5.1. Testing before the job
Testing for drug or alcohol use sometimes takes place before a person is hired, transferred or promoted into a position (“pre-employment” or “certification” testing) or is allowed, as a contractor, to start work on a client’s job site (“pre-access” testing). The principles around these types of testing are similar.
The OHRC takes the position that drug and alcohol testing as part of the initial applicant screening process is prohibited under subsection 23(2) of the Code.
While the case law has not ruled out the possibility of testing for alcohol or drugs after a person receives a conditional offer of employment for a safety-sensitive position, the OHRC recommends against this practice.
If testing leads to refusing to hire someone based on an addiction or perceived addiction, it may be prima facie discriminatory. Negative consequences flowing from a positive test result may contribute to a finding that a job applicant has a “perceived” disability – even if he or she does not have an addiction.
If employers do put in place before-the-job drug and alcohol testing for safety-sensitive positions, they should make sure that a positive test result does not lead to automatically revoking the offer of employment or other negative consequences. The testing should be one part of a larger qualifying process, which could include examining the required licencing or other legitimate qualifications. The employer also must meet their duty to accommodate people with addictions.
Any medical testing should provide an effective assessment of the applicant’s ability to do the essential job duties. Pre-employment and pre-access drug and alcohol tests have been found to be insufficient to show that an employee has or will attend work impaired by alcohol or drugs. Because of this, if testing leads to negative consequences based on someone’s addiction or perceived addiction, it may be difficult to justify as a bona fide requirement.
5.2. Reasonable grounds and post-incident testing
“Reasonable grounds” (“for cause” or “reasonable cause”) and “post-incident” testing for either alcohol or drugs may be acceptable in specific circumstances, such as where there has been a link established between impairment and performing safety-sensitive job duties. “Reasonable grounds” should be informed by objective evidence, such as specific observed behaviours or other indicators, including:
- Seeing someone use alcohol or drugs at work
- An employee appearing or acting in a way that is consistent with someone impaired by alcohol or drugs (such as the person smelling like alcohol or drugs)
- Substances or substance paraphernalia in the vicinity of the employee or the area where the employee works (e.g. empty bottles).
Drug and alcohol testing is not automatically necessary for employees who appear impaired by drugs or alcohol. Other methods, such as allowing the person a chance
to explain their behaviour, temporarily removing them from safety-sensitive job duties to ensure immediate safety, offering accommodation to the point of undue hardship (such as referral to an employee assistance program or the support needed to attend a rehabilitative program), progressive performance management, and where there is objective evidence that there are legitimate reasons to be concerned, asking the person to attend a medical assessment, could address the issue. The OHRC’s Policy on preventing discrimination based on mental health disabilities and addictions has more information about these approaches.
An employer will have a legitimate interest in post-incident testing following accidents or reports of dangerous behaviour that have resulted in “near-misses,” and where looking at the condition of the employee is a reasonable part of the investigation. This may involve assessing if the employee consumed mind- or behaviour-altering substances that could have contributed to the incident. The inquiry could also include looking at other factors that may have contributed to the incident, such as lack of training, fatigue, or other factors that can increase risk.
Where a workplace accident or incident appears to result from external factors such as mechanical or structural failure or environmental factors, post-incident testing should not be conducted.
Both reasonable grounds and post-incident testing should only be conducted if they are necessary as part of a larger process of assessing drug or alcohol addiction. This process includes a broader medical assessment by a substance use disorder expert or under the care of a physician. Additional components of a larger assessment may include employee assistance programs (EAPs), peer reviews and supervisory reviews.
5.3. Random testing
Random on-the-job testing should be done only where a link has been established between impairment and performing job duties, such as in the case of employees
who are in safety-sensitive positions where the employer is able to demonstrate risk
in the workplace.
As stated earlier, the focus of drug and alcohol testing should be on determining actual impairment of an employee's ability to perform or fulfil the essential duties or requirements of the job at the time of the test. In random alcohol testing, the use of breathalyzers has been found to be permissible under the Code. Alcohol testing by breathalyzer is seen to be minimally intrusive (compared to blood tests, for example) and a highly accurate measure of both levels of consumption and actual impairment. Consequently, random alcohol testing is acceptable in safety-sensitive positions, but only where staff supervision is minimal or non-existent, there is evidence of risk in the particular workplace, and the employer meets its duty to accommodate the needs of people with addictions who test positive.
Although many technological advances have been made, the scientific research has not yet confirmed that a method of drug testing exists that is analogous to the breathalyzer for alcohol in terms of its:
- Ability to measure current impairment 
- High level of accuracy
- Minimal level of intrusion, and
- Rapid response time.
Drug testing methods that incorporate these criteria may help employers justify random testing of employees in safety-sensitive positions as a bona fide requirement, but only where staff supervision is minimal or non-existent, there is demonstrated risk in the particular workplace, and the employer meets its duty to accommodate the needs of people with addictions who test positive.
However, even drug and alcohol testing policies that may meet the requirements of the Code are vulnerable to challenge based on employee privacy.
In Irving, the Supreme Court of Canada considered the legal issue of whether implementing a random alcohol testing policy in a safety-sensitive workplace was a valid exercise of the employer’s management rights under the collective agreement.
The Court affirmed that random testing is not automatically justified on the basis that the workplace is dangerous and employees are in safety-sensitive positions. The Supreme Court held that while the dangerousness of a workplace is highly relevant, evidence of enhanced safety risks, such as evidence of a general problem with substance abuse in the workplace, is also required.
In that case, the majority held that implementing a random alcohol testing policy was not a valid exercise of the employer's rights. The testing policy was not justified because the risks to safety in the workplace did not outweigh the severe impacts on employees’ privacy.
The case pertained to employees’ rights under a collective agreement. However, the Supreme Court stated that even in a non-unionized workplace, “an employer must justify the intrusion on privacy resulting from random testing by reference to the particular risks in a particular workplace. There are different analytic steps involved, but both essentially require attentive consideration and balancing of the safety and privacy interests.”
5.4. Testing as part of a rehabilitation plan
Where an employee is returning to a safety-sensitive job after treatment for alcohol or drug addiction, post-reinstatement testing may be justified. An employee may
be expected to meet certain conditions when they return to work, which may include unannounced testing. Any conditions should be tailored to the person’s individual circumstances to meet the employer’s duty to accommodate. In such cases, the drug or alcohol testing period set should be reasonable and the testing frequency should not be overly onerous or intrusive.
Post-reinstatement testing may be part of a back-to-work agreement (e.g. a last-chance agreement or contingency behaviour contract), where breaching the agreement could result in terminating a person’s job. However, similar to people with other chronic disabilities, people with addictions may experience relapse after treatment. Having an agreement in place does not negate the employer’s duty to accommodate an employee if they have a relapse.
Example: A mill worker in a safety-sensitive position discloses a drug addiction after a work-related incident and attends treatment. One of the conditions of returning to work is that she has to undergo random drug testing. The employee passes her first drug test. Soon after, she has a car accident and starts using substances again. She fails a second drug test. Looking at all of the circumstances, including her prognosis, her accommodation plan and her recovery to date, the employer accommodates the relapse by giving the employee time off for further treatment while initiating another return-to-work plan.
At the same time, the employer’s obligation to accommodate is not limitless. An employer has a duty to accommodate a person with an addiction to the point of undue hardship. There may also be limited circumstances where an accommodation that otherwise would not amount to undue hardship is not required because it would fundamentally alter the nature of the employment, or it would still not allow the person to “fulfill the essential duties attending the exercise of the right.”
Example: An employee who is employed in a safety-sensitive position has a drug addiction and has repeatedly gone to rehabilitation. Despite multiple attempts to return him to work, he is not able to pass a post-reinstatement drug test. His doctor says that the employee is unable to work in the foreseeable future, and needs to take an undetermined leave of absence to attend rehabilitation. It would not cause undue hardship based on either cost or health and safety for the employer to continue to accommodate him. However, based on these unique circumstances, and after the employer has made repeated attempts to accommodate, the individual still cannot perform the essential duties of the job or perform alternative work and the employer’s duty to accommodate may end.
Or, after accommodation has been tried and exhausted, there may be no further accommodation available that will help the person complete the essential requirements of the job. There may also be situations where someone is continually unable or unwilling to take part in the accommodation process, despite the employer’s attempts. In these cases, the employer’s duty to accommodate may end.
 Mechanical Contractors Assn Sarnia, 2013, supra note 29.
 In Weyerhaeuser Company Limited v Ontario (Human Rights Commission) ex rel Chornyj, 2007 CanLII 65618 (Ont Div Ct) [“Chornyj”], a pre-employment drug testing policy was found not to be prima facie discriminatory based on perceived disability. The claimant, who admitted to using drugs recreationally, did not experience automatic dismissal or revocation of his employment and the company did not perceive him to be disabled. Therefore, his claim under the ground of “perceived disability” was not tenable.
 In Entrop, supra note 9, pre-employment drug testing by urinalysis was found to be a violation of the Code because, in the words of the Court, “…a positive test does not show future or even likely future impairment on the job, yet an applicant who tests positive only once is not hired” (at para 103). The Ontario Court of Appeal did not make a finding on pre-employment alcohol testing, as it was not included in Imperial Oil’s drug and alcohol policy. See also Mechanical Contractors Assn Sarnia, 2013, supra note 29 at paras 183, 217 and 218. In Mechanical Contractors Assn Sarnia, 2014, supra note 29, the Ontario Divisional Court upheld the arbitrator’s analysis of the collective agreement issue, but did not find it necessary to address the Human Rights Code issue.
 Employers may also be expected to show the particular safety risks that exist in the workplace. See Mechanical Contractors Assn Sarnia, 2014, supra note 29; Irving, supra note 7 at para 20.
Irving, supra note 7 at paras 30, 45.
 Adapted from ENFORM, “Alcohol and Drug Policy Model for the Canadian Upstream Petroleum Industry” Ed. 1.0. (2007), online: ENFORM The Safety Association for Canada’s Upstream Oil and Gas Industry www.enform.ca/resources/detail/70/alcohol-and-drugs-policy-model-for-the-upstream-petroleum-industry at 18 (retrieved August 4, 2016).
 Ontario Human Rights Commission, Policy on preventing discrimination based on mental health disabilities and addictions, (Toronto: Queen’s Printer for Ontario, 2014) online: Ontario Human Rights Commission www.ohrc.on.ca/en/policy-preventing-discrimination-based-mental-health-disabilities-and-addictions.
 Entrop, supra note 9 at para 114. See also Sterling Crane,  OLRD No 4623 (QL) [“Sterling Crane”], in which the Ontario Labour Relations Board noted: “It is apparent that the jurisprudence now accepts the authority of an employer to conduct post incident urinalysis testing in a safety-sensitive workplace both as a legitimate exercise of management rights and as a BFOR under human rights legislation, where that testing is conducted as part of an investigation to determine the cause of the incident in question. I agree that a policy providing for such testing constitutes a legitimate exercise of management's authority so long as the testing does not become random” (at para. 74).
 Irving, supra note 7 at paras 20, 45.
 See Entrop, supra note 9 at para 99; Imperial Oil Ltd v Communications, Energy and Paperworkers Union of Canada, Local 900  OLAA No. 721 (QL) at para 124, [“Imperial Oil Ltd, 2006”]; Imperial Oil Limited v Communications, Energy & Paperworkers Union of Canada, Local 900, 2008 CanLII 6874 (Ont Div Ct), [“Imperial Oil Ltd, 2008”], and Imperial Oil Ltd. v Communications, Energy & Paperworkers Union of Canada, Local 900, 2009 ONCA 420 (CanLII) [“Imperial Oil Ltd, 2009”].
 Entrop, supra note 9.
 See section 7.1. on the duty to accommodate for more information.
 Urinalysis, for example, can detect past use, but it cannot detect how much of the drug was used, or whether the person is currently impaired. See Frone, supra note 3; Leo J. Kadehjian, “Specimens for Drugs-of-Abuse Testing” In Forensic Science and Medicine: Drugs of Abuse: Body Fluid Testing, ed by R.C.Wong & H.Y. Tse (Totawa, NJ: Humana Press, 2005). See also Entrop, supra note 9. Oral fluid testing techniques are rapidly improving [see, for example, Nathalie A. Desrosiers et al., “On-Site Test for Cannabinoids in Oral Fluid” (2012) 58:10 Clinical Chemistry 1418]. There are still limits to establishing impairment through oral fluid at the time of testing, due to a variety of factors. See Marilyn A. Huestis, et al., “Oral Fluid Testing: Promises and Pitfalls,” (2011) 57:6 Clinical Chemistry 805; V. Vindenes, et al., “Detection of Drugs of Abuse in Simultaneously Collected Oral Fluid, Urine and Blood from Norwegian Drug Drivers” (2012) 219 Forensic Science International 165; Frone, supra note 3; Alain Verstraete, “Detection Times of Drugs of Abuse in Blood, Urine, and Oral Fluid” (2004) 26:2 Ther Drug Monit 200; Australian National Council on Drugs, ANCD Position Paper: Drug Testing (2013) online: ATODA www.atoda.org.au/archived-ancd-website/ at 4-5; Scott Macdonald, “Submission to the Society of Energy Professionals and the Power Workers’ Union Comment on the Canadian Nuclear Safety Commission Discussion Paper Fitness for Duty: Proposals for Strengthening Alcohol and Drug Policy, Programs and Testing” (2012), online: www.suretenucleaire.gc.ca.eng/pdfs/Discussion-Papers/12-03/20120919-DIS-.... Blood testing may be better than other methods at detecting levels of drugs associated with impairment but it is highly intrusive. See Leo J. Kadehjian, ibid.; Macdonald, 2010, supra note 5. A breathalyzer for drugs is being developed. See Olof Beck, “Exhaled Breath for Drugs of Abuse Testing – Evaluation in Criminal Justice Settings” (2014) 54 Science and Justice 57; Sarah K. Himes et al., “Cannabinoids in Exhaled Breath following Controlled Administration of Smoked Cannabis” (2013) 59:12 Clinical Chemistry 1780. However, there is not a large body of scientific literature validating its use.
 In Imperial Oil Ltd, 2008, supra note 44, the Ontario Divisional Court stated, “There is no dispute that the current drug test [oral fluid] does disclose impairment by cannabis, although the result of the test is not available at the time it is administered” (at para. 10). However, in the scientific literature, limitations are still evident (see footnote 47, above).
 In Imperial Oil Ltd, 2006 supra note 44, testing for drugs in oral fluid using a cheek swab was found
not to be justifiable under the collective agreement partly because the results could not be immediately determined – the test had to be sent to a laboratory for analysis. Employees who would eventually receive a positive result were sent back to their safety-sensitive jobs right after being tested. As such,
this method could not ensure immediate safety in the workplace, and was not analogous to the alcohol breathalyzer, which was allowed by the Ontario Court of Appeal in Entrop (at paras. 112-113). This finding was upheld in Imperial Oil Ltd, 2008, supra note 44 and Imperial Oil Ltd, 2009, supra note 44. Drug testing guidelines advise that initial screening tests for oral fluid be analysed in a laboratory, confirmed using high standard laboratory techniques, and verified by a trained medical review officer.
See European Workplace Drug Testing Society, European Guidelines for Workplace Drug Testing in Oral Fluid (2011) online: EWDTS www.ewdts.org/ewdts-guidelines.html (retrieved July 21, 2015) at 6; and Substance Abuse and Mental Health Services Administration, “Mandatory Guidelines for Federal Workplace Drug Testing Programs – Oral Fluid” (proposed) (2015) 80:94 Federal Register online: Department of Health and Human Services www.samhsa.gov/workplace/drug-testing#proposed-mandatory-guidelines (retrieved July 22, 2015) at 28086. .
 There is some disagreement about the permissibility of pre-employment and random drug testing in human rights case law across jurisdictions. For example, despite the CHRT’s finding that urinalysis drug testing for cannabis did not indicate impairment on the job, in Milazzo, supra note 34, it was found to be “reasonably necessary” to meet the legitimate work-related purpose. This was because bus drivers were unsupervised most of the time, and a positive test result, although not conclusive, was a “red flag” that helped to identify drivers who were more likely to have accidents. Drug and alcohol testing served as a deterrent to employees who had control over their drug and alcohol use, but there was insufficient evidence to conclude that individuals with substance abuse problems would be deterred. The policy also reflected requirements to comply with U.S. legislation. However, the CHRT found that the employer’s policy failed to accommodate people who tested positive to the point of undue hardship. See also Chornyj, supra note 36; Dennis v Eskasoni Band Council  CHRD No 38 (QL); Alberta (Human Rights and Citizenship Commission) v Kellogg Brown & Root (Canada) Company, 2007 ABCA 426, (leave to appeal to SCC denied).
 Irving, supra note 7.
 Irving, supra note 7, at para 31.
 In Irving, supra note 7, the dissenting minority noted in footnote 2 (para 86):
“While Entrop was decided in the context of a non-unionized workplace under human rights legislation, it remains relevant to any analysis concerning the reasonableness of drug and alcohol testing policies. Indeed, the board here relied on Entrop in assessing the invasiveness of the breathalyser test (para. 116). Whether an arbitrator applies the test developed by this Court for the human rights context in British Columbia (Public Service Employee Relations Commission) v. BCGSEU, 1999 CanLII 652 (SCC),  3 S.C.R. 3 (“Meiorin”), or traditional labour relations law and the KVP test, at bottom, the inquiry in both cases is concerned with the reasonableness of the company policy. In some provinces, arbitrators may adjudicate grievances challenging these policies under both KVP and Meiorin and we have difficulty accepting that a policy would fail under one test but pass muster under the other. See, e.g., Labour Relations Act, 1995, S.O. 1995, c. 1, Sch. A, s. 48(12)(j).”
 Irving, supra note 7, at para 20.
 Entrop, supra note 9. The Ontario Court of Appeal did not find it necessary to interfere with the Board
of Inquiry’s conclusion that post-reinstatement testing did not contravene the Code, as long as it was necessary as part of a larger assessment to determine whether an employee was not merely using but abusing alcohol or drugs (at para. 129).
 In a review of the literature, McLellan, et al. found that the relapse rates for substance dependence (which includes alcohol and drugs) were between 40-60% in the first year after discharge from treatment, which was similar to the relapse rates for type 1 diabetes (30-50%), hypertension (50-70%) and asthma (50-70%). McLellan, et al., “Comparison of Relapse Rates Between Drug Addiction and Other Chronic Illnesses” (2000) 284 JAMA 1689 at 1693.
Colonial Cookies Corp v United Food and Commercial Workers Canada, Local 175 (Grant Grievance),  OLAA No. 468 (QL). In Milazzo, supra note 34 at para 34, the CHRT stated:
“Accordingly, the ‘last chance agreement’ is in the Tribunal's opinion unenforceable in regards to the [Canadian Human Rights] Act. As the case law indicates, an analysis must be made in each case to determine whether or not it is impossible for the employer to accommodate the needs of the employee to the point of undue hardship. While it is certainly open to the Respondent to warn employees returning to work after rehabilitation that any relapse could result in termination of there [sic] employment, the imposition of a last chance agreement cannot serve to nullify the duty of accommodation established under human rights legislation.”
58 Hydro-Québec v Syndicat des employé-e-s de techniques professionnelles et de bureau d’Hydro-Québec, section locale 2000,  2 SCR 561; Hall v Chief of Police, Ottawa Police Service, 2008 CanLII 65766 (Ont Div Ct).
 Section 17 of the Ontario Human Rights Code.