board refused to allow the policy to be implemented. The dissenting member found that Suncor had overwhelmingly demonstrated evidence of a serious problem with drug and alcohol use in the workplace as well as an enhanced safety risk, meeting the Supreme Court of Canada's test in its recent Irving decision. Suncor has already announced it intends to appeal the decision.
In May, 2012, Suncor announced that it was introducing random drug and alcohol testing at its Oil Sands facilities. Implementation was to begin in October 2012. Unifor Local 707a, a union representing thousands of Suncor workers at several sites, filed a grievance against the testing program. It also obtained an injunction preventing Suncor from introducing the random testing until an Arbitration Board ruled on its grievance. As discussed in a previous story in HR Space, "Temporary Injunction Against Random Drug and Alcohol Testing", Suncor's appeal of the injunction was rejected in December 2012.
The arbitration case proceeded along a parallel track with the injunction case. A three-member Board heard evidence over 23 days throughout 2013.
The Supreme Court's July 2013 decision in Communications, Energy and Paperworkers Union of Canada, Local 30 v. Irving Pulp & Paper Ltd. also informs the arbitration. In Irving, the Supreme Court ruled that the employer's random alcohol testing policy was not justified on the facts of that case. The Court did, however, leave the door open for the validity of random testing in different circumstances.
As in Irving, the main issue in the Suncor arbitration was whether the benefits of the random testing program were proportionate to the harm caused in violating employees' privacy rights. In particular, the Board evaluated whether Suncor's evidence met the threshold necessary to prove that drug and alcohol use was a "general problem in the workplace," and that it posed "legitimate safety concerns."
Proportionality of Random Testing
The Board found that Suncor failed to prove that drug and alcohol usage was a significant problem or legitimate safety risk on its worksites. This was not due to the quantity of evidence Suncor provided- - - Suncor had introduced evidence of 2, 276 incidents it attributed to drugs and alcohol, as well as testimony and reports from multiple experts and employees- - - but rather the quality of this evidence. The Board preferred the evidence of the Union's experts. They also found that much of the evidence Suncor presented was either problematic or lacked applicability to the Union members grieving the policy. They noted Suncor's failure, in the evidence provided, to differentiate between types of employees, length of employment, work locations and work areas within sites.
The Board also criticized Suncor's policy for failing to first exhaust other, less intrusive means to achieve safety (i.e. random checks, improvement of existing programs etc.) and for its inconsistent standards for drug and alcohol use. The policy did not prevent alcohol consumption immediately before a shift, while simultaneously having zero tolerance for drugs.
Finally, the Board found that Suncor's proposed method of testing, urinalysis, was a more invasive and inaccurate form of testing than the use of oral swabs. Such a testing method, which might have been acceptable when doing "for-cause" testing, was not reasonable in a random testing scenario.
Taken together, these issues caused the Board to side with the Union, preventing Suncor from introducing its random testing policy.
Lessons for Employers
Following the Supreme Court's lead in Irving, the Board did not prohibit random drug and alcohol testing in all circumstances. It based its decsion on the evidence before it. The Board noted that, had it been given jurisdiction to determine what would be a reasonable policy, it would have applied the Drug and Alcohol Risk Reduction Pilot Project ("DARRPP") principles. They involve, among other things, a time-limited pilot project, measurement of effects and results (including false positives), a dispute resolution mechanism, a clear and unequivocal "under the influence of alcohol or drugs" prohibition, consistent training and using oral fluid as the method of testing. A policy guided by the DARRPP principles would be applied initially to employees with less than 2 years of service and coincide with a peer-based initiative. It would be reviewed and evaluated every six months to allow for modification based upon actual experience and good data.
The decision provides guidance as to what evidence arbitrators or courts may in the future be looking for to justify a random drug and alcohol testing policy:
- The policy should be consistent. If jobs are safety-sensitive such that there is no tolerance for drugs, this "no tolerance policy" should extend to alcohol;
- Employers should use random testing on select worksites rather than across all of its operations. As random testing is more invasive than for-cause testing, it is more likely to be upheld for an individual facility where problems have been experienced, rather than a policy applying to multiple facilities with varying types of conditions, employees and drug and alcohol use patterns;
- The method for random testing under the policy should be the least invasive method possible;
- The evidence presented to justify the policy should be as narrow and explicit as possible. In particular:
- Any statistics relied on should be as concrete and specific as possible. This includes delineating between the type of employee involved in an incident (union, non-union or contract), seniority and the worksite involved;
- Comparisons with worksites and communities that do not have endemic drug and alcohol problems are also useful.
- Non-events should be weeded out of the evidence in order to give credibility to the statistics; and
Finally, employers should stay tuned. The strong dissenting opinion disagreed with the majority on a number of major points, not least the amount and type of evidence necessary to meet the test laid out by the Supreme Court in Irving. Suncor has already indicated its intention to appeal, so there is little doubt that the law will continue to evolve in this area.