Ten years after British Columbia declared toxic drugs a public health emergency, Canada is still struggling to turn promising responses into lasting interventions that save lives.
Since 2016, more than 56,000 apparent opioid-toxicity deaths have been recorded nationwide. Yet Canada has not lacked ideas for addressing this crisis. It has developed and advanced treatment, overdose prevention, harm reduction and drug policy innovations that have attracted international attention. The deeper problem is what happens after innovation; whether promising approaches are implemented at sufficient scale, evaluated properly and sustained long enough to matter.
The latest numbers offer some reason for cautious optimism. Canada recorded 5,704 apparent opioid-toxicity deaths in 2025, down 23 per cent from the previous year. But the reasons remain uncertain, including possible changes in the drug supply, wider naloxone access and a shrinking population at risk after years of cumulative loss of life.
Still, 5,704 deaths amount to roughly 15 deaths a day, and the drug supply continues to evolve. In 2025, 82 per cent of opioid toxicity deaths involved non-pharmaceutical opioids, while 70 per cent also involved a stimulant. The crisis is indeed changing, not disappearing.
Canada has adopted, adapted and scaled a wide range of responses, from opioid agonist treatments, take-home naloxone, supervised consumption, drug checking, prescribed alternatives, decriminalization, recovery programming and rapid-access addiction care. Many were first demanded, created or sustained by people who use drugs and community organizations before governments embraced them.
Yet access remains uneven. In practice, what help someone can receive may depend on their postal code, which government is in power and whether a local service survives the next funding cycle.
Recent provincial shifts make this problem tangible. Ontario is expanding its Homelessness and Addiction Recovery Treatment Hubs (HART Hubs) while ending provincial funding for supervised consumption sites, with its last provincially funded site in Kingston closing on Sept. 30. Those Hubs may provide valuable treatment services, but treatment and supervised consumption perform different functions. British Columbia offers a different example: its decriminalization exemption was substantially narrowed in 2024 and expired without renewal this January, making the original policy difficult to evaluate as a stable three-year experiment. Saskatchewan meanwhile has legislated involuntary addiction treatment, while Alberta is building its own “compassionate intervention” system. These approaches are not identical. What they share is that Canada keeps changing the mix of services while struggling to build the continuity and evaluation needed to learn what works, for whom and under what conditions.
That is why the familiar treatment-versus-harm reduction debate misses the point. Asking which one will “solve” the overdose crisis is like asking whether an emergency department or a family doctor matters more. They serve different purposes. Treatment can help people reduce or stop substance use while harm reduction services can prevent disease, injury or death at the moment drugs are consumed while connecting people with further care if needed. Nor are policing and incarceration substitutes for health care; criminalization can deter people from seeking help while incarceration can increase overdose risk, particularly after release. A functioning system needs the right services in the right places, connected to one another.
This system also needs follow-through, including stable funding, common quality standards and timely evaluation. Canada’s divided jurisdictions make that harder. One level of government can fund or authorize a service while another regulates it, and municipalities can determine where the service operates. What looks like a debate over evidence can instead become a tug of war over jurisdiction, ideology and political tolerance. Services may be opened, restricted or closed before governments have adequately measured their effects.
There is also a limit to what even a well-run health system or drug law enforcement can accomplish while the product causing much of the crisis remains unpredictable. Potency, contamination and combinations of opioids and stimulants can change faster than clinical services and police can respond. Canada’s experience with cannabis shows that governments can create legal access, product standards and supply oversight, although it does not tell us how other drugs should be regulated.
The relevant question is not whether one model should simply be copied, but which upstream approaches can reduce exposure to products of unknown composition and potency without creating new harms.
A decade into this emergency, Canada does not primarily need another dashboard documenting the crisis. It needs a system capable of learning from what already exists, measuring what works, adapting when conditions change, sustaining effective services at sufficient scale and evaluating what happens when governments change course.
