Alberta Arbitrators Strike Oil Company’s Random Alcohol and Drug Testing Policy

28 mars 2014

( Disponible en anglais seulement )

Employers looking to Alberta for a legal precedent to support their random alcohol and drug testing policies will need to keep looking, after Alberta arbitrators deemed Suncor Energy Inc.’s random testing policy to be an unreasonable exercise of Suncor’s management rights.

Industry observers had been following Unifor, Local 707A v Suncor Energy Inc closely after the Supreme Court of Canada’s decision last year in Irving[1], which determined that employers must meet a high standard to justify random alcohol testing of employees. Until the Suncor decision, which was released last week, the high Irving threshold had been challenged only once in Canadian law.

In that case (Sarnia[2]), an Ontario arbitrator applied the Irving standards and struck an alcohol and drug policy, because not enough specific evidence was provided to justify testing contractors for alcohol and drugs before allowing them to enter the employer’s worksite. (We published articles about Irving and Sarnia last year.)

Historically, Alberta arbitrators and courts have been more tolerant of alcohol and drug testing policies than have their colleagues in other provinces[3], and given that alcohol and drug use is commonly thought to be prevalent among Alberta’s oil patch workers (whether accurate or not), many observers believed the Suncor decision would provide an example of a random alcohol and drug testing policy that is deemed reasonable in Canadian law. As Suncor argued in the arbitration, Suncor has an alcohol and drug problem among its employees at the impugned Alberta workplace that is pervasive, profound and “unparalleled in any other workplace in Canada”[4].

However, two of the three arbitrators (the “Board”) in the Suncor case held that the Policy was an unreasonable exercise of Suncor’s management rights. The case provides a further precedent to demonstrate how difficult it is for employers to institute random alcohol and drug testing in Canada. It must be noted, however, the decision included a dissent, and Suncor indicated it will seek judicial review.


In June 2012, Suncor announced it was amending its alcohol and drug testing policy for employees in two of its massive oil sands operations in the Fort McMurray, Alberta area. Its existing policy, which provided for post-incident, reasonable grounds, follow-up, post-rehabilitation, and return to work alcohol and drug testing (commonly cited in the jurisprudence as “for cause” testing) was amended to include random alcohol and drug testing of employees deemed to work in “safety sensitive” positions (the “Policy”). Of Suncor’s 3,383 unionized employees (as of July 2013), 2,771 (82 percent) of them were deemed to occupy safety-sensitive positions, which was a fact the Union did not challenge.

Of note, prior to making the decision to implement the Policy, Suncor had participated in the development of the Drug and Alcohol Risk Reduction Pilot Project (DARRPP), a two-year trial evaluation project monitoring the introduction of random workplace alcohol and drug testing. However, around the same time DARRPP was announced, Suncor announced it would proceed with its own policy. The Union filed a policy grievance in July 2012, stating the “random testing standard was unreasonable and unjustified”.

Suncor intended to implement the Policy on October 15, 2012. On October 12, 2012, the Union applied for and received an injunction to prevent implementation of the Policy, pending the result of the arbitration[5]. The Court of Appeal upheld the injunction in December 2012[6]. Irving came out in June 2013, setting an uphill road for Suncor to justify the Policy.

The arbitration took 23 days. The parties adduced 19 witnesses, including four experts. A total of 188 pages of written arguments were submitted.

The Union’s Position

Relying heavily on Irving, the Union argued that Suncor failed to meet its onus to demonstrate that the benefits to be gained through the implementation of the Policy (a reduction of the safety risk at the workplace) were proportional to the harm that will occur as a result of the implementation (an infringement of the tested employee’s privacy)[7]. In particular, the Union argued:

  • Suncor did not demonstrate a “significant safety concern” at the impugned workplace with respect to the bargaining unit;
  • The Policy is ineffective as a means to reduce safety risks at the impugned workplace; and
  • Suncor had not exhausted less intrusive, effective means to address its drug and alcohol use concerns at the impugned workplace.

Suncor’s Position

Suncor stated and adduced evidence that there is an “out-of-control drug culture” in the Fort McMurray area, and that cocaine use was a particular concern[8]. Suncor also highlighted, among other things:

  • In the nine year period between October 1, 2003 and December 31, 2012, there were 224 positive alcohol and drug tests of Suncor Employees at the impugned worksites, and of the 224 positive tests, 216 were Union employees[9];
  • There have been 20 fatalities at the impugned worksites, and alcohol or drugs were factors in three of them (although the last death of a bargaining unit employee occurred in 1994)[10];
  • Between 2004 and August 2013, there were 2,276 security incidents involving alcohol and drugs, including the discovery of devices used to defeat urine drug tests, such as whizzinators, bottles of urine and urine testing kits[11];
  • In 2009 through 2012 there were 115 positive employee alcohol and drug tests at the impugned workplace compared to only five positive alcohol tests and zero positive drug tests at all of Suncor’s other operations in Canada[12];
  • Suncor is subject to onerous statutory obligations to ensure a safe workplace, including those imposed by the Alberta Occupational Health and Safety Act and the Criminal Code[13];
  • Suncor has tried other methods to address its concerns about alcohol and drug use at the impugned workplace, including the use of sniffer dogs and extensive safety training and education with respect to alcohol and drugs[14]; and
  • The Union’s grievance did not specifically refer to the testing methods and procedures Suncor planned to use to implement the Policy, and, as such, the Board did not have jurisdiction to analyze whether the methods and procedures Suncor planned to use to carry out testing (urinalysis, in particular) were reasonable.


As set out in Irving, the Board followed a “balancing of interests proportionality approach” to determine whether the Policy’s claimed benefits of a reduced safety risk outweighed the harm to employee privacy[15].

With regard to Suncor’s argument that the Board did not have jurisdiction to determine the reasonableness of the method and procedures Suncor planned to use when it implemented the Policy (because the Union did not include this as part of its grievance), the Board stated, “[b]oth the fact of random testing, and the method used to random test are relevant in the proportionality assessment this Board must undertake[16].”

In analyzing the procedures and methods of testing, the Board focused on urinalysis testing and Suncor’s claim that random urinalysis testing would improve workplace safety. However, the Board highlighted the differences between urinalysis testing in “for cause” testing and in random testing. Whereas a positive “for cause” urinalysis test provides only one piece of evidence to use to build a case against an employee and improve safety at the workplace, the random test is purported to determine if the tested employee is fit for work, but a positive urinalysis drug test does not gauge current impairment. The Board did not find Suncor’s evidence persuasive that a positive urinalysis test creates a “red flag” and identifies “high risk employees” who had recently used drugs[17].” As such, the Board cited the failure of urinalysis drug testing to confirm current impairment as one factor that contributed to the unreasonableness of the Policy.

The Board also noted “for cause” testing only occurs after a triggering incident, but the Policy states a 50 percent minimum of the 2771 bargaining unit members would be randomly tested each year, which was another reason the Board cited to indicate the Policy was not proportional[18].

With regard to what constitutes “a problem sufficient to justify random alcohol testing”, the Board Majority indicated the Court in Irving had provided “very little guidance” on what is required to justify a privacy infringement[19]. Nevertheless, the Board stated that Suncor’s evidence, which included positive “for cause” testing results, statistics of security incidents and evidence of alcohol and drug problems in the community-at-large and of non-bargaining unit employees, was not specific enough to meet the Irving standard, vague as it is. For example, the Board noted that the reference to three fatalities linked to drugs or alcohol involved contractor employees and not bargaining unit members, and the Board’s decision must concern only the employers’ relationship with the bargaining unit members. In this manner, with regard to the specificity of evidence required, the Board decision echoed the decision in Sarnia.

The Board also noted the high volume of positive alcohol and drug test results at the impugned workplace compared to other Suncor worksites in Canada neither provided enough details of the testing nor addressed how many Suncor employees worked at the impugned workplace relative to how many Suncor employees worked at other sites in Canada.

The Board also stated Suncor failed to make a causal connection between alcohol/drug problems and accidents, injuries and near misses at the impugned workplace, as required in Irving[20].

The Board further noted that the steps Suncor had been taking to improve safety before attempting to implement the Policy were working—the incidence of positive “for cause” tests were decreasing, and the safety record was improving, which found to be was evidence that the Policy was unreasonable[21]. In particular, the Board noted that the results of positive tests taken from 2012 and 2013 indicated that, statistically, Suncor’s incidences of positive tests were fewer than the amount presented in Irving[22]. In general, the Board felt that the evidence was “simply lacking” that an out-of-control a drug or alcohol culture exists with respect to the impugned bargaining unit members[23].

As an aside, the Board endorsed the DARRPP principles with regard to alcohol and drug policies. These principles include a time-limited trial project prior to implementation, measurement of effects and results (including false positives), a dispute resolution mechanism for addressing employee concerns or complaints, a clear and unequivocal “under the influence of alcohol or drugs” prohibition, consistent training, the use of oral fluids as the testing method (which can confirm current impairment for both alcohol and drugs), and regularly scheduled (every six months) re-evaluations with regard to expanding, reducing, maintaining or discontinuing the policy based on clear data indicating results[24].

What does the Decision mean?

  • While it remains difficult (but still possible) for employers in Canada to justify random alcohol testing of employees, it is even more difficult to justify random drug testing, particularly random drug testing by urinalysis.
  • To justify the implementation of random testing for alcohol and drugs, employers are required to provide specific evidence of a drug and alcohol problem among the members of the bargaining unit who are subject to the random testing.
  • The evidence of a workplace problem with alcohol and drugs must be causally linked to accidents, injuries or near misses at the workplace.
  • “For cause” testing for alcohol and drugs, even urinalysis drug testing “for cause”, is reasonable and does not appear to be controversial when it is only used as one piece of evidence and is not used to decide an issue determinatively.
  • Employers may look to DARRPP principles and possibly apply them if they wish to implement random alcohol and drug testing policies.

Given Suncor’s intentions to seek judicial review, it is possible the reviewing court may quash the majority decision. However, the current state of the law indicates employers should read Irving, Sarnia and the Suncor decision carefully before attempting to implement random alcohol and drug testing policies.

[1] Communications, Energy and Paperworkers Union of Canada, Local 30 v Irving Pulp & Paper, Ltd, 2013 SCC 34 (“Irving”)

[2] Mechanical Contractors Association Sarnia (Re) 2013 CanLII 54951 (ONLA) (« Sarnia »)

[3] Sarnia, supra, para 110

[4] Para 158

[5] CEP, Local 707 v Suncor Energy Inc. 2012 ABQB 627 (Macklin, J)

[6] CEP, Local 707 v Suncor Energy Inc. 2012 ABCA 373

[7] Paras 136 and 199.

[8] Para 179.

[9] Para 51

[10] Paras 36 to 39

[11] Para 53.

[12] Para 268.

[13] Para 163.

[14] Para 168.

[15] Para 223.

[16] Para 216.

[17] Paras 291 and 295.

[18] Paras 201 to 206.

[19] Para 239.

[20] Para 312.

[21] Para 250.

[22] Pars 252.

[23] Para 320.

[24] Para 346.

Avis de non-responsabilité

Les renseignements affichés sur ce blogue contiennent des points de droit variés fournis uniquement à des fins informatives et non commerciales. Ces renseignements ne constituent pas un avis juridique de la part de l’auteur. Nous mettons en garde les lecteurs de ne pas prendre de décision particulière sans avoir préalablement obtenu l’avis juridique d’un professionnel qualifié. Toute personne qui décide de prendre une décision en s’appuyant sur ces renseignements le fait à ses propres risques.