Assessing Random Drug and Alcohol Policies: Recent Guidance from BC

  • May 10, 2018
  • Tyler Boggs

The recent case of USW Locals 7884 and 9346 v Teck Coal Ltd (Fording River and Elkview Operations)[1] affirms the general arbitral consensus that, absent sufficient evidence of a workplace problem with drug and alcohol use, the implementation of a random drug and alcohol testing policy is likely to be dismissed as an unreasonable exercise of management rights. Arbitrator Kinzie's analysis reinforces the fundamental nature of an employee's right to personal privacy, and is consistent with other recent authorities on this issue.

What separates Teck Coal from other related decisions, however, is the arbitrator's unique approach to the evidence placed before him, his rejection of a commonly accepted analytical approach to such disputes, and his creation of a new legal test for evaluating the reasonableness of a unilaterally introduced random drug and alcohol testing policy.


Consistent with other recent authorities, the Union in Teck Coal argued that the applicable legal test was the one endorsed by Justice Abella in Irving Pulp and Paper Ltd., which applies the following general arbitral consensus on random drug and alcohol testing policies:

"[…] a unilaterally imposed policy of mandatory, random and unannounced testing for all employees in a dangerous workplace has been overwhelmingly rejected by arbitrators as an unjustified affront to the dignity and privacy of employees unless there is reasonable cause".[2]

Arbitrator Kinzie rejected this argument on the basis that the Supreme Court did not articulate a new legal test by tendering this commentary. Rather, as recognized by the dissenting justices in Irving Pulp and Paper Ltd., adjudicators are free to depart from the general arbitral consensus described above where a reasonable rationale for doing so exists.[3]

On this understanding, Arbitrator Kinzie opted to create his own test for assessing the reasonableness of the unilaterally introduced random drug and alcohol testing policy before him. His analysis proceeded in the following manner:

  1. Have employees' privacy rights been infringed and, if so, to what extent?
  2. Is there sufficient or adequate cause to justify the search, seizure, and resulting privacy intrusions resulting from the random testing?
  3. If so, is random testing a proportionate response to a demonstrable workplace problem?[4]

Arbitrator Kinzie noted that a failure on behalf of the employer to establish sufficient or adequate cause to justify the impugned policy's introduction at stage 2 of the test would be fatal to its case.


Arbitrator Kinzie found with little difficulty that the employees' privacy rights had been infringed by the introduction of the policy at issue.[5] Accordingly, the first stage of the test was satisfied.

Proceeding to the second stage, Arbitrator Kinzie took a permissive approach to the evidence and determined that evidence from the employer's operations throughout the Elk Valley Region was relevant to the matter before him. Interestingly, the Arbitrator accepted that this included evidence of the relevant circumstances of non-bargaining unit members and workers to whom the policy did not apply, including contractors.[6]   

This did not assist the employer's claim, however. Distinguishing between a "risk" and a "problem", Arbitrator Kinzie noted that the evidentiary burden placed on an employer under the second stage of the test requires evidence of a "problem" concerning an actual state of affairs, which is deemed “difficult to control” or “unruly” in nature.[7] Though managing risk may constitute a sufficient justification in typical disputes concerning an exercise of management rights, Arbitrator Kinzie held that a higher degree of safety concern is required where privacy rights are engaged.[8]

The union was able to rebut the employer's assertion that there was a systemic safety concern throughout its operations. The union presented data which demonstrated that the number of positive post-incident and reasonable cause tests in the workplace was relatively insignificant. Moreover, a downward trend in the employer's premium rate payments to the Workers' Compensation Board of British Columbia was accepted as evidence of the absence of a problem with drugs, alcohol, or safety in the workplace. In fact, such evidence confirmed the union's position that there was insufficient cause to justify the policy's introduction:

"[…] The evidence considered as a whole does not demonstrate that the Employer was facing a 'difficult to control' or 'unruly' situation in this regard. What we have in fact […] is an improving situation on top of an already relatively safe operation".[9]

The employer had therefore failed to satisfy stage 2 of Arbitrator Kinzie's test.

Although unnecessary to do so in light of the above, Arbitrator Kinzie also considered the proportionality of the policy. Relying on the Supreme Court's commentary in Irving Pulp and Paper, Ltd., supra, the arbitrator maintained that an employer must provide cogent evidence of a demonstrable workplace problem. Given his findings at stage 2 of the test, the arbitrator concluded that the safety benefits arising from the policy were not proportionate to the harm that would inure to employees’ privacy rights. The employer was therefore not justified in implementing the random drug and alcohol testing policy.


Though the extent to which Arbitrator Kinzie's three-stage test will be adopted in subsequent jurisprudence remains unknown, the decision is notable given its comprehensive – and lengthy – review of the applicable jurisprudence on this issue. In striking down the policy before him, Arbitrator Kinzie issued a decision that broadly affirms the Canadian approach to such matters, which requires proof of a "general" problem in a workplace before intrusions on employee privacy will be tolerated.

The decision reinforces that disputes with respect to the enforceability of random testing policies are largely fought and won (or lost) on the facts. Arbitrator Kinzie's approach to the evidence presented by the parties provides useful guidance to counsel seeking to defend or challenge such policies in the future. As this area of the law continues to evolve, it will be interesting to see how adjudicators grapple with the "risk" vs "problem" distinction referred to in the decision. Undoubtedly, the evidentiary threshold required to establish a "problem" sufficient to justify the introduction of a random drug and alcohol testing policy will continue to be a hotly contested issue.


Tyler Boggs is an associate at Cavalluzzo LLP. He practices primarily in the areas of labour law, human rights, and constitutional law. He joined the firm as an associate in 2017 after summering and articling with the firm.


[1]          USW Locals 7884 and 9346 v Teck Coal Ltd (Fording River and Elkview Operations), 134 CLAS 126, 286 LAC (4th) 1 ["Teck Coal"].

[2]          Communications, Energy and Paperworkers Union of Canada, Local 30 v Irving Pulp & Paper, Ltd., 2013 SCC 34 at para 6, [2013] 2 SCR 458 ["Irving Pulp & Paper"].

[3]          Teck Coal, supra note 1, at para 266.

[4]          Ibid at para 293.

[5]          Ibid at para 326.

[6]          Ibid at paras 331-333. On the issue of contractors, however, Arbitrator Kinzie clarified that such evidence would be entitled to little weight. In his view, if there was evidence of a problem among only a contractor's employees, it would be more reasonable for an employer to terminate its relationship with that particular contractor than to require its own employees to submit to testing absent sufficient cause.

[7]          Ibid at para 353.

[8]          Ibid at para 355.

[9]          Ibid at para 384.