Part One: COVID-19 & the Canadian Constitution

Sujit Choudhry*
May 11, 2020

Part 1 of 2

Over the past two months, the federal government, the provinces, and municipalities have exercised a variety of legal powers to respond to the COVID-19 pandemic. Not surprisingly, a large number of constitutional issues have emerged — and could soon give rise to constitutional challenges in the courts.

While physical distancing measures are beginning to generate controversy as infringements of the freedoms of assembly and association, they are only one of a long and growing number of constitutional issues, including those arising from: domestic violencecontact tracing via cellphone data; resource allocation decisions in hospitals for end-of-life care; the federal-provincial division of powers with respect to public health (including interprovincial transport and the Emergencies Act); and the delegation of legislative powers by Parliament and provincial legislatures to the executive.

In this article I provide a brief overview of the constitutional landscape in Canada, through the lens of COVID-19.

  1. Lockdowns, Canadian-style

Canada vs. the United States: Provinces and municipalities across Canada have enacted a range of measures that have closed down non-essential businesses, government offices and schools, and sharply restricted the use of public property. While the media has taken to referring to these measures in aggregate as “lockdowns”, from a constitutional perspective, it is important to look at the details.

To see why, let’s quickly compare the COVID-19 measures in California and Ontario:

  • In California, Governor Gavin Newsom’s Executive Order N-33–20 (19 March 2020)ordered “all individuals living in the State of California to stay home or at least their place of residency” subject to exceptions for individuals employed in “critical infrastructure sectors” (including communications, financial institutions, food and agriculture, health care, and transportation) and essential government services. Individuals can leave the house to access “necessities” such as food, prescriptions, and health care — and when they leave the house, must maintain physical distancing. This is known as a shelter-in-place order.

California’s shelter-in-place order is much more restrictive than the sum total of the various measures in Ontario, because it establishes a default rule that all residents must remain at home, subject to limited exceptions. While the Canadian Civil Liberties Association has criticized the enforcement of physical distancing restrictions on the bases that police officers have not educated or warned individuals before issuing tickets, and that the fines imposed are disproportionately high, we should not lose sight of the relative narrowness of Ontario’s approach, and how that keeps it on relatively safe constitutional terrain.

Self-isolation orders infringe a number of rights under the Canadian Charter of Rights and Freedoms, including the freedoms of religion, expression, assembly and association. However, all Charter infringements can potentially be justified under s. 1, the Charter’s limitation clause. A central issue to any potential Charter challenge would be whether the government has chosen the approach that infringes rights only to the extent necessary. The decision to not issue a blanket shelter-in-place order, and instead rely on self-isolation directed at a class of individuals based on objective criteria for a limited timeframe, would count in favour of the constitutionality of the order.

  • Charterduties to protect health care workers and the vulnerable: The Charter may actually require lockdowns of some fashion. It is arguable that governments have positive Charter duties to take measures to protect those most at risk from infection: the elderly and the immunocompromised (who are at risk involuntarily), and health care workers (who are at risk because of their professional duties). A critical constitutional fact is the risk of transmission by persons who are COVID-19 positive but asymptomatic, and who may therefore be unaware that they are infected and pose a threat to others.

Under European human rights law, it has long been accepted that governments have a duty to protect individuals from threats to their life posed by natural risks or other persons under certain circumstances. European commentators have accordingly argued that shelter-in-place orders are obligatory for this reason. A parallel line of argument could be developed under the Charter. The potentially fatal nature of COVID-19 engages s. 7 of the Charter, which protects the right to life, liberty and security of the person. In R v. Bedford, the Supreme Court held that the government had acted unconstitutionally by criminally prohibiting individuals (sex workers) from engaging in conduct (solicitation). Notably, the prohibition was found to be unconstitutional because it put sex workers at greater risk of physical harm from other persons (pimps), not the state. In Dunmore v. Ontario (Attorney General), the Supreme Court held that the exclusion of an industry (agriculture) from private sector collective bargaining legislation violated employees’ freedom of association, because it breached a positive duty to extend to workers the scope of protective legislation that protected them from management, not the state. Bedford and Dunmore in combination arguably create a positive duty on the state to protect human life, even at the hands of non-state actors, at least where the state knows or ought to know of the risk.

  • Domestic violence: Globally, one of the most horrific consequences of lockdowns has been a steep rise in domestic violence, because abused partners, generally women, are now forced to spend more time at home with, and are more limited in their ability to flee, their abusive partners. The risk of domestic violence is exacerbated by the economic stress created by COVID-19. The situation is so serious that it has led to public statements from the United Nationsand the Government of Canada. The increased risk of domestic violence has constitutional implications under the Charter. Using Bedford (see above) as a model, if lockdowns place abused partners at greater risk of abuse, they breach s. 7. Under a s. 1 analysis, governments could more easily justify lockdowns if they adopted measures such as exemptions for victims of domestic violence, providing and/or increasing funding to emergency shelters, and providing mechanisms for abused partners fleeing domestic violence to alert authorities (e.g. at pharmacies or by texting).
  1. Contact tracing and privacy

As governments chart a path out of lockdowns and to reopen the economy and public institutions, a key element will be contact tracing, which has long been an important weapon in the arsenal of public health authorities. The basic idea behind contact tracing is simple: to identify individuals who have come into contact with those who have tested positive for COVID-19, and to notify them that they may be infected and should therefore go into self-isolation.

Until recently, contact tracing has taken place manually — that is, by interviews and old-fashioned detective work. The widespread use of mobile phones offers new possibilities for contact tracing at the population level using cellphone dataSouth Korea and Singapore have taken this approach, and Germany is considering itAn individual downloads an app, which, through GPS or Bluetooth, tracks information about other cellphone users with whom they have been in close proximity. If the individual tests positive, notifications are sent out accordingly.

However, there is no consensus as to how any such apps should use, disclose, and retain individuals’ COVID-19 information. Nor do we know what governance mechanisms would be implemented for the contact tracing programs that such apps would enable. Practical questions abound about how such app-based programs would actually work. For example, is participation in such programs to be voluntary, mandatory, or a condition for receiving a COVID-19 test? Are data collected about persons going to be individually identifiable to government authorities, or de-identified? What types and volumes of data will be collected and shared with governments: just physical location history and COVID-19 status, or other demographic information such as age and sex? What about other potentially relevant personal information, such as the geolocations of contactless payments? What data will be shared with infected individuals’ contacts? Who else will have access to this information — paramedics, nurses and doctors, law enforcement — and under what circumstances and according to what procedures? What about employers and co-workers? How long will the information be retained? Would the data be held by government or a newly created arms-length agency? What would the forms of legal and political oversight and accountability be?

At present, it is projected that a COVID-19 vaccine is 12 to 18 months away, so contact tracing programs based on mobile app technology may be in place for some time. Privacy groups, such as the Electronic Frontier Foundation, have raised the alarm. Canadian commentators have also urged caution. This has led to proposals, such as Pan-European Privacy-Preserving Proximity Tracing, to create digital platforms that preserve privacy while achieving the health benefits of population-level contact tracing.

If Canadian governments head in this direction, there are a number of complex legal issues they would face. Would the approach be national, led by the federal government, or would some provinces attempt to move ahead on their own, as has been suggested might happen in Alberta? Would governments attempt to invoke existing statutory powers, or would a new legislative basis be required? If the former route is taken, federal and provincial privacy legislation — such as the federal Privacy Act and Personal Information Protection and Electronic Documents Act (PIPEDA), or provincial statutes such as the Ontario Freedom of Information and Protection of Privacy Act — would apply.

If new legislation were required, it would raise a host of novel constitutional questions. The Charter does not expressly protect the right to privacy, unlike many other constitutions. The Supreme Court’s privacy jurisprudence under the Charter has largely developed in the context of the criminal legal process with respect to the right to be secure against unreasonable search or seizure. The Court has not had many opportunities to develop the right to informational privacy in the non-criminal context, unlike courts in other jurisdictions. What can be expected, though, is that differences in app design and governance will have a direct bearing on whether a contact tracing program is constitutional.

  1. End-of-life decisions

In the battle against COVID-19, a major focus has been on the capacity of hospitals to admit and treat patients that require it. Provincial governments are aggressively trying to expand the necessary resources, especially intensive care unit (ICU) beds and ventilators. The goal is to avoid a situation where hospital admissions rise to the point where the demand for these items exceeds the available supply, as appears to be happening in some parts of the United States. In a worst-case scenario, it is possible this could happen in Canada — although based on current trends, it is possible that ICU admissions will prove to be lower than predicted.

Nonetheless, the Ontario government has reportedly made contingency plans to triage access to ICUs, in the form of draft “last resort guidelines”. According to news reports, the draft guidelines — the “Clinical Triage Protocol for Major Surge in COVID Pandemic” (28 March 2020) — would kick in when hospitals are operating at 200 percent capacity. The Protocol has three levels of triage: level 1 would deny life-saving treatment to patients with more than an 80 percent chance of death; level 2 would deny life-saving treatment to patients with more than a 50 percent chance of death; level 3 would deny such treatment to patients with more than a 30 percent of death. It is not clear if the guidelines have been formally approved. The Ontario government has insisted that it will not make triaging decisions on the basis of age. The draft guidelines would apply both to initial access to ventilators, as well as to decisions to withdraw life support.

Rationing access to ventilators on the basis of the likelihood of survival might breach the equality rights provision of the Charter, s. 15. Under federal and provincial human rights codes, disability has been interpreted to include medical conditions in some circumstances, and the same may apply under the Charter. If so, triaging patients on the basis of the likelihood of survival might amount to disability discrimination under the CharterAn open letter to the Ontario government advances a related argument — that “the Triage Protocol identifies particular disabilities, such as cognitive disabilities and advanced neurodegenerative diseases including Parkinson Disease, and Amyotrophic Lateral Sclerosis” — which could give rise to a claim of disability discrimination, depending on how they are used.

The chances of a successful Charter challenge to the Protocol under s. 15 are unclear. In Auton v. British Columbia, the Supreme Court rejected a Charter challenge to the failure to publicly fund a particular treatment for autistic children, on the basis that it was novel. Auton did not address access to a hospital service that is already publicly funded and is not novel. But on the other hand, Auton might also have established a political questions doctrine around the content of publicly funded health care services, sending the message that the courts will avoid being pulled into disputes over the allocation of scare health care resources under the Charter.

Withdrawing life-sustaining treatment raises different issues. Legal disputes may arise in situations where a patient is incapable, the Protocol provides that a patient may be weaned off a ventilator and the medical team wishes to do so, but the patient’s family disagrees. Until now, such disputes have arisen without the involvement of formal government policy and have been governed by the statutory and common law framework for consent to treatment and substitute decision-making. The core legal question has been whether the substitute decision-maker’s refusal to consent to the withdrawal of treatment is in the patient’s best interests, as the Ontario Court of Appeal explained in Rasouli (Litigation Guardian of) v. Sunnybrook Health Sciences Centre. The Protocol would bring such disagreements under the Charter, and would raise a novel issue that falls outside the scope of the Supreme Court’s landmark decision in Carter v. Canada on Medical Assistance in Dying (MAID).

To be continued. Part 2 will be available tomorrow.

[*]Gastwissenschaftler at the WZB Berlin Social Science Research Center, and the Principal of This blog post is published with permission from Medium:

Subscription Form


Protection of Privacy – Personal information provided is collected in accordance with Section 33(c) of the Alberta Freedom of Information and Protection of Privacy Act (the FOIP Act) and will be protected under Part 2 of that Act. It will be used for the purpose of managing CCS’ email subscription lists. Should you require further information about collection, use and disclosure of personal information, or to unsubscribe, please contact: Administrator, Centre for Constitutional Studies, 448D Law Centre, University of Alberta, Edmonton AB, T6G 2H5, Tel: 780-492-5681, Email: You may unsubscribe from our email lists at any time.
Centre for Constitutional Studies
448D Law Centre
University of Alberta
Edmonton, AB T6G 2H5
chevron-down linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram