When a physician or nurse is diagnosed with a substance use disorder, three things are at stake at once: patient safety, the professional's licence and career, and the organization's liability. Removing the person permanently protects no one in the long run. It loses a highly trained clinician, discourages others from coming forward, and pushes the illness underground.

Medical monitoring offers a different path. The professional completes treatment, then returns to practice under a structured agreement with random testing, clinical follow-up and regular reporting. The evidence for this model is unusually strong.

What the outcome studies show

United States. The largest study followed 904 physicians admitted to 16 state physician health programs. Over the five-year monitoring period, 78% had no positive test for alcohol or drugs, and 72% were still practising medicine at follow-up [1]. A companion analysis of the same cohort found that 631 of 802 physicians with known outcomes (79%) were licensed and working at five years [2].

Canada. The Ontario Physician Health Program reviewed its first 100 doctors, each on a five-year monitoring contract that included addiction medicine follow-up, support groups, workplace monitoring and random urine testing. 71% had no known relapse, a further 14% completed the program after a relapse, and 85% completed successfully [3].

Across studies. A 2021 meta-analysis pooled 29 studies of monitoring for health professionals. It found 72% abstinence and 77% work retention at follow-up. Abstinence was higher when monitoring began after treatment was completed (79%) than when it began at the start of treatment [4].

Nurses: the same pattern, with lessons on design

The largest nursing study tracked 7,737 nurses in 13 US monitoring programs between 2007 and 2015. About 61.5% completed their program. The strongest predictors of completion were program features, not just the individual [5]:

• More random testing. Each additional drug test selection raised the odds of completion. Benefits levelled off at about 26 tests a year.
• Enough time. Completion rose with time in the program. The authors recommend a minimum of three years.
• Zero relapses. A relapse was the strongest negative predictor, which is why early detection matters.
• Daily check-ins and support meetings. Both tracked with higher completion.

In Canada, regulators are moving the same way. Ontario's Nurses' Health Program, launched in 2019 by the College of Nurses of Ontario and the province's nursing associations, gives nurses with substance use or mental health disorders a voluntary path to recover and return to safe practice outside the disciplinary process [6]. In British Columbia, BCCNM has described a case in which a nurse returned to work with medical monitoring and temporary narcotics restrictions, set out in a consent agreement that mirrored her workplace accommodations [7].

What makes monitoring work

Across the research, effective programs share the same building blocks:

• Treatment first. Monitoring supports recovery. It does not replace treatment.
• Frequent, truly random testing with chain of custody, laboratory confirmation and Medical Review Officer review, so results are accurate and defensible.
• Multi-year duration. Five years is standard for physicians in both the US and Ontario studies.
• Clinical and peer support, including addiction medicine follow-up and mutual support groups.
• Workplace conditions where needed, such as restricted access to controlled drugs.
• Clear, consistent reporting so the employer, regulator or insurer can act on facts.

How Canada differs from the US

Most of the published evidence comes from US programs that are closely tied to state licensing boards. In Canada, the structure varies by province and profession. Physician health programs operate provincially, nursing regulators use their own processes, and employers carry a duty under human rights law to accommodate a substance use disorder as a disability, up to undue hardship. Reporting of health information also requires the participant's informed consent under federal and provincial privacy law. A Canadian monitoring program has to fit all of these, not just copy the US model.

What this means for employers and insurers

For a hospital, health authority, clinic or disability insurer, monitoring turns an uncertain return to work into a documented, measurable one. It protects patients through early detection, gives the professional a fair and structured path back, and gives the organization evidence that it took reasonable steps and met its duty to accommodate.

The studies above are observational. The 2021 meta-analysis notes that differences between programs and publication bias limit firm conclusions, and no randomized trials exist. Even so, the consistency of results across countries, professions and decades is hard to ignore.

Next: What substance use costs Canadian employers, and how to reduce it

Substance use is one of the most expensive problems a Canadian employer can have and one of the least visible. Most of the cost never shows up as a single line item. It is spread across absences, turnover, incidents, disability claims and legal exposure.

The national bill

The most recent Canadian Substance Use Costs and Harms report found that substance use cost the Canadian economy $49.1 billion in 2020, or $1,291 for every person in the country [1]. The largest share was lost productivity: $22.4 billion, or 45.6% of the total, from premature death and long- and short-term disability [1].

• Alcohol was the costliest substance at $19.7 billion, just over 40% of all costs [1].
• Opioids cost $7.1 billion in 2020, the highest of any year studied. Nearly three-quarters of that was lost productivity, mostly from early deaths [1].

What it looks like at work

A 2025 national study by the Canadian Centre on Substance Use and Addiction surveyed workers and managers in safety-sensitive jobs, including construction, natural resources and health care [2]. It found:

• About 1 in 5 workers and 2 in 5 managers reported using alcohol or other drugs within two hours before work or during work, or being hungover or high at work.
• About half of those who used before or during work said they usually kept working and tried to hide it.
• 19% said their workplace culture generally accepted drinking or cannabis use before or during work.
• Only 56% felt they could safely tell a manager about a substance use problem.
• 43% of managers did not feel trained to intervene with a worker who might be impaired.

Injuries

An Institute for Work & Health study followed 1,715 Canadian workers through three survey waves between 2018 and 2020. Workers who used cannabis before or during work were about twice as likely to be injured on the job as non-users. Workers who used cannabis only outside work hours had no added injury risk [3]. The lesson for employers is that the risk sits with impairment at work, which is exactly what fitness-for-duty testing and monitoring are designed to catch.

Absences and turnover

US data adds the day-to-day cost. A 2017 analysis found that workers with a substance use disorder take nearly 50% more unscheduled leave and have 44% higher turnover than the overall workforce. Workers in recovery did better than average, with about 10% fewer unscheduled leave days and 12% lower turnover [4]. Replacing an employee typically costs around a fifth of their annual salary and benefits [4]. Comparable Canadian figures have not been published, so treat these as indicative.

Legal exposure in Canada

Since 2004, section 217.1 of the Criminal Code has placed a legal duty on everyone who directs how others work to take reasonable steps to prevent bodily harm to workers and the public. Organizations convicted on indictment face fines with no upper limit [5]. Provincial safety law adds to this. In British Columbia, section 4.20 of the Occupational Health and Safety Regulation restricts working while impaired by alcohol or drugs in a way that could endanger anyone, and the BC Human Rights Tribunal has accepted stricter zero-tolerance policies in safety-sensitive settings [6].

At the same time, Canadian employers face limits that US employers do not:

• In Irving Pulp & Paper (2013 SCC 34), the Supreme Court held that random alcohol testing in a dangerous workplace still needs evidence of a general substance problem in that workplace.
• In Stewart v. Elk Valley Coal (2017 SCC 30), the Court upheld the termination of a worker involved in an incident who had not disclosed drug use, under a policy that encouraged disclosure before an incident.
• A substance use disorder is a disability under human rights law, so employers have a duty to accommodate up to undue hardship.

US federal rules, such as Department of Transportation testing, often mandate random testing. Canadian law generally does not permit blanket random testing. Employers need a policy built for Canadian law.

What reduces the cost

• A clear impairment policy that covers alcohol, cannabis, other drugs and medications, and encourages early disclosure.
• Post-incident and reasonable cause testing, collected under chain of custody and reviewed by a Medical Review Officer, so decisions rest on confirmed results.
• Manager training to recognize and respond to impairment.
• A return-to-work path with monitoring. Workers in recovery have better attendance and lower turnover than average [4], and 83% of managers in the CCSA study were comfortable with a co-worker returning to the same job after treatment [2]. Monitoring gives the employer objective evidence that the return is working.

This article is general information, not legal advice. Have your drug and alcohol policy reviewed by employment counsel in your province.

Next: Why medical monitoring works for doctors and nurses returning to practice