Alternatives to Prohibition and an Integration of the Course
Drugs and Society
Learning objectives for this lesson:
- Explain the public health approach to drug policy, using Vancouver’s Four Pillars framework (2001) and Canada’s federal drug strategies from 1987 to 2016 as examples.
- Define prohibition, decriminalization (de jure and de facto), legal regulation and commercialization, and classify real, dated policies with these terms.
- Describe the design, amendments, dates and evaluated outcomes of decriminalization in Portugal (from 2001), Oregon (2021 to 2024) and British Columbia (2023 to 2026).
- Explain why evaluations of the same policy can reach conflicting conclusions, and calculate a simple difference-in-differences estimate from published rates.
- Compare the cannabis regulatory models of Canada, Uruguay, American states and Germany on the main levers of legal regulation, including the purposes and structure of Canada’s Cannabis Act.
- Summarize evidence since legalization on cannabis use, legal sources, potency, poisonings in young children and criminal charges in Canada, and on adolescent use in Uruguay and the United States, and calculate a ratio of monthly rates.
- Weigh prohibition, decriminalization and legal regulation against Canadian evidence, including the record of legal alcohol and tobacco, stating the criteria and weights behind a conclusion.
- Summarize how the lessons of the course fit together, define a syndemic, and link substance use to mental health, violence and trauma, stigma, HIV and other infections, and structural drivers, naming the series course that develops each strand.
This course was developed by Dr. Kiffer G. Card.
Public Health Frameworks for Drug Policy: Four Pillars, National Strategies and Definitions
Content note. This section mentions overdose deaths and HIV and hepatitis C epidemics among people who inject drugs. If you would like support, the 9-8-8 Suicide Crisis Helpline is available by call or text at 9-8-8, the National Overdose Response Service can be reached at 1-888-688-6677, and the Hope for Wellness Help Line for Indigenous people can be reached at 1-855-242-3310.
Introduction
Lessons 2 and 11 traced how Canada and other countries came to control psychoactive substances through criminal law. Lesson 2 followed the Opium Act of 1908 and the racialized panics that shaped early drug laws, and Lesson 11 examined the international conventions, the war on drugs, policing and incarceration, and the iron law of prohibition, the observation that enforcement against a drug market tends to push it toward more potent products. This final lesson asks what the alternatives to prohibition are, what happened when governments tried them, and how a student can weigh the evidence about them.
This first section supplies the vocabulary for the rest of the lesson. It begins with the idea of a public health approach to drug policy, follows that idea through Vancouver’s Four Pillars framework and Canada’s national drug strategies, and ends with precise definitions of decriminalization and legal regulation. Section 2 then applies these definitions to decriminalization in Portugal, Oregon and British Columbia, and Section 3 applies them to the legal regulation of cannabis.
Learning Objectives
- 1.1 Explain what a public health approach to drug policy is and how it differs in aims from a policy built mainly on criminal prohibition.
- 1.2 Describe Vancouver’s Four Pillars framework (endorsed by city council in May 2001), the conditions that produced it, and what followed from it, including Insite and the Supreme Court of Canada’s 2011 decision.
- 1.3 Trace Canada’s federal drug strategies from 1987 to 2016, including the removal of harm reduction as a pillar in the late 2000s and its restoration in December 2016.
- 2.1 Define prohibition, decriminalization (de jure and de facto), legal regulation and commercialization, and explain how diversion differs from each.
- 2.2 Classify real Canadian and international drug policies on a spectrum of control, giving the date of each policy and stating what each approach means for the drug supply.
1.1 From Prohibition to a Public Health Framing
Every drug policy rests on assumptions about what drug use is and what the state should do about it. This part contrasts two sets of assumptions that run through the history covered in this course.
1.1.1 What Prohibition Means
Prohibition is a policy in which the production, sale and possession of a substance are criminal offences, except for narrow medical or scientific purposes. As of 2026, most psychoactive drugs in Canada other than alcohol, tobacco, caffeine and (since October 2018) cannabis are controlled this way under the Controlled Drugs and Substances Act of 1996. A person found in possession of a controlled drug without authorization can be charged, prosecuted and given a criminal record. Lesson 11 explained that this approach rests on the international drug control conventions of 1961, 1971 and 1988, and it described the costs that critics attribute to it: large numbers of arrests for possession, unequal enforcement against Black and Indigenous people, violence in drug markets and a drift toward more potent drugs.
Supporters of prohibition argue that criminal penalties deter people from starting to use drugs, signal social disapproval, keep prices high and give police the tools to disrupt organized crime. These are empirical claims, and the evidence on each is examined in this lesson and in HSCI 402, which develops the methods used to evaluate drug and alcohol policies.
1.1.2 The Public Health Approach
A public health approach treats drug use and its harms mainly as questions of health and social conditions. It asks which policies reduce deaths, disease and social harm across a whole population, and it judges a policy by those outcomes. Under this approach, enforcement remains one tool among several. When Health Canada announced the Canadian Drugs and Substances Strategy in December 2016, it described a “comprehensive, collaborative, compassionate and evidence-based approach to drug policy” (Health Canada, 2016), language that reflects this framing.
The public health approach also includes harm reduction, which aims to reduce the health and social harms of drug use without requiring that people stop using drugs. Needle distribution, naloxone, supervised consumption sites and drug checking are examples, and HSCI 402 develops these services in depth. Harm reduction became a contested element of Canadian drug strategy, as the rest of this section shows.
These two framings did not arrive in Canadian policy as abstractions. They met in a particular neighbourhood in the 1990s, and the next part describes what happened there.
1.2 Vancouver’s Four Pillars
In Vancouver, a public health framework for drugs took shape as a municipal policy. Case 12.1 summarizes the conditions in Vancouver’s Downtown Eastside that led to it.
In the mid-1990s, Vancouver’s Downtown Eastside faced several linked crises at once. A 2005 report to Vancouver City Council by the city’s drug policy coordinator, Donald MacPherson, recalled that public health emergencies had been declared over HIV and hepatitis C among people who inject drugs and over drug overdose deaths, and that the neighbourhood had a large open drug market. The report described the Four Pillars framework as a response to the failure of all levels of government to respond adequately to illicit drug sale and use in the area (MacPherson, 2005).
The city’s draft A Framework for Action: A Four-Pillar Approach to Drug Problems in Vancouver, written by MacPherson, went to council in April 2001, and council unanimously endorsed it on 15 May 2001 (City of Vancouver, 2001; MacPherson, 2005).
Case 12.1 shows the clustering of overdose, infection and an open drug market in one place. Section 4 returns to this pattern, because it is an example of what researchers call a syndemic.
1.2.1 The Four Pillars
The framework organized the city’s response under four headings, which it called pillars: prevention, treatment, harm reduction and enforcement, applied together in a coordinated way (MacPherson, 2005). Click each card below to read what each pillar covers in general terms.
The framework gave harm reduction a place as one of four pillars beside enforcement, and it asked that all four be applied in a comprehensive and integrated way (MacPherson, 2005). Both features distinguished it from a response led mainly by policing.
1.2.2 What Followed
Two projects associated with the framework drew international attention. Insite, which opened in the Downtown Eastside in 2003, is described by Vancouver Coastal Health as North America’s first legal supervised consumption site, and as of October 2026 it operated under a Health Canada exemption from prosecution under federal drug laws (Vancouver Coastal Health, n.d.). A research trial known as the North American Opiate Medication Initiative (NAOMI) also began under the framework (MacPherson, 2005). HSCI 402 develops supervised consumption and prescribed opioid treatments.
Insite’s exemption was made under section 56 of the Controlled Drugs and Substances Act, which allows the federal Minister of Health to exempt people or activities from the Act for medical, scientific or public interest purposes. When the federal government declined to extend the exemption, the case reached the Supreme Court of Canada. In Canada (Attorney General) v PHS Community Services Society, decided on 30 September 2011, the Court held that the Minister’s refusal breached section 7 of the Canadian Charter of Rights and Freedoms and ordered the Minister to grant the exemption (Canada (Attorney General) v PHS Community Services Society, 2011). Section 56 reappears in Section 2, because British Columbia’s decriminalization in 2023 used the same legal mechanism.
Vancouver’s framework was local. The next part turns to the national strategies that set federal priorities over the same decades.
1.3 Canada’s Federal Drug Strategies, 1987 to 2016
Federal drug strategies set out how Ottawa divides its effort and money among enforcement, prevention, treatment and harm reduction. Figure 12.1 places the federal strategies and the Vancouver events from Part 1.2 on one timeline, and Table 12.1 summarizes each strategy.
1.3.1 1987 to 1998: A Health-Led Strategy with Shrinking Funds
On 25 May 1987, the federal government launched a five-year strategy, National Drug Strategy: Action on Drug Abuse, the year after Prime Minister Brian Mulroney had declared drug abuse an epidemic. It committed $210 million, about 77 percent of it for demand reduction (prevention, treatment and research), at a time when federal effort had been almost entirely devoted to reducing supply. The Canadian Centre on Substance Abuse (now the Canadian Centre on Substance Use and Addiction, CCSA) was created in 1988 (House of Commons Special Committee on the Non-Medical Use of Drugs, 2002). The word “abuse” in these titles was standard at the time; this course uses “substance use” and explained in Lesson 1 why the older term is avoided.
Cabinet renewed the strategy in March 1992 as Canada’s Drug Strategy, with planned funding of $270 million over five years, although Health Canada later estimated that only about $104.4 million was spent. When the strategy was renewed again in 1998, its stated principle was that “substance abuse is primarily a health issue,” and harm reduction was listed as one of four pillars alongside enforcement, prevention and treatment. The parliamentary committee that reviewed it in 2002 nonetheless described the late 1990s as the effective sunset of the strategy, because its funding had been cut sharply (House of Commons Special Committee on the Non-Medical Use of Drugs, 2002).
1.3.2 2007 to 2016: Harm Reduction Removed and Restored
The National Anti-Drug Strategy, which the federal government announced on 4 October 2007, organized federal action under three action plans, prevention, treatment and enforcement, and harm reduction was no longer a pillar (Department of Justice Canada, 2012). Health Canada’s 2016 announcement dates the removal of the harm reduction pillar to 2006, the year before the strategy was announced (Health Canada, 2016), so sources give either 2006 or 2007 for the change. On 12 December 2016, the federal Minister of Health announced the Canadian Drugs and Substances Strategy, which replaced the National Anti-Drug Strategy and formally restored harm reduction as a pillar alongside prevention, treatment and enforcement (Health Canada, 2016).
Table 12.1. Canada’s federal drug strategies from 1987 to 2016 and the place of harm reduction in each.
| Year | Strategy | Main features | Harm reduction |
|---|---|---|---|
| 1987 | National Drug Strategy: Action on Drug Abuse | $210 million over five years; about 77 percent for demand reduction; CCSA created in 1988 | Not among the strategy’s six fronts of action |
| 1992 | Canada’s Drug Strategy (Phase II) | Planned $270 million over five years; about $104.4 million spent | Not among the phase’s stated priorities |
| 1998 | Canada’s Drug Strategy (renewal) | “Primarily a health issue”; funding reduced again | One of four pillars |
| 2007 | National Anti-Drug Strategy (announced 4 October 2007) | Prevention, treatment and enforcement | Removed as a pillar (dated to 2006 by Health Canada) |
| 2016 | Canadian Drugs and Substances Strategy | Prevention, treatment, harm reduction and enforcement | Restored as a pillar |
Sources: House of Commons Special Committee on the Non-Medical Use of Drugs (2002); Department of Justice Canada (2012); Health Canada (2016).
Table 12.1 shows that the four pillars of Vancouver’s framework and of the 2016 national strategy were nearly the same, while the weight given to harm reduction changed with governments. In each strategy, though, possession of controlled drugs remained a criminal offence. In 2021, Health Canada’s Expert Task Force on Substance Use went further. Its first report unanimously recommended ending criminal penalties for simple possession, and most members recommended bringing the Controlled Drugs and Substances Act, the Tobacco and Vaping Products Act and the Cannabis Act under a single public health legal framework (Expert Task Force on Substance Use, 2021). These recommendations use the terms decriminalization and regulation, which the next part defines.
2.1 Defining the Alternatives
Public debate often uses “decriminalization” and “legalization” loosely, and the two are sometimes treated as the same thing. They describe different policies, with different effects on people who use drugs and on the drug supply. This part defines the main terms used in the rest of the lesson.
2.1.1 Decriminalization
The Expert Task Force on Substance Use (2021) defined decriminalization as the removal of criminal penalties for an activity. For drugs, the activity is usually possession of a small amount for personal use. Decriminalization leaves the substance controlled: producing and selling it remain crimes, so people who use drugs still buy from the unregulated market described in Lesson 5. Decriminalization can be achieved in two ways. In de jure decriminalization, a law or a legal instrument changes, as when Portugal amended its law in 2001 or when Health Canada granted British Columbia an exemption in 2023. In de facto decriminalization, the law stays the same but police or prosecutors stop enforcing it for possession (Expert Task Force on Substance Use, 2021).
Decriminalization schemes differ in two design features. The first is whether possession carries an administrative sanction, a non-criminal penalty such as a fine or a referral: Portugal kept administrative sanctions, Oregon used a $100 fine that could be waived, and British Columbia had no sanction below its limit. The second is the threshold quantity below which possession is presumed to be for personal use. Thresholds set too low leave many people exposed to trafficking charges, and the Expert Task Force (2021) recommended thresholds high enough to reflect people’s actual purchasing and consumption habits.
Decriminalization also raises a question of international law, because Article 3(2) of the 1988 Convention, quoted in Lesson 11, requires each party to make possession for personal consumption a criminal offence “subject to its constitutional principles and the basic concepts of its legal system.” The International Narcotics Control Board (INCB), which monitors compliance with the treaties, has treated Portugal’s approach, in which possession of small quantities for personal use remains prohibited and is sanctioned mainly by administrative measures, as consistent with the treaties: an INCB mission in 2004 described the practice as “consistent with the international drug control treaties,” and INCB President Werner Sipp said in Vienna on 9 December 2015 that Portugal’s dissuasion procedure, run by the commissions described in Section 2, was “in full compliance with the three drug control conventions” (Sipp, 2015). On the Board’s reading, the qualifier therefore stretches to decriminalization with administrative sanctions and stops at a legal market for non-medical use, which the Board said could not be reconciled with Canada’s international obligations when Canada legalized cannabis (Lesson 11, Case 11.2).
2.1.2 Legal Regulation, Legalization and Commercialization
Legal regulation goes a step further. It creates a lawful supply of a substance under government oversight of its production, manufacture, importation, distribution, promotion and sale, which is how the Expert Task Force on Substance Use (2021) defined regulation. The word legalization is often used for the same change, but it can suggest an absence of rules, which does not describe any of the systems in this lesson. Alcohol, tobacco and, since 17 October 2018, cannabis are legally regulated in Canada.
Legal regulation itself varies along a scale of commercialization, the degree to which private companies may produce, promote and sell the substance for profit. As of 2026, Canada’s cannabis market and its alcohol market allow licensed for-profit producers. Uruguay’s cannabis law of 2013, which gave the state control of large-scale production, and Germany’s law of 2024, which allowed home growing and non-profit clubs, set up less commercial systems. Section 3 compares these models.
2.1.3 Where Diversion Fits
Lesson 11 introduced drug treatment courts and police diversion, which steer some people charged with or suspected of drug offences toward treatment or services. Diversion leaves the criminal offence in place; police or courts use discretion about how to respond to it. Diversion therefore sits within prohibition, and Oregon’s 2024 law, discussed in Section 2, combined a restored criminal offence with optional diversion programs.
2.2 Classifying Policies on a Spectrum of Control
Figure 12.2 places these approaches on a spectrum of control, from criminal prohibition at one end to commercial legal regulation at the other. The spectrum is schematic: it shows how the approaches differ in their legal design and makes no claim about which produces the least harm.
Table 12.2 restates the definitions in Figure 12.2 and adds what happens to a person found with a small amount, and to the supply, under each approach.
Table 12.2. Four approaches to drug control, what each means for a person found with a small amount, and what each means for the supply.
| Approach | Person found with a small amount | Supply | Dated example |
|---|---|---|---|
| Prohibition | May be charged and receive a criminal record; diversion possible | Illegal and unregulated | Opioids and stimulants in Canada (2026) |
| Decriminalization | No criminal charge below the threshold; an administrative sanction may apply | Illegal and unregulated | Portugal since 1 July 2001 |
| Legal regulation, non-commercial | Lawful within limits | Lawful, from the state, home growing or non-profit clubs | Germany, law in force from 1 April 2024 |
| Legal regulation, commercial | Lawful within limits | Lawful, from licensed businesses, with product rules | Cannabis in Canada since 17 October 2018 |
Why the supply column matters
The supply column of Table 12.2 explains a point that recurs in Section 2. Under both prohibition and decriminalization, the drugs people use come from the unregulated market, so their content is unknown. Lesson 5 described how that market came to be dominated by fentanyl and its analogues, with sedatives such as xylazine, medetomidine and bromazolam added. A change in the penalty for possession leaves this supply unchanged. Only legal regulation, or a prescribed supply of the kind HSCI 402 examines, changes what is in the drug itself. Gaudreault and colleagues (2025), whose study Section 2 describes, concluded that their findings did not support decriminalization alone as an effective way to reduce overdose harms, and evaluations of decriminalization therefore also examine outcomes such as arrests and contact with services.
Interactive 12.1 asks you to apply these definitions. Each item is a real policy with its dates. Choose the approach that best describes it, then select Check my answers to read the reasoning for each item.
For each policy, choose one of the four approaches from Figure 12.2.
Several items in Interactive 12.1 changed category over time, which is why every policy in this lesson carries a date. Try it 12.1 asks you to use the definitions in writing.
Read the two recommendations of the Expert Task Force on Substance Use (2021) summarized at the end of Part 1.3. For each, write two or three sentences that (1) name the approach from Table 12.2 that it describes, (2) state what would change for a person found with a small amount of a drug, and (3) state what would change, if anything, for the drug supply. Then compare your answer with the box on why the supply column matters and explain why the two recommendations would be expected to affect overdose deaths differently.
Summary
A public health approach judges drug policy by its effects on health and social harm across a population. Vancouver’s Four Pillars framework (May 2001) applied it to the overlapping crises of the Downtown Eastside and gave harm reduction a place beside prevention, treatment and enforcement, leading to Insite (2003) and the Supreme Court of Canada’s 2011 PHS decision. Federal strategies from 1987 to 2016 called drug use mainly a health issue while keeping possession a crime, and harm reduction moved from pillar (1998) to omission (late 2000s) to restoration (2016). Decriminalization removes criminal penalties for possession and leaves the supply illegal; legal regulation creates a lawful, rule-bound supply; and diversion operates within prohibition. Section 2 applies these definitions to Portugal, Oregon and British Columbia.
Key Takeaways
- 1.1 Prohibition makes the production, sale and possession of a substance criminal offences, and its supporters cite deterrence, social disapproval and the disruption of organized crime. A public health approach treats drug use and its harms mainly as questions of health and social conditions, judges a policy by its effects on deaths, disease and social harm across a population, and includes harm reduction among its tools.
- 1.2 Vancouver’s Four Pillars framework, endorsed unanimously by city council on 15 May 2001, responded to linked crises of overdose, HIV, hepatitis C and an open drug market in the Downtown Eastside. It placed harm reduction beside prevention, treatment and enforcement, and it was followed by Insite (2003), which, as of October 2026, operated under a section 56 exemption of the kind that the Supreme Court of Canada ordered the Minister to grant in its 2011 PHS decision.
- 1.3 Federal strategies from 1987 to 2016 shifted money toward demand reduction and described drug use as primarily a health issue, while possession remained a crime throughout. Harm reduction was a pillar in 1998, was dropped under the National Anti-Drug Strategy (dated to 2006 or 2007), and was restored by the Canadian Drugs and Substances Strategy on 12 December 2016.
- 2.1 Decriminalization removes criminal penalties for possession, de jure through a change in law or de facto through a change in enforcement, and leaves production and sale illegal. Legal regulation creates a lawful supply under government rules and varies in how commercial it is, while diversion steers people toward services but keeps the criminal offence in place.
- 2.2 The four approaches can be placed on a spectrum from criminal prohibition to commercial legal regulation, and every policy needs a date because several have changed category over time. Under both prohibition and decriminalization the supply stays illegal and unregulated, so only legal regulation or a prescribed supply changes what is in the drug itself.
Reflection
Canada’s federal drug strategy listed harm reduction as one of four pillars in 1998; the National Anti-Drug Strategy of the late 2000s dropped it; and the Canadian Drugs and Substances Strategy of 12 December 2016 restored it. Under all three strategies, possession of controlled drugs remained a criminal offence, and the 1998 strategy described substance use as “primarily a health issue.” In 2021, Health Canada’s Expert Task Force on Substance Use recommended ending criminal penalties for simple possession. Using the definitions of prohibition, decriminalization and legal regulation from this section, explain how a strategy can call drug use a health issue while keeping possession a crime. Then explain what would change, and what would stay the same, for a person who buys fentanyl from the unregulated supply if the Task Force’s recommendation were adopted.
A strategy can call drug use a health issue while keeping possession a crime because the two operate at different levels. The strategies of 1998 and 2016 directed money and attention toward prevention, treatment and harm reduction, but they left the legal status of drugs under the Controlled Drugs and Substances Act unchanged. In the terms of this section, Canada remained under prohibition, with public health services added around it; when harm reduction was dropped in the late 2000s and restored in 2016, the law on possession did not move.
If the Task Force’s recommendation were adopted, this would be decriminalization. The person buying fentanyl could no longer be charged or given a criminal record for possessing an amount below the threshold, which could reduce fear of police, make it easier to seek care and remove a barrier to jobs and housing. What would stay the same is the supply. Production and sale would remain crimes, so the person would still buy from the unregulated market, where Lesson 5 showed that the content and potency of drugs are unknown and sedatives such as medetomidine may be added. Only legal regulation, or a prescribed supply, would change what is in the drug. For this reason I would expect decriminalization to affect arrests and contact with services more clearly than it affects deaths.
Minimum 20 characters required.
Question 1: Under the decriminalization of drug possession, which statement is accurate?
Question 2: What change did the Canadian Drugs and Substances Strategy of 12 December 2016 make to federal drug policy?
Question 3: Vancouver’s Four Pillars framework, endorsed by city council on 15 May 2001, organized the city’s response under which four pillars?
Question 4: A city’s police stop charging people for possessing small amounts of drugs, but the law itself is unchanged. Which term describes this?
Decriminalization in Portugal, Oregon and British Columbia, and Policy Evaluation in Brief
Content note. This section discusses overdose and drug toxicity deaths and studies that count them. If you would like support, the 9-8-8 Suicide Crisis Helpline is available by call or text at 9-8-8, the National Overdose Response Service can be reached at 1-888-688-6677, and the Hope for Wellness Help Line for Indigenous people can be reached at 1-855-242-3310.
Introduction
This section examines three jurisdictions that decriminalized possession of all or most drugs: Portugal, which did so in 2001 and extended its approach in 2023; Oregon, which did so in 2021 and reversed course in 2024; and British Columbia, whose federal exemption ran from 2023 until 31 January 2026. For each, the section describes what the policy changed and what the evaluations found.
The three cases share a difficulty that the final two parts address: researchers studying the same policy have reached conflicting conclusions. Learning to read these disagreements, by asking what each study compared and measured, is a skill that applies to many other policy questions.
Learning Objectives
- 3.1 Describe Portugal’s decriminalization of 1 July 2001, summarize the findings of Hughes and Stevens (2010), and explain how the same Portuguese data were used to call the reform both a resounding success and a disastrous failure.
- 3.2 Describe Oregon’s Measure 110 (1 February 2021 to 31 August 2024), its replacement, and the conflicting evaluations of its effect on overdose deaths.
- 3.3 Describe British Columbia’s exemption (31 January 2023 to 31 January 2026), its amendments, its end and the conflicting evaluations of its effects.
- 4.1 Explain four evaluation designs and calculate a difference-in-differences estimate from published rates.
- 4.2 Identify why evaluations of the same policy disagree, including differences in comparison groups, handling of fentanyl, follow-up, baselines, outcomes and the years compared.
3.1 Portugal, 2001
Portugal’s reform, in place since 2001, is cited by people on every side of the debate. This part sets out what the law did and what an independent evaluation by Hughes and Stevens (2010) found.
3.1.1 What the Law Changed
On 1 July 2001, Law 30/2000 made the acquisition and possession of any illicit drug for personal use an administrative offence. Before the reform, possession for personal use could be punished by up to one year in prison. The 2001 law applied to quantities up to ten days’ supply, set at 0.1 gram of heroin, ecstasy or amphetamines, 0.2 gram of cocaine and 2.5 grams of cannabis. People found with these amounts were referred to regional Commissions for the Dissuasion of Drug Addiction, panels of three members drawn from law, social work and medicine. For people judged to be dependent, a commission could recommend a treatment or education program in place of a sanction; for others, it could suspend proceedings or impose a fine, community service or other administrative penalties. Possession for sale remained a criminal matter for the courts, and under the 2001 law so did amounts above the threshold (Hughes & Stevens, 2010); a 2023 law changed how larger amounts are treated, as described below.
The legal change came with a national drug strategy, adopted in May 1999, that expanded prevention, harm reduction, treatment and social reintegration; the number of people in drug treatment rose from 23,654 in 1998 to 38,532 in 2008 (Hughes & Stevens, 2010). Any evaluation of the reform therefore measures decriminalization together with this expansion of services.
3.1.2 What Hughes and Stevens Found
Caitlin Hughes and Alex Stevens (2010) drew on independent evaluations, national data and interviews with 13 key informants, and compared Portugal’s trends with those in neighbouring Spain and Italy. Table 12.3 summarizes the indicators they reported.
Table 12.3. Selected indicators before and after Portugal’s decriminalization of 1 July 2001, as reported by Hughes and Stevens (2010).
| Indicator | Before | After |
|---|---|---|
| Lifetime use of any illicit drug, ages 15 to 64 | 7.6% (2001) | 12.0% (2007) |
| Past-year use of any illicit drug, ages 15 to 64 | 3.4% (2001) | 3.7% (2007) |
| Estimated problematic drug use, per 1,000 aged 15 to 64 | 7.6 (2000) | 6.8 (2005; not statistically significant) |
| Estimated injecting drug use, per 1,000 | 3.5 (2000) | 2.0 (2005) |
| New HIV cases among people who use drugs | 907 (2000) | 267 (2008) |
| New AIDS cases among people who use drugs | 506 (2000) | 108 (2008) |
| Share of prisoners with drug-related offences | 44% (1999) | 21% (2008) |
Hughes and Stevens concluded that, contrary to predictions, decriminalization did not lead to major increases in drug use, and that the evidence indicated reductions in problematic use, drug-related harms and prison overcrowding. They were explicit about the limits of this evidence: the legal change arrived together with expanded services, up to 38 percent of the commissions were not operating at times between 2005 and 2008, and it was “impossible to state that any of these changes were the direct result of the decriminalization policy” (Hughes & Stevens, 2010, p. 1017).
A resounding success or a disastrous failure?
By the late 2000s, two opposing accounts of Portugal were circulating. A 2009 report for the Cato Institute by Glenn Greenwald called the reform a resounding success, citing a fall in drug-related deaths from close to 400 in 1999 to 290 in 2006. A 2010 report by Manuel Pinto Coelho, chair of the Association for a Drug Free Portugal, called it a disastrous failure, citing a 45 percent rise, from 216 to 314, in deaths with a positive toxicology test for drugs between 2006 and 2007. Hughes and Stevens (2012) compared both accounts with the official data in a peer-reviewed paper. The accounts used different years and different counts: 2006 appears as 290 deaths in one and 216 in the other. The indicator both relied on counted any death with a drug detected at autopsy, so it rose when more toxicology tests were done; deaths that physicians attributed to drugs fell after 2001 and, despite a slight rise from 2005, remained much lower than at the time of the reform. Hughes and Stevens concluded that both accounts used evidence selectively, through their choice of indicators, years and datasets. When you read a claim about any policy in this lesson, ask which years, which measure and which comparison it uses.
3.1.3 Portugal since 2001
Portugal extended its approach in 2023. Law 55/2023 of 8 September 2023, in force from 1 October 2023 (Godinho, 2024), allows possession above the ten-day threshold to be treated as personal use when it can be shown to be so, and it brought synthetic drugs within the decriminalized framework; a legal commentary noted that the law sets no maximum quantity (PLMJ, 2023). Oregon, the next case, shows a decriminalization that lasted three and a half years.
3.2 Oregon’s Measure 110, 2021 to 2024
Oregon decriminalized possession of small amounts of all drugs in 2021. Its experience is closely studied because the policy began just as fentanyl was spreading into the drug supply of the western United States.
3.2.1 What Measure 110 Changed
Oregon voters approved Measure 110 on 3 November 2020, with 58.46 percent voting yes, and it took effect on 1 February 2021 (Ballotpedia, n.d.). Possession of small amounts became a Class E violation with a maximum fine of $100, dismissed if the person completed a health screening through a statewide telephone hotline, or with another approved screener, within 45 days. Cannabis tax revenue above $11.25 million a quarter went to a Drug Treatment and Recovery Fund, which paid for Behavioral Health Resource Networks offering screening, peer support, harm reduction, housing and low-barrier treatment in every county (Oregon Legislative Policy and Research Office, 2021). Measure 110 was thus a decriminalization with an administrative sanction, paired with new funding for services.
3.2.2 How It Worked in Practice
Court data show how the citation pathway was used. For offences from 1 February 2021 to 31 August 2024, Oregon courts recorded 10,028 cases with Class E possession violations. All Class E charges were dismissed in 704 cases (7 percent), with a screening verification filed in 85 of these, and 7,925 cases ended in a conviction after the person failed to appear (Oregon Judicial Department, 2024). These data do not capture treatment funded outside the citation pathway, and Joshi and colleagues (2023) noted that the new treatment funding was distributed slowly, mostly after the first year.
3.2.3 Recriminalization in 2024
In 2024 the legislature passed House Bill 4002, which Governor Tina Kotek signed on 1 April 2024 (Ballotpedia, n.d.). From 1 September 2024, possession of small amounts became a misdemeanour again. Counties and Tribal governments could choose to run deflection programs, which avoid a charge for people who complete them, and 23 counties covering about 85 percent of the population had opted in when the bill passed (Oregon Health Authority, 2024). In the terms of Section 1, Oregon moved from decriminalization back to prohibition with diversion.
3.2.4 What the Evaluations Found
Oregon’s measure took effect just as fentanyl was spreading into the state’s drug supply, so evaluations of its effect on overdose deaths must separate the two changes, and the studies described here did so differently. Noah Spencer (2023) used a synthetic control method, which builds a weighted combination of other states to estimate what would have happened in Oregon without the policy, and estimated 182 additional unintentional overdose deaths in 2021, about 23 percent more than predicted. Joshi and colleagues (2023), using the same general method for the first year, found no statistically significant association. Zoorob and colleagues (2024) noted that fentanyl spread rapidly in Oregon’s supply in the first half of 2021, as the measure took effect, and found a positive association before adjusting for fentanyl’s spread and none after. A National Bureau of Economic Research working paper issued in July 2026, which had not been peer reviewed as of October 2026 (Hall, Hansen & Matsuzawa, 2026), estimated about 1,186 excess deaths in Oregon from 2021 through 2023. Its authors argued that a control for fentanyl’s share of drug seizures may partly capture the fall in seizures of other drugs that followed reduced enforcement, and that other measures of fentanyl’s spread still left a positive effect on deaths. As of October 2026, the evidence on Measure 110 and deaths is conflicting, largely over how the arrival of fentanyl is handled, and research on the measure is still evolving.
British Columbia’s exemption began two years after Measure 110, in a drug supply that was already dominated by fentanyl, which changes what its evaluations can and cannot show.
3.3 British Columbia’s Exemption, 2023 to 2026
British Columbia’s decriminalization used the federal power described in Section 1: an exemption under section 56 of the Controlled Drugs and Substances Act, which allows the Minister to exempt a class of persons from the Act if the Minister considers it necessary for a medical or scientific purpose or otherwise in the public interest (Controlled Drugs and Substances Act, 1996, s. 56(1)).
3.3.1 What the Exemption Did
From 31 January 2023, adults aged 18 and older in British Columbia could possess up to a cumulative 2.5 grams of opioids, cocaine (including crack), methamphetamine and MDMA without criminal penalty or fine; larger amounts could still lead to arrest, charges and seizure, and police could offer information and voluntary referrals (Toward the Heart, 2026; Health Canada, 2024b). From 18 September 2023, the exemption no longer applied within 15 metres of playgrounds, spray pools, wading pools and skate parks. At the province’s request, from 7 May 2024 it applied only in private residences, designated health care clinics, places where people were lawfully sheltering, and overdose prevention and drug checking sites (Health Canada, 2024a). On 14 January 2026, Health Minister Josie Osborne announced that the province would not seek renewal, saying the pilot “has not delivered the results we hoped for” (BC Ministry of Health, 2026), and the exemption ended on 31 January 2026 (BC Centre for Disease Control, n.d.).
The province reported police data to Health Canada. Figure 12.3 shows the monthly average number of police-recorded offences in which possession was the most serious offence, in three periods.
Possession offences fell by about two thirds under the original exemption and rose to about 83 percent of their earlier level after public spaces were excluded; 72.8 percent of offences in the later period occurred in public spaces. The province cautioned that these are police-recorded suspected offences, that some trafficking investigations are first coded as possession, and that attributing changes in services or deaths to decriminalization requires further evaluation (Government of British Columbia, 2025).
3.3.2 What the Evaluations Found
Two studies published in 2025 reached different conclusions about health effects, and a third, published in 2026, found no detectable change in health outcomes. Gaudreault and colleagues (2025) compared British Columbia with the rest of Canada for 2021 to 2022 and for the exemption’s first year, 2023. Police-reported possession incidents fell immediately by about 57 percent, with no significant change in trafficking incidents or, relative to the rest of Canada, in opioid or stimulant hospitalizations or deaths. The authors concluded that decriminalization can reduce harms from criminal legal involvement, and that their findings did not support decriminalization alone as an effective way to reduce overdose harms.
Nguyen and colleagues (2025) built a synthetic British Columbia from six other provinces, using quarterly data from 2016 to 2023 and a baseline set before the province began prescribing a safer supply of opioids in 2020. Compared with that baseline, opioid poisoning hospitalizations were 33 percent higher during the safer supply period and 58 percent higher once decriminalization was added, with no statistically significant change in deaths. The authors suggested that the rise could reflect greater willingness to seek care or diversion of prescribed opioids. A BC Centre for Disease Control analysis found that the drug toxicity death rate in the first 12 months of decriminalization was not statistically different from the earlier rate (Government of British Columbia, 2025). HSCI 402 develops safer supply.
A 2026 study by Gaudreault and colleagues used a controlled interrupted time series (Part 4.1) of monthly rates from August 2020 to March 2025, a period that covers decriminalization in February 2023 and the partial recriminalization of May 2024, when public spaces were excluded. British Columbia was compared with the rest of Canada for possession incidents, with the rest of Canada excluding Quebec for hospitalizations, and with Alberta and Ontario for deaths. The results are ratios: a ratio below 1 means a fall and a ratio above 1 a rise, a ratio of rate ratios compares the change in British Columbia with the change in the comparison regions, and a 95 percent confidence interval that includes 1 means no statistically significant difference. Possession incidents in British Columbia fell after decriminalization (incidence rate ratio 0.61, 95 percent confidence interval 0.42 to 0.86) and rose after recriminalization (1.68, 1.13 to 2.51). Opioid hospitalizations did not differ significantly from the comparison regions after decriminalization (ratio of rate ratios 0.87, 0.54 to 1.42) or after recriminalization (1.03, 0.56 to 1.90), and neither did stimulant hospitalizations (0.77, 0.50 to 1.28, and 1.02, 0.59 to 1.76). Deaths after decriminalization did not differ significantly from those in Alberta (0.83, 0.64 to 1.06) or Ontario (1.07, 0.64 to 1.78). The authors concluded that decriminalization reduced possession incidents but was “insufficient in isolation to affect health outcomes” (Gaudreault et al., 2026).
The three British Columbia studies agree that the exemption sharply reduced police contact for possession while it applied in full, and the 2026 study found that possession incidents rose again after recriminalization. They disagree about hospitalizations: Nguyen and colleagues (2025) found a rise from a baseline set before safer supply, while both studies by Gaudreault and colleagues found no significant difference from other provinces. None of the three found a statistically significant change in deaths. A 2026 systematic review of 27 studies of the pilot concluded that its early outcomes may reflect design features, institutional readiness and system capacity rather than legal change alone, and that its longer-term impacts remained uncertain (Ali et al., 2026). The next part explains the design choices behind such disagreements.
4.1 Policy Evaluation in Brief
A policy evaluation tries to estimate the counterfactual: what would have happened in the same place, over the same period, without the policy. The counterfactual can never be observed, so every evaluation substitutes an estimate of it. HSCI 402 develops evaluation methods in depth; this part introduces the four designs used in the studies above.
4.1.1 Four Evaluation Designs
The designs differ in how they estimate the counterfactual. Open each item below for a plain description and the main weakness of each.
Compares an outcome in the period before a policy with the period after it, in one place. Its counterfactual is that nothing would have changed. It is weak when other things change at the same time, such as the arrival of fentanyl or a pandemic.
Uses many measurements before and after the policy to model the existing trend, then asks whether the level or slope changed when the policy began. Its counterfactual is that the earlier trend would have continued. A controlled interrupted time series adds a comparison series, as Gaudreault and colleagues did for police data in 2025 and for police data, hospitalizations and deaths in 2026 (Gaudreault et al., 2025, 2026).
Compares the change over time in the place with the policy with the change in a comparison group without it. Its counterfactual is that the treated place would have followed the same trend as the comparison group, an assumption called parallel trends.
Builds a weighted combination of several untreated places whose past outcomes closely match the treated place, and uses that synthetic comparison as the counterfactual. It was used by Spencer (2023), Joshi and colleagues (2023), Zoorob and colleagues (2024) and, in a synthetic difference-in-differences form, by Nguyen and colleagues (2025).
4.1.2 Calculating a Difference-in-Differences
The simplest version of difference-in-differences subtracts the change in the comparison group from the change in the place with the policy, as the formula below shows.
Worked Example 12.1: British Columbia’s opioid deaths
Gaudreault and colleagues (2025) reported average quarterly opioid toxicity death rates per 100,000 people. In British Columbia the rate was 11.21 before decriminalization (2021 to 2022) and 11.80 during its first year (2023). In the rest of Canada it was 4.10 and 4.29.
Change in British Columbia = 11.80 − 11.21 = 0.59. Change in the rest of Canada = 4.29 − 4.10 = 0.19. Difference-in-differences = 0.59 − 0.19 = 0.40 per 100,000 per quarter.
On the relative scale, British Columbia’s ratio is 11.80 ÷ 11.21 = 1.053 and the rest of Canada’s is 4.29 ÷ 4.10 = 1.046, so the ratio of ratios is 1.053 ÷ 1.046 = 1.006 (calculated from the unrounded ratios), a difference of less than 1 percent. The published estimate, from a statistical model fitted to quarterly data, was 1.01 with a 95 percent confidence interval of 0.88 to 1.16, which is consistent with no change. The scales differ because British Columbia’s starting rate was almost three times higher, so always check which scale a study uses.
Use Calculator 12.1 to repeat Worked Example 12.1 and to see how strongly the answer depends on the comparison group. Two of its presets use hypothetical numbers that are not drawn from any real place.
4.2 Why Evaluations of the Same Policy Disagree
Table 12.4 sets the evaluations described in Parts 3.2 and 3.3 side by side, so that their designs, comparisons, outcomes and periods can be compared directly.
Table 12.4. Evaluations of decriminalization in Oregon and British Columbia described in this section, with their designs, outcomes and findings.
| Study | Design and comparison | Outcome and period | Main finding |
|---|---|---|---|
| Spencer (2023), Oregon | Synthetic control from other states | Unintentional overdose deaths, 2021 | 182 additional deaths, about 23% above prediction |
| Joshi et al. (2023), Oregon | Synthetic control from 48 states and Washington, DC | Fatal overdose rate, first year after the policy | No statistically significant association |
| Zoorob et al. (2024), Oregon | Synthetic control from 48 states and Washington, DC, adjusted for fentanyl’s spread | Fatal overdose rate, data from 2008 to 2022 | Positive association before adjustment; none after |
| Hall, Hansen and Matsuzawa (2026), Oregon; working paper | Synthetic control with alternative fentanyl measures | Overdose deaths, 2021 to 2023 | About 1,186 excess deaths; not peer reviewed as of October 2026 |
| Gaudreault et al. (2025), British Columbia | Difference-in-differences with the rest of Canada | Hospitalizations, deaths and police data, 2021 to 2023 | Possession incidents down about 57%; no significant change in hospitalizations or deaths |
| Nguyen et al. (2025), British Columbia | Synthetic difference-in-differences from six provinces; baseline before safer supply | Opioid hospitalizations and deaths, 2016 to 2023 | Hospitalizations up 58% from baseline; no significant change in deaths |
| Gaudreault et al. (2026), British Columbia | Controlled interrupted time series with the rest of Canada (possession), the rest of Canada excluding Quebec (hospitalizations), and Alberta and Ontario (deaths) | Possession incidents, opioid and stimulant hospitalizations and deaths, monthly, August 2020 to March 2025, including recriminalization from May 2024 | Possession incidents down after decriminalization (ratio 0.61) and up after recriminalization (1.68); no significant difference in hospitalizations or deaths |
Table 12.4 shows five recurring reasons for disagreement: different comparison groups, different handling of fentanyl’s arrival, different lengths of follow-up, baselines that include other policy changes such as safer supply, and different outcomes, such as hospitalizations or deaths. The box on Portugal’s opposing accounts in Part 3.1 adds a sixth, the choice of years compared. Interactive 12.2 asks you to identify the main reason in four real disagreements.
For each pair, choose the design difference that best explains why the conclusions differ, then select Check my choices.
Each design answers a slightly different question, so the conclusions are best read together, with attention to what each could rule out. Try it 12.2 applies this reading to a pair of public claims.
A commentator says that “British Columbia’s decriminalization increased overdoses,” and another replies that “decriminalization had no effect on overdoses.” Using Table 12.4 and Part 3.3, write a short paragraph that (1) names the study each speaker is probably relying on, (2) states the outcome and comparison each study used, (3) explains how both statements can be partly supported by evidence, and (4) proposes one wording that a careful writer could use in place of both statements.
Summary
Portugal decriminalized possession of all drugs on 1 July 2001 and expanded treatment; Hughes and Stevens (2010) found no major increase in use and falls in problematic use, HIV infections and drug-related imprisonment, without being able to attribute them to decriminalization alone. Oregon replaced criminal penalties with a waivable $100 fine from February 2021 until September 2024, and evaluations of its effect on deaths conflict. British Columbia’s exemption (31 January 2023 to 31 January 2026) sharply reduced police-recorded possession offences while it applied in full, and evaluations disagree about hospitalizations and find no significant change in deaths. Evaluations disagree when they use different comparisons, periods, adjustments and outcomes. Section 3 turns to legal regulation, using cannabis as the main example.
Key Takeaways
- 3.1 Portugal made possession of up to ten days’ supply of any drug an administrative offence on 1 July 2001 and expanded treatment at the same time. Hughes and Stevens (2010) found no major increase in use and falls in problematic use, HIV infections and drug-related imprisonment, without being able to attribute them to decriminalization alone. Opposing accounts that called the reform a resounding success or a disastrous failure reached their conclusions by choosing different years, indicators and death counts.
- 3.2 Oregon’s Measure 110 replaced criminal penalties with a $100 fine that could be waived after a health screening, from 1 February 2021 until House Bill 4002 restored a misdemeanour with optional deflection on 1 September 2024. Evaluations of its effect on overdose deaths conflict, largely over how the arrival of fentanyl is handled.
- 3.3 British Columbia’s section 56 exemption (31 January 2023 to 31 January 2026) allowed adults to possess up to 2.5 grams without penalty, was narrowed in 2023 and 2024, and was not renewed. Police-recorded possession offences fell sharply while it applied in full; studies disagree about hospitalizations and none found a statistically significant change in deaths.
- 4.1 Every evaluation estimates a counterfactual, using a before-and-after comparison, an interrupted time series, difference-in-differences or a synthetic control. Difference-in-differences subtracts the change in a comparison group from the change in the place with the policy, and its absolute and relative versions can give different impressions when starting rates differ.
- 4.2 Evaluations of the same policy disagree when they use different comparison groups, handle fentanyl’s arrival differently, follow people for different periods, set baselines that include other changes such as safer supply, or measure different outcomes. The choice of years compared, as in the Portuguese accounts, is a further source of disagreement.
Reflection
Two published evaluations of British Columbia’s decriminalization exemption (31 January 2023 to 31 January 2026) reached different conclusions. Gaudreault and colleagues (2025) compared British Columbia with the rest of Canada, using 2021 to 2022 as the period before and 2023 as the period during the exemption; they found a 57 percent immediate fall in police-reported possession incidents and no significant change in opioid or stimulant hospitalizations or deaths. Nguyen and colleagues (2025) compared British Columbia with a weighted synthetic comparison built from six other provinces (a synthetic difference-in-differences design), using a baseline set before British Columbia began prescribing a safer supply of opioids in 2020; they found opioid hospitalizations 58 percent higher by 2023 than at baseline, with no significant change in deaths. Explain two design differences that could account for the different findings on hospitalizations, and describe what a careful reader can and cannot conclude from the two studies together.
The first design difference is the baseline period. Nguyen and colleagues set their baseline before safer supply began in 2020, so their estimate for 2023 combines the effects of safer supply and decriminalization, along with anything else that changed after early 2020. Gaudreault and colleagues compared 2021 to 2022 with 2023, when safer supply was already in place, so their estimate isolates the change around the start of decriminalization. A study that starts earlier can find a larger change simply because more time and more policy changes are included.
The second difference is the comparison group. Gaudreault and colleagues compared British Columbia with the rest of Canada as a whole, while Nguyen and colleagues built a weighted synthetic British Columbia from six provinces chosen to match its earlier trends. Different comparison groups imply different counterfactuals.
Together, the studies support a few conclusions: the exemption sharply reduced police contact for possession while it applied in full, and neither study found a significant change in deaths. A reader cannot conclude that decriminalization alone raised or did not raise hospitalizations, because the two designs answer different questions, both covered only about one year after the policy, and hospitalizations could rise either because more people sought care or because harms increased.
Minimum 20 characters required.
Question 1: Which finding did Hughes and Stevens (2010) report about Portugal’s decriminalization of 1 July 2001?
Question 2: Zoorob and colleagues (2024) found a positive association between Oregon’s Measure 110 and fatal overdoses before one adjustment and no association after it. What did they adjust for?
Question 3: What did the amendment to British Columbia’s exemption that took effect on 7 May 2024 change?
Question 4: In a hypothetical difference-in-differences analysis, a province’s overdose rate rises from 10 to 14 per 100,000, and the comparison group’s rate rises from 5 to 7 per 100,000. What is the absolute difference-in-differences?
Legal Regulation of Cannabis: Canada, Uruguay, the United States and Germany
Introduction
Cannabis is the drug for which the debate about prohibition has moved furthest in Canada. Lesson 2 described the Le Dain Commission, whose majority recommended in the early 1970s that cannabis possession offences be repealed. Lesson 7 explained cannabis pharmacology, rising THC potency and the evidence on health effects, and Lesson 11 noted that the United Nations Commission on Narcotic Drugs removed cannabis from Schedule IV of the 1961 Convention in December 2020. On 17 October 2018, the Cannabis Act came into force and created a national legal market for non-medical cannabis in Canada.
This section treats cannabis as the main Canadian test of legal regulation. It first sets out the levers that any regulatory system uses, then describes Canada’s model and three others, and finally examines what has happened since 2018 to cannabis use, the legal and illegal markets, potency, poisonings in children and criminal charges.
Learning Objectives
- 5.1 Identify the main levers of legal regulation, such as minimum age, who supplies, purchase limits, home growing and registration, and compare four jurisdictions on them.
- 5.2 State the purposes of Canada’s Cannabis Act (in force 17 October 2018), how responsibilities are divided between federal and provincial governments, and what the 2024 legislative review concluded.
- 5.3 Compare the cannabis models of Canada, Uruguay, American states and Germany, placing each on the spectrum from non-commercial to commercial regulation.
- 6.1 Summarize Canadian evidence since legalization on use, legal sources and potency, with the years each figure covers.
- 6.2 Calculate the ratio of two average monthly rates from published Ontario data on poisonings in children, and explain why a trend-adjusted estimate can differ from an unadjusted one.
- 6.3 Summarize the fall in cannabis possession charges since legalization, and describe concerns about the Act’s treatment of First Nations, Inuit and Métis governments.
- 6.4 Summarize evidence on adolescent cannabis use after legalization in Uruguay and the United States.
5.1 The Tools of Legal Regulation
Section 1 defined legal regulation as a lawful supply governed by rules. The rules are where regulatory systems differ. Every system decides who may buy and how much, who may produce and sell and whether they may profit, which products are allowed and how they are packaged and labelled, whether people may grow their own, and how prices and taxes are set. Table 12.5 compares four jurisdictions on the levers that the sources in this section describe.
Table 12.5. Selected levers of cannabis regulation in Canada, Uruguay, American states with adult-use laws and Germany.
| Lever | Canada (from 2018) | Uruguay (law of 2013) | American states (2026) | Germany (law of 2024) |
|---|---|---|---|---|
| Minimum age | 18 federally; provinces may raise it | Over 18, citizens and permanent residents | Set by each state | 18 |
| Who supplies | Federally licensed for-profit producers; provinces and territories oversee sale | State-controlled production sold in pharmacies; registered clubs; home growers | For-profit commercial models in a number of states | Home growers and non-profit cultivation associations; no commercial sales |
| Possession or purchase limits | 30 g dried or equivalent in public | Pharmacy purchases of no more than 10 g a week | Set by each state | 25 g in public and 50 g at home |
| Home growing | Up to 4 plants per residence; provinces may lower | Up to 6 flowering plants per household | Set by each state | Up to 3 plants per adult |
| Registration of users | None | Required; one access route per person | Set by each state | Membership in one association only |
Sources: Department of Justice Canada (2021); Ramsey (2016); Laqueur et al. (2020); Pew Research Center (2026); Osborne Clarke (2024).
Table 12.5 shows that “legal cannabis” covers systems with very different degrees of commercialization, which Figure 12.2 in Section 1 placed on a spectrum. The next part looks more closely at Canada’s system.
5.2 Canada’s Cannabis Act
The Cannabis Act replaced criminal prohibition with a legal framework for producing, distributing, selling and possessing cannabis. Its design follows from its stated purposes.
5.2.1 The Purposes of the Act
Section 7 of the Act states that its purpose is “to protect public health and public safety,” and lists seven specific purposes (Cannabis Act, 2018, s. 7). Click each card below to read them, grouped into four themes.
These purposes give evaluators a list of outcomes to examine, and Parts 6.1 to 6.3 are organized around them. They also include possible tensions: a legal industry that grows quickly may displace the illegal market while also creating new inducements to use.
5.2.2 Who Does What
As of 2026, the federal government licenses producers and sets national rules for products, packaging and tracking, while provinces and territories oversee distribution and sale. Adults aged 18 and older may possess up to 30 grams of legal dried cannabis or its equivalent in public and grow up to four plants per residence; provinces and territories may raise the minimum age and lower the number of plants. Giving or selling cannabis to anyone under 18 is a criminal offence with a maximum penalty of 14 years in prison (Department of Justice Canada, 2021), so criminal law remains for selling to young people and for the illegal market.
5.2.3 Edibles and Other Products
At first, legal sales were limited mainly to dried flower, seeds and oils. Edible cannabis products reached stores in Ontario, British Columbia and Alberta from January 2020, under rules that required plain, child-resistant packaging and a maximum of 10 milligrams of THC per package. Quebec prohibited edible products such as candies, desserts and baked goods (Ottawa Hospital Research Institute, 2022). This difference between provinces later allowed researchers to study poisonings in children, as Part 6.2 describes.
5.2.4 The Legislative Review
The Act required a review of its operation. An expert panel chaired by Morris Rosenberg published its final report in March 2024. It concluded that “the evidence suggests that the framework includes appropriate controls and measures to manage the risks,” while identifying areas for improvement, particularly the growing consumption of higher-THC products (Expert Panel on the Legislative Review of the Cannabis Act, 2024). Its findings appear throughout Parts 6.1 to 6.3. Before turning to outcomes, the next part compares Canada’s model with three others.
5.3 Uruguay, the United States and Germany
Three other jurisdictions show the range of choices available within legal regulation, from state control to commercial markets to non-commercial home growing and clubs. The four tabs below describe each, with dates.
Laqueur and colleagues (2020) describe Uruguay as the first country in the world to legalize recreational cannabis, in 2013. Law No. 19,172, enacted in December 2013 and administered by the Institute for the Regulation and Control of Cannabis (IRCCA), let registered citizens and permanent residents over 18 use one of three routes: home growing of up to six flowering plants per household, non-profit clubs of 15 to 45 members, or pharmacy purchases of no more than 10 grams a week (Ramsey, 2016). The law gave the government control of all large-scale production. Registration of home growers began in August 2014 and of clubs in October 2014, while pharmacy sales began only in July 2017 (Laqueur et al., 2020). Uruguay’s law is an example of non-commercial state regulation.
As of May 2026, 24 states and the District of Columbia allowed recreational as well as medical cannabis use under state law, according to the Pew Research Center (2026), drawing on the National Conference of State Legislatures. Many of these states adopted for-profit commercial models, in which licensed private companies grow, process, market and sell cannabis (Laqueur et al., 2020). Rules on age, possession, home growing, product types, advertising and taxes are set state by state, so the United States is a set of separate experiments that researchers can compare.
Germany’s parliament passed the Cannabis Act (Cannabisgesetz) on 23 February 2024, and the law dated 27 March 2024 came into force on 1 April 2024 (Osborne Clarke, 2024; Bundesgesetzblatt, 2024). As passed, the law let adults possess up to 25 grams in public and 50 grams at home and grow up to three plants. From 1 July 2024, non-profit cultivation associations of up to 500 adult members could supply members with up to 25 grams a day and 50 grams a month; members aged 18 to 21 could receive up to 30 grams a month, with THC content limited to 10 percent. The law did not permit commercial sales, and when it was passed a planned second stage of regional commercial pilot projects was uncertain (Osborne Clarke, 2024).
The European Union Drugs Agency reported in June 2025 that Malta legislated in December 2021 for limited home growing and non-profit growing clubs, and that Luxembourg legislated in July 2023 to permit limited home growing and private use. It also reported that the Netherlands had tolerated sales of up to 5 grams to adults in coffeeshops for decades and was experimenting with a closed, regulated supply chain in 10 municipalities, and that Switzerland had begun pilot trials of sales in certain cities (European Union Drugs Agency, 2025).
The four models differ most in who may supply cannabis and whether they may profit from it. Canada and many American states sit at the commercial end of Figure 12.2, while Uruguay and Germany sit nearer the non-commercial middle. The next part asks what has happened in Canada since 2018.
6.1 Outcomes in Canada since 2018
Evaluating legalization is difficult for the reasons set out in Section 2: there is no Canada without legalization to compare with, the legal market opened gradually, and the COVID-19 pandemic arrived 17 months after the Act came into force. The evidence below is therefore mostly descriptive, and the strongest designs are those that compare provinces with different rules.
6.1.1 Cannabis Use and Young People
Health Canada’s Canadian Cannabis Survey is an annual online survey of people aged 16 and older; the 2023 cycle had about 11,690 respondents (Health Canada, 2024c). In 2024, 26 percent of respondents reported using cannabis in the past 12 months, compared with 22 percent in 2018 (Health Canada, 2024d). Among people aged 16 to 19, past-year use was 36 percent in 2018, 44 percent in 2019 and 41 percent in 2024. The expert panel described youth use as relatively stable since legalization, while noting that Canada continued to report among the highest rates of youth cannabis use in the world and that more than 4 in 10 people aged 20 to 24 reported past-year use (Expert Panel on the Legislative Review of the Cannabis Act, 2024). Interactive 12.3 lets you explore these survey results by indicator.
Choose an indicator to see the percentages Health Canada reported for each year shown. Years marked “not reported” were not given for that indicator in the 2024 summary. “Consumers” means people who used cannabis in the past 12 months, and a legal usual source means a legal store or website.
Source: Health Canada (2024d), Canadian Cannabis Survey 2024 summary. Percentages of respondents aged 16 and older, or of past-year consumers where stated.
Interactive 12.3 shows a modest rise in overall use, no clear trend at ages 16 to 19, and a large shift in where people obtain cannabis. The survey measures self-reported behaviour among people willing to answer online, so small changes between years should be read with caution.
6.1.2 The Legal and Illegal Markets
Among people who used cannabis in the past 12 months, the share whose usual source was a legal store or website rose from 37 percent in 2019 to 72 percent in 2024, and the share using an illegal source fell from 28 percent in 2018 to 3 percent in 2024 (Health Canada, 2024d). The expert panel reported “steady progress in shifting adult consumers to the legal cannabis market,” uneven across the country, and judged that “the illicit cannabis market remains entrenched” (Expert Panel on the Legislative Review of the Cannabis Act, 2024). The two descriptions measure different things: the survey asks about each consumer’s usual source, while the panel considered the persistence of illegal production and sale.
6.1.3 Potency
Lesson 7 described the long-term rise in THC concentrations. Under legalization, the expert panel reported in 2024 that the majority of dried cannabis sold contained more than 20 percent THC, and that high-potency vaping products and infused pre-rolled joints appeared to be gaining market share. The panel said it was increasingly concerned about this shift and cited studies linking high-potency products to increases in health care visits for cannabis-related psychosis (Expert Panel on the Legislative Review of the Cannabis Act, 2024). A legal market can regulate potency, but when the panel reported in 2024, Canada’s rules permitted high-potency dried products.
6.2 Unintentional Poisonings in Young Children
A systematic review by Allaf and colleagues (2023) found that most studies of acute poisoning reported more cannabis poisonings after legalization or decriminalization, with a larger increase in studies of children, and identified edibles as an important driver. An Ontario study by Myran and colleagues (2022a) illustrates this for young children who eat cannabis products by accident. The study examined every emergency department visit for cannabis exposure among children under 10 in Ontario across three periods: before legalization (January 2016 to September 2018), after legalization of dried flower, seeds and oils (October 2018 to January 2020), and after edibles and other products reached stores (February 2020 to March 2021). Figure 12.4 shows the average number of visits per month in each period.
Comparing periods of different lengths requires a rate, here the number of visits per month. The ratio of two such rates tells you how many times higher one period was than another.
Worked Example 12.2: Edibles and emergency visits in Ontario
Before legalization there were 81 visits in 33 months, an average of 81 ÷ 33 = 2.45 visits a month. After edibles reached stores there were 317 visits in 14 months, an average of 317 ÷ 14 = 22.64 visits a month.
Rate ratio = 22.64 ÷ 2.45 = 9.2. Visits were about nine times as frequent each month after edibles became available as before legalization.
Myran and colleagues (2022a), who used population rates of children, reported an unadjusted incidence rate ratio of 9.12 (95 percent confidence interval 7.15 to 11.65). When they adjusted for the steady upward trend across the whole study period, the ratio for the edibles period was 2.23 (1.17 to 4.27), still a statistically significant increase. The adjusted figure asks how much the edibles period exceeded what a continuing trend alone would predict. The share of visits that led to admission to hospital also rose, from 24.7 percent before legalization to 38.5 percent in the edibles period.
Use Calculator 12.2 to repeat Worked Example 12.2 for both periods after legalization. The third preset uses hypothetical numbers to show why the length of each period matters.
A second study by Myran and colleagues (2022b) compared provinces. In the provinces studied, hospitalizations for cannabis poisoning in children aged 0 to 9 rose 2.6-fold after flower was legalized. In Ontario, British Columbia and Alberta, which allowed edibles, rates then rose to 7.5 times the pre-legalization rate, while in Quebec, which prohibited edibles such as candies and desserts, they stayed at about 3.0 times (Ottawa Hospital Research Institute, 2022). The expert panel reported that such hospitalizations among children aged 0 to 11 outside Quebec rose from 0.5 per 100,000 in 2015 to 6.4 per 100,000 in 2021 (Expert Panel on the Legislative Review of the Cannabis Act, 2024). These observational findings associate a specific regulatory choice, the legal sale of edibles, with a specific harm.
6.3 Criminal Charges and First Nations, Inuit and Métis Jurisdiction
Purpose (e) of the Act was to reduce the burden on the criminal justice system. The expert panel cited a 95 percent reduction in charges for cannabis possession between 2017 and 2022 (Expert Panel on the Legislative Review of the Cannabis Act, 2024). Lesson 11 described how cannabis arrests before legalization fell unequally on Black and Indigenous people, so this reduction also concerns equity.
First Nations, Inuit and Métis jurisdiction
The expert panel found that First Nations, Inuit and Métis “were not adequately consulted when the Act and related measures were developed,” and that the Act “is silent on the authority of First Nations, Inuit and Métis governments,” which it called a significant gap. It cited an industry count from February 2023 in which fewer than 1 percent (24 of more than 3,300) of provincially or territorially authorized retail stores were on reserves, and it recommended nation-to-nation agreements that would let communities exert more control over cannabis activities in their territories (Expert Panel on the Legislative Review of the Cannabis Act, 2024). The issue concerns self-determination and economic participation as well as health, which HSCI 214 develops in its treatment of colonialism.
A systematic review of Canadian research by Hall and colleagues (2023) drew a similar picture: legalization appears to have reduced cannabis arrests and increased the share of purchases from legal sources, along with access to more potent products and acute harms among adults and children. The review noted that the outcomes assessed had been limited to the first 6 to 24 months after legalization, so this evidence comes mainly from early studies and research is still evolving. Evidence from other countries helps to judge whether these results are typical.
6.4 Evidence from Uruguay and the United States
Adolescent use is often raised in debates about legalization, and studies in other countries offer comparisons. Laqueur and colleagues (2020) compared Uruguay with a synthetic control built from 15 Chilean regions, using school surveys of students in their eighth, tenth and twelfth school years from 2014 to 2018. They found no meaningful change in adolescent cannabis use or in the perceived risk of frequent use, with weak evidence that perceived availability rose. They noted that pharmacy sales began only in July 2017, so most of their study period preceded full implementation.
Laqueur and colleagues (2020) also reviewed American studies, which were mixed: one found increased past-month use among younger students in Washington and no effect in Colorado (Cerdá et al., 2017), and a study using data from all states found no increase in adolescent use but small increases in cannabis use disorder among adolescents who used cannabis (Cerdá et al., 2020). A systematic review of quasi-experimental studies, most of them from the United States, by Windle and colleagues (2026) found that legalization may increase cannabis use among young adults and adults, with weaker and less consistent effects among youth. Research on legalization and youth use is still evolving. Try it 12.3 asks you to bring this evidence together for one purpose of the Canadian Act.
Choose one purpose of the Cannabis Act from Part 5.2. Using Interactive 12.3, Calculator 12.2 and the evidence in Parts 6.1 to 6.4, write a short assessment that (1) names two pieces of evidence relevant to that purpose, with their years and sources, (2) states whether each suggests progress, harm or no clear change, (3) names one limitation of each piece of evidence, and (4) proposes one regulatory lever from Table 12.5 that could be changed to improve results on that purpose, with the trade-off it would involve.
Summary
Legal regulation creates a lawful supply governed by rules, and real systems differ most in how commercial they are. Canada’s Cannabis Act (17 October 2018) pursues seven purposes under the aim of protecting public health and safety, and its 2024 expert panel judged its controls appropriate while urging action on high-THC products. Uruguay built a non-commercial state-controlled model, many American states built commercial markets, and Germany (1 April 2024) allowed home growing and non-profit associations. In Canada since 2018, past-year use rose modestly, youth use showed no clear trend, most consumers moved to legal sources, possession charges fell, high-potency products gained ground, and poisonings in young children rose where edibles were sold. Section 4 uses this evidence to weigh the three approaches and draw the course together.
Key Takeaways
- 5.1 Legal regulation is defined by its rules, and every system decides who may buy and how much, who may supply and whether they may profit, which products are allowed, whether people may grow their own, and whether users must register. Comparing jurisdictions lever by lever shows that “legal cannabis” covers systems with very different degrees of commercialization.
- 5.2 Canada’s Cannabis Act (17 October 2018) aims to protect public health and public safety through seven purposes, from protecting young people to reducing the burden on the criminal justice system. As of 2026, the federal government licenses producers and sets product rules, provinces and territories oversee sale, and the 2024 expert panel judged the controls appropriate while urging action on higher-THC products.
- 5.3 Uruguay (law of 2013) built a non-commercial, state-controlled model with registration, Germany (1 April 2024) allowed home growing and non-profit associations without commercial sales, and many of the 24 American states with adult-use laws as of May 2026 built for-profit markets. Canada and many American states sit at the commercial end of the spectrum, and Uruguay and Germany nearer the non-commercial middle.
- 6.1 In the Canadian Cannabis Survey, past-year use rose from 22 percent in 2018 to 26 percent in 2024, with no clear trend at ages 16 to 19, and the share of consumers whose usual source was legal rose from 37 percent in 2019 to 72 percent in 2024. In 2024 the expert panel judged the illegal market still entrenched and reported that most dried cannabis sold contained more than 20 percent THC.
- 6.2 A ratio of two average monthly rates compares periods of different lengths: in Ontario, emergency visits for cannabis exposure among children under 10 were about nine times as frequent each month after edibles reached stores as before legalization. Adjusting for the existing upward trend reduced the estimate to 2.23, still a significant increase, and provinces that sold edibles saw larger rises than Quebec, which prohibited many of them.
- 6.3 Charges for cannabis possession fell by 95 percent between 2017 and 2022, a change that also concerns equity because earlier arrests fell unequally on Black and Indigenous people. The expert panel found that First Nations, Inuit and Métis were not adequately consulted and that the Act is silent on their governments’ authority.
- 6.4 In Uruguay, a comparison with Chilean regions found no meaningful change in adolescent cannabis use from 2014 to 2018, although pharmacy sales began only in July 2017. American studies were mixed, and a 2026 systematic review found weaker and less consistent effects of legalization on use among youth than among adults.
Reflection
Canada’s Cannabis Act (in force 17 October 2018) lists seven purposes, including (a) protecting the health of young persons by restricting their access to cannabis and (c) providing for the licit production of cannabis to reduce illicit activities. The Canadian Cannabis Survey reported past-year use among people aged 16 to 19 of 36 percent in 2018, 44 percent in 2019 and 41 percent in 2024, and the share of past-year consumers whose usual source was legal rose from 37 percent in 2019 to 72 percent in 2024. In Ontario, emergency department visits for cannabis exposure among children under 10 averaged about 2.5 a month before legalization and 22.6 a month after edible products reached stores in 2020. Write a balanced assessment of how well the Act is meeting purposes (a) and (c), and propose one change to a regulatory lever, stating the trade-off it would involve.
On purpose (c), the evidence suggests clear progress. The share of past-year consumers whose usual source was legal roughly doubled between 2019 and 2024, to 72 percent, so by 2024 most people who used cannabis bought mainly from a quality-controlled supply. The expert panel judged in 2024 that the illegal market remained entrenched, and the survey measures usual source among people willing to answer online, so the size of the remaining illegal market is uncertain.
Purpose (a) is mixed. Past-year use among 16 to 19 year olds rose between 2018 and 2019 and then sat at 41 percent in 2024, which Health Canada described as no clear trend; the Act has not clearly increased or reduced youth use, and the expert panel noted in 2024 that Canada’s rates were among the highest in the world. The poisoning data show a clearer failure for young children: monthly emergency visits rose about ninefold after edibles reached stores, and Quebec, which banned candy-like edibles, saw a smaller rise in hospitalizations.
One change would be to restrict edible products that resemble candies and desserts, as Quebec did. The likely benefit is fewer poisonings in young children. The trade-off is that some adults who prefer edibles to smoking might return to the illegal market, which would work against purpose (c), so the change would need monitoring of both outcomes.
Minimum 20 characters required.
Question 1: Under Canada’s Cannabis Act, which level of government oversees the distribution and sale of cannabis?
Question 2: Which feature of Germany’s cannabis law, as passed (in force 1 April 2024), distinguishes it from Canada’s Cannabis Act?
Question 3: In Ontario, emergency visits for cannabis exposure among children under 10 averaged about 2.5 a month before legalization and 22.6 a month after edibles reached stores (Myran et al., 2022a). What is the approximate ratio of these monthly rates?
Question 4: Which statement about youth cannabis use in Canada since legalization best matches the evidence in this section?
Integrating the Course: Weighing Approaches and Substance Use in Syndemic Terms
Content note. This section discusses overdose deaths, violence and trauma, and colonial policies, including their links with substance use and HIV. If you would like support, the 9-8-8 Suicide Crisis Helpline is available by call or text at 9-8-8, the National Overdose Response Service can be reached at 1-888-688-6677, and the Hope for Wellness Help Line for Indigenous people can be reached at 1-855-242-3310.
Introduction
The first three sections of this lesson described alternatives to prohibition and the evidence on them. This final section does two things. It asks how a careful reader can weigh prohibition, decriminalization and legal regulation against one another when the evidence is incomplete and sometimes conflicting, and it steps back to look at the whole course, linking substance use to the other health problems with which it clusters.
The aim of the first three parts is a method of weighing evidence. It leaves the choice of policy to the reader, because the evidence alone does not settle it: people who agree on the facts can still disagree about which outcomes matter most.
Learning Objectives
- 7.1 Identify the criteria on which drug policies can be compared, and explain why the weight given to each criterion shapes the conclusion.
- 7.2 Explain what the costs and deaths attributed to legal alcohol and tobacco in Canada suggest about commercial legal regulation.
- 7.3 Weigh prohibition, decriminalization and legal regulation against Canadian evidence from cannabis legalization, British Columbia’s exemption and alcohol, stating the criteria and weights that drive a conclusion.
- 8.1 Summarize how Lessons 1 to 11 build an account of what drugs do, why people use them, why some develop addiction and how societies control them.
- 8.2 Define a syndemic, distinguish it from comorbidity, and state the three criteria proposed by Singer and colleagues.
- 8.3 Describe how substance use clusters and interacts with mental illness, violence and trauma, stigma, HIV and other sexually transmitted and blood-borne infections, and structural drivers, and name the course in the HEAL Learning Lab syndemics series that develops each of these strands.
7.1 Criteria for Comparing Drug Policies
Weighing policies means comparing them on several outcomes at once. This part names the criteria; the next part adds the evidence from legal alcohol and tobacco, and Part 7.3 sets out the Canadian evidence in a single table and an interactive tool.
The evidence in this course bears on at least six criteria: deaths and acute health harms; substance use, especially by young people; criminal justice harms, including arrests, records and their unequal distribution; control over what is in the product; the size and power of illegal markets; and public order and community concerns about public drug use. These criteria match the purposes stated in Canada’s Cannabis Act (Section 3) and the concerns that led British Columbia to narrow and then end its exemption (Section 2). A policy can perform well on one criterion and poorly on another, which is why the weight a person gives to each criterion shapes the conclusion.
7.2 What Legal Alcohol and Tobacco Teach
Alcohol and tobacco have been sold legally by commercial businesses in Canada for far longer than cannabis, and they show that legality does not mean low harm. The Canadian Substance Use Costs and Harms project estimated that substance use cost Canada $49.1 billion in 2020, of which alcohol accounted for $19.7 billion, just over 40 percent, and that substance use caused nearly 74,000 deaths in 2020, with alcohol and tobacco accounting for over 85 percent of them (Canadian Centre on Substance Use and Addiction [CCSA], 2023). Lesson 4 introduced Canada’s Guidance on Alcohol and Health (January 2023) as an example of risk communication. These figures reflect how widely alcohol and tobacco are used as well as how they are regulated, and HSCI 402 develops the evidence on alcohol and tobacco policies such as pricing, availability and marketing rules. For the comparison here, alcohol shows that a legal, commercial market can coexist with very large harms, so the content and enforcement of the rules within legal regulation deserve as much scrutiny as the decision to legalize.
7.3 Weighing Prohibition, Decriminalization and Legal Regulation
Table 12.6 places the main Canadian and comparative evidence from this course under each criterion and approach. Each cell summarizes evidence presented earlier, with its source; where the evidence conflicts or is missing, the cell says so.
Table 12.6. Evidence from this course on prohibition, decriminalization and legal regulation, by criterion.
| Criterion | Prohibition | Decriminalization | Legal regulation |
|---|---|---|---|
| Deaths and acute harms | The unregulated supply came to be dominated by fentanyl and sedative adulterants (Lesson 5); while cannabis was prohibited, emergency visits for cannabis exposure among children under 10 in Ontario averaged 2.5 a month, fewer than after legalization (Myran et al., 2022a) | No significant change in deaths in British Columbia relative to the rest of Canada in the first year (Gaudreault et al., 2025) or relative to Alberta and Ontario in a series to March 2025 (Gaudreault et al., 2026); conflicting estimates for Oregon (Section 2) | Child cannabis poisonings rose where edibles were sold (Myran et al., 2022a, 2022b); alcohol and tobacco caused over 85% of substance use deaths in 2020 (CCSA, 2023) |
| Use, especially by young people | 36% of 16 to 19 year olds reported past-year cannabis use in 2018, the year legalization began (Health Canada, 2024d) | Portugal: adult lifetime use rose from 7.6% to 12.0% and past-year use from 3.4% to 3.7% (2001 to 2007) (Hughes & Stevens, 2010) | Canada: past-year use 22% to 26% (2018 to 2024), no clear trend at ages 16 to 19; Uruguay: no meaningful change among adolescents (Laqueur et al., 2020) |
| Criminal justice harms | Arrests and records for possession, falling unequally on Black and Indigenous people (Lesson 11) | British Columbia: possession offences fell from 509 to 165 a month (Government of British Columbia, 2025); Portugal: share of prisoners with drug offences fell from 44% to 21% (1999 to 2008) (Hughes & Stevens, 2010) | Cannabis possession charges fell 95% from 2017 to 2022 (Expert Panel, 2024) |
| Control over the product | Content unknown to the buyer (Lesson 5) | Supply unchanged and still unregulated (Section 1, 2.2 Classifying Policies on a Spectrum of Control) | Product rules such as 10 mg THC per edible package; most dried cannabis above 20% THC (Expert Panel, 2024) |
| Illegal markets | Enforcement associated with a shift toward more potent drugs, the iron law (Lesson 11) | Supply remains with illegal markets | 72% of consumers used legal sources in 2024 (Health Canada, 2024d); the panel judged the illegal market “entrenched” |
| Public order and community concerns | Enforcement aimed at public order was one of Vancouver’s four pillars (Section 1) | British Columbia asked for public spaces to be excluded from 7 May 2024 over concerns about public drug use (Health Canada, 2024a) | Provinces and territories oversee where cannabis is sold (Department of Justice Canada, 2021) |
Note on deaths and acute harms: deaths stayed high under both prohibition and decriminalization because the supply was unregulated under both (Section 1, 2.2 Classifying Policies on a Spectrum of Control). For example, there were 1,826 drug toxicity deaths in British Columbia in 2025, while the exemption applied (preliminary; BC Coroners Service, 2026).
Table 12.6 shows that the approaches are rarely compared on the same drug at the same time. The evidence on legal regulation comes mostly from cannabis, alcohol and tobacco, while the evidence on decriminalization comes from opioids and stimulants in a toxic supply. Both bodies of evidence are still evolving: a 2026 systematic review judged the longer-term impacts of British Columbia’s pilot uncertain (Ali et al., 2026), and a 2023 review of Canadian cannabis research found that the outcomes assessed covered only the first two years after legalization (Hall et al., 2023). Any overall judgement has to bridge these differences, and Interactive 12.4 makes the role of weights visible.
For each criterion, set how important it is to you, then judge how each approach performs on it (1 performs poorly, 2 mixed or unclear, 3 performs well), using the evidence note and Table 12.6. All judgements start at 2. The weighted totals update as you go, and the two emphasis buttons change only the weights. The tool records nothing; it shows how your own weights and judgements combine.
Two students who make identical judgements about the evidence in Interactive 12.4 can reach different totals if one gives most weight to deaths and the other to criminal justice harms. That is the main lesson of the tool: disagreements about drug policy are often disagreements about values and weights, and evidence can narrow them without always resolving them. The next part turns from weighing to looking back over the whole course.
8.1 Looking Back Across the Course
Each lesson of this course added one layer to an account of drugs and society. The four items below summarize how the lessons fit together; open each to review the ideas that this lesson has drawn on.
Lesson 1 defined psychoactive substances, showed that the categories of licit, illicit, prescribed and regulated are legal and social constructions, introduced Health Canada’s spectrum of substance use, and showed how language can carry stigma. Lesson 2 traced how substances became licit or illicit, from ceremonial tobacco and the opium trade to temperance, the Opium Act of 1908 and the Le Dain Commission (1969 to 1973). Those categories are what this lesson’s alternatives propose to change.
Lesson 3 introduced routes of administration, pharmacokinetics, pharmacodynamics and the influence of set and setting. Lessons 4 to 7 applied this pharmacology to alcohol and other depressants, to opioids and the toxic unregulated supply, to stimulants, tobacco and vaping, and to cannabis, psychedelics and new psychoactive substances. The point about the drug supply in Section 1 (2.2 Classifying Policies on a Spectrum of Control) and the evidence on potency and edibles in Section 3 depend on this pharmacology.
Lesson 8 explained the neurobiology of reward, learning and neuroadaptation, including gambling. Lesson 9 examined why some people develop addiction and others do not, from genetics and development to adversity and natural recovery, and the debate over the brain disease model. Lesson 10 described motives for use, the social ecological model, the social determinants of substance use harms and populations with unique needs. These lessons explain why the same policy affects different people differently.
Lesson 11 examined the international drug control system, the war on drugs in the United States and Canada, policing, incarceration, race and human rights, diversion, and the iron law of prohibition. This lesson examined public health frameworks, decriminalization, the legal regulation of cannabis and the evidence on each, and how to weigh them.
Read as a whole, the course moves from the molecule to the person to the society. The final two parts show why these levels cannot be studied in isolation, by introducing the idea of a syndemic.
8.2 Substance Use in Syndemic Terms
Substance use often clusters with other health problems in a population. People who are most harmed by drugs are often also affected by mental illness, violence, infection, stigma and poverty, and these conditions make one another worse. The syndemic concept gives this pattern a name.
Singer and colleagues (2017) define a syndemic as the aggregation of two or more diseases or health conditions in a population in which an adverse biological, behavioural or social interaction worsens the effects of one or more of them. Syndemics are most likely under conditions of inequality, such as poverty, stigma, stress or structural violence. The concept differs from comorbidity, which describes conditions that simply occur together in the same person; a syndemic approach also asks how the conditions interact and what social conditions drive that interaction (Singer et al., 2017). The box below lists the three criteria Singer and colleagues proposed.
Three criteria for a syndemic
Singer and colleagues (2017) proposed that a syndemic involves (1) two or more conditions that cluster within a population; (2) social factors, such as poverty, stigma or structural violence, that create that clustering; and (3) an adverse interaction between the conditions that increases the overall burden of disease. Comorbidity describes conditions occurring together, without the second and third criteria.
Singer and colleagues (2017) describe SAVA, which stands for substance abuse, violence and AIDS, as the first syndemic identified in the research literature and the one most heavily investigated. Merrill Singer (1996) used the term to describe how drug use, violence and HIV in low-income urban neighbourhoods in the United States formed a tightly linked set of health and social crises (the older word “abuse” is part of the name). HSCI 431 develops syndemic theory and its critiques; this section uses the concept only to connect the strands of the course.
8.3 The Strands of the Course
Figure 12.5 places substance use at the centre of five strands that this course has touched, and names the lesson where each appeared and the series course that develops it.
Each strand in Figure 12.5 interacts with substance use in both directions, and each has appeared in this course. The paragraphs below orient you to each strand and to where it is developed; they do not teach the content of those courses.
Mental health and illness. Lesson 10 introduced the self-medication hypothesis, in which some people use substances to cope with distress, and substance use can in turn worsen mental health. HSCI 214 develops mental illness and the stress response, and HSCI 402 develops concurrent disorders and their treatment.
Violence and trauma. Lesson 10 described childhood adversity as a risk factor, and Lesson 11 described violence in illegal drug markets under prohibition. HSCI 214 develops adverse childhood experiences and intergenerational trauma.
Stigma. Lesson 1 showed how stigmatizing language shifts clinicians’ judgements, and Lesson 2 showed how racialized drug panics shaped early laws. Stigma can deter people from seeking care, and reducing stigma around access to care was one stated purpose of British Columbia’s decriminalization (Toward the Heart, 2026). HSCI 214 develops stigma theory.
HIV, hepatitis C and other infections. Case 12.1 described public health emergencies over HIV and hepatitis C among people who inject drugs in Vancouver in the 1990s, and Table 12.3 showed new HIV cases among people who use drugs in Portugal falling from 907 in 2000 to 267 in 2008. HSCI 212 develops infections among people who use drugs, and HSCI 431 develops the global HIV epidemic and syndemic theory.
Structural drivers. Lesson 10 described poverty, housing, colonialism, racism, dislocation and criminalization as social determinants of substance use harms. Criminalization is itself a policy choice, which is why the alternatives in this lesson belong in a syndemic account. HSCI 214 develops colonialism and minority stress.
Return to Case 12.1. In the mid-1990s, overdose deaths, HIV and hepatitis C among people who inject drugs, and an open drug market were concentrated in one neighbourhood (MacPherson, 2005). In the terms of the three criteria for a syndemic (Part 8.2), the conditions clustered in one population; social factors such as poverty, housing and criminalization plausibly drove that clustering; and the conditions interacted, for example through the sharing of injection equipment. The Four Pillars response addressed several strands together, with harm reduction services aimed at both overdose and infection. Seeing the problem as a syndemic changes the evaluation question: a policy is judged by its effects across all the strands, and an evaluation that counts only one outcome, such as deaths or arrests, sees only part of the picture.
Case 12.2 connects the syndemic idea back to the evaluation problems of Section 2, where each study measured one or two outcomes. Table 12.7 lists where each strand is developed in the series.
Table 12.7. The strands of a syndemic account of substance use, where this course introduced each, and the HEAL Learning Lab course that develops it.
| Strand | Introduced in HSCI 180 | Developed in |
|---|---|---|
| Screening, surveillance, burden, overdose mortality, prevention, treatment, harm reduction, safer supply, alcohol and tobacco regulation, policy evaluation | Lessons 1, 4, 5 and 12 | HSCI 402, Substance Use, Addiction and Public Health |
| Stigma, the stress response, adverse childhood experiences, colonialism, intergenerational trauma, minority stress, mental illness | Lessons 1, 2 and 10 | HSCI 214, Perspectives on Mental Health and Illness |
| Infections among people who use drugs | Lessons 3, 6 and 12 | HSCI 212, Perspectives on Infectious and Immunological Diseases |
| The global HIV epidemic and syndemic theory | Lesson 12 | HSCI 431, The Global HIV/AIDS Epidemic |
| Psychedelic therapy, its trials and its regulation | Lesson 7 | HSCI 485, psychedelic medicine seminar |
| Community-based research and implementation science | Lesson 12 (evaluation) | HSCI 8XX, graduate course |
Table 12.7 completes the map of the series. The remaining parts summarize the section and apply the lesson to a real Canadian decision.
Summary
Weighing prohibition, decriminalization and legal regulation requires several criteria at once, including deaths, use, criminal justice harms, product control, illegal markets and public order, and the conclusion depends on the weight given to each as well as on the evidence. Canadian evidence shows that decriminalization in British Columbia reduced police contact for possession without a clear change in deaths, that cannabis legalization moved most consumers to legal sources and cut possession charges while raising poisonings in young children where edibles were sold, and that legal alcohol and tobacco accounted for over 85 percent of substance use deaths in Canada in 2020, with alcohol alone accounting for just over 40 percent of the cost of substance use. Across the course, the lessons moved from molecule to person to society, and the syndemic concept, which describes conditions that cluster, interact and share social causes, links substance use to mental health, violence and trauma, stigma, HIV and other infections, and structural drivers. HSCI 402, HSCI 214, HSCI 212, HSCI 431 and HSCI 485 develop these strands.
Key Takeaways
- 7.1 Drug policies can be compared on at least six criteria: deaths and acute harms, use by young people, criminal justice harms, control over the product, illegal markets, and public order. A policy can perform well on one criterion and poorly on another, so the weight a person gives each criterion shapes the conclusion.
- 7.2 Substance use cost Canada an estimated $49.1 billion in 2020, and alcohol and tobacco, both legal and commercially regulated, accounted for over 85 percent of the nearly 74,000 substance use deaths that year. Legality does not mean low harm, so the content and enforcement of the rules within legal regulation deserve as much scrutiny as the decision to legalize.
- 7.3 Set side by side, the Canadian evidence, much of it from early studies, suggests that decriminalization reduced police contact for possession without a clear change in deaths, and that cannabis legalization moved most consumers to legal sources and cut charges while raising poisonings in young children where edibles were sold. Because the approaches are rarely tested on the same drug at the same time, people who agree on the evidence can still reach different conclusions by weighting the criteria differently.
- 8.1 The course moved from the molecule to the person to the society: Lessons 1 and 2 set out concepts and history, Lessons 3 to 7 explained how drugs act, Lessons 8 to 10 explained why people use substances and why some develop addiction, and Lessons 11 and 12 examined control and its alternatives.
- 8.2 A syndemic is the clustering of two or more conditions in a population, driven by social factors such as poverty, stigma or structural violence, with an adverse interaction that increases the overall burden of disease. Comorbidity describes conditions that occur together without the second and third criteria; SAVA is the syndemic that Singer and colleagues (2017) call the first identified in the research literature.
- 8.3 Substance use interacts in both directions with mental health and illness, violence and trauma, stigma, HIV, hepatitis C and other infections, and structural drivers such as poverty, housing, colonialism, racism and criminalization. HSCI 402 develops policy, prevention, treatment and harm reduction; HSCI 214 mental health, trauma, stigma and colonialism; HSCI 212 infections among people who use drugs; HSCI 431 the global HIV epidemic and syndemic theory; and HSCI 485 psychedelic medicine.
Applying This Lesson
The final task applies the whole lesson to the decision with which this course’s account of decriminalization ends.
On 14 January 2026, British Columbia announced that it would not seek renewal of its decriminalization exemption, which ended on 31 January 2026 (Section 2). Prepare a one-page evidence brief on that decision. In it, (1) summarize what the exemption did and how it changed in September 2023 and May 2024; (2) state the strongest evidence that supports ending the exemption and the strongest evidence that supports continuing it, naming the study, outcome and comparison for each; (3) identify which criteria in Interactive 12.4 the province’s statement appears to weight most heavily; (4) name two strands from Figure 12.5 that the exemption touched and the series course you would consult on each; and (5) propose one study design from Part 4.1 in Section 2 that could answer a question the existing evaluations leave open.
Reflection
This section defined a syndemic, following Singer and colleagues (2017), as two or more conditions that cluster in a population, with social factors that create the clustering and an adverse interaction that increases the overall burden of disease. Consider Vancouver’s Downtown Eastside in the mid-1990s, where public health emergencies were declared over HIV and hepatitis C among people who inject drugs and over drug overdose deaths, alongside a large open drug market. Using the three criteria, explain whether this situation fits the definition of a syndemic. Then describe how a syndemic view would change the way you evaluate a drug policy such as decriminalization, and name the HEAL Learning Lab courses you would turn to for two of the strands involved.
The Downtown Eastside in the mid-1990s appears to meet all three criteria. First, several conditions clustered in one population: overdose deaths, HIV and hepatitis C among people who inject drugs, all concentrated in one neighbourhood. Second, social factors plausibly created the clustering, including poverty, poor housing and the criminalization of drug use, which concentrated an open drug market and its risks in the same streets. Third, the conditions interacted: injection drug use from an unregulated supply raised the risk of both overdose and infection through shared equipment, and illness and stigma made it harder to reach care. A list of separate problems would describe comorbidity; the interactions and shared causes make it a syndemic.
A syndemic view would change how I evaluate decriminalization. Most of the studies in Section 2 measured one or two outcomes, such as deaths or hospitalizations. I would also want to measure contact with services, new HIV and hepatitis C infections, experiences of stigma and police contact, and whether effects differed across groups, since a policy might have small effects on deaths and larger effects on other strands. For infections among people who use drugs I would turn to HSCI 212, and for stigma, trauma and colonialism to HSCI 214; HSCI 431 develops syndemic theory itself.
Minimum 20 characters required.
Question 1: According to Singer and colleagues (2017), what distinguishes a syndemic from comorbidity?
Question 2: According to the Canadian Substance Use Costs and Harms project (CCSA, 2023), which statement about 2020 is accurate?
Question 3: In Interactive 12.4, two students make identical judgements about how each approach performs, yet the weighted totals rank the approaches differently. What best explains this?
Question 4: Which course in the HEAL Learning Lab syndemics series develops infections among people who use drugs?
Final Assessment
Bringing It All Together
This lesson asked what the alternatives to prohibition are and how to weigh the evidence about them. Section 1 defined the public health approach through Vancouver’s Four Pillars framework and Canada’s federal strategies, and distinguished prohibition, decriminalization, legal regulation and diversion. Section 2 examined decriminalization in Portugal (from 2001), Oregon (2021 to 2024) and British Columbia (2023 to 2026), and explained why evaluations of the same policy can conflict.
Section 3 used cannabis to examine legal regulation, comparing Canada, Uruguay, American states and Germany, and summarized Canadian evidence on use, legal sources, potency, child poisonings and charges. Section 4 brought the evidence together in a weighing exercise, showed that conclusions depend on weights as well as evidence, and placed substance use within a syndemic of mental illness, violence and trauma, stigma, infections and structural drivers. The final assessment below draws on all four sections.
Key Takeaways from this lesson
- A public health approach judges drug policy by its effects on health and social harm across a population. Vancouver’s Four Pillars framework (endorsed 15 May 2001) placed harm reduction beside prevention, treatment and enforcement and led to Insite (2003) and the Supreme Court of Canada’s 2011 PHS decision, and Canadian strategies from 1987 to 2016 adopted public health language while keeping possession a crime.
- Decriminalization removes criminal penalties for possession and leaves the supply illegal and unregulated; legal regulation creates a lawful, rule-bound supply; diversion operates within prohibition. Every policy should be classified with its date, because several have changed category over time.
- Portugal decriminalized possession of all drugs on 1 July 2001 with an expansion of treatment, and Hughes and Stevens (2010) found no major rise in use and falls in problematic use and HIV, without being able to attribute them to decriminalization alone. Oregon’s Measure 110 (1 February 2021 to 31 August 2024) used a waivable $100 fine, and British Columbia’s exemption (31 January 2023 to 31 January 2026) sharply reduced police-recorded possession offences while it applied in full.
- Evaluations estimate a counterfactual, and they disagree when they choose different comparison groups, baselines, follow-up periods, adjustments, outcomes and years. Evaluations of Oregon conflict largely over the arrival of fentanyl, and studies of British Columbia disagree about hospitalizations and find no significant change in deaths.
- Legal regulation varies in how commercial it is: Canada’s Cannabis Act (17 October 2018) and many American states allow for-profit supply, while Uruguay and Germany chose non-commercial models built on state control, home growing and non-profit clubs or associations.
- Under the Cannabis Act, most consumers moved to legal sources and possession charges fell by 95 percent from 2017 to 2022, while poisonings in young children rose where edibles were sold and high-potency products gained ground. In Uruguay, a comparison with Chilean regions found no meaningful change in adolescent use.
- Legal alcohol and tobacco accounted for over 85 percent of substance use deaths in Canada in 2020, so the content and enforcement of the rules within legal regulation deserve as much scrutiny as the decision to legalize. Conclusions about drug policy depend on the weights given to each criterion as well as on the evidence.
- The course moved from the molecule to the person to the society. A syndemic is a cluster of interacting conditions with shared social causes; HSCI 402, 214, 212, 431 and 485 develop the strands that this course introduced.
Core Concepts Reviewed
Section 1: the public health approach, the Four Pillars, Canada’s federal drug strategies from 1987 to 2016, section 56 exemptions, and the definitions of prohibition, de jure and de facto decriminalization, administrative sanctions, threshold quantities, legal regulation, commercialization and diversion.
Section 2: Portugal’s 2001 decriminalization and its commissions, Oregon’s Measure 110 and House Bill 4002, British Columbia’s exemption and its amendments, counterfactuals, before-and-after, interrupted time series, difference-in-differences and synthetic control designs, and the reasons evaluations conflict.
Section 3: the levers of legal regulation, the purposes and structure of Canada’s Cannabis Act, the Uruguayan, American and German models, the Canadian Cannabis Survey, potency, child poisonings and the rate ratio, criminal charges, and First Nations, Inuit and Métis jurisdiction.
Section 4: criteria and weights for comparing policies, the evidence from alcohol and tobacco, the arc of the course from molecule to society, the definition of a syndemic and its difference from comorbidity, SAVA, and the series courses that develop each strand.
The final reflection asks you to use the whole lesson to brief a provincial health committee on a real policy choice.
Reflection
Imagine that you have been asked to brief a provincial health committee in a Canadian province (not British Columbia) on whether to request a federal exemption to decriminalize possession of small amounts of drugs, similar to British Columbia’s exemption of 31 January 2023 to 31 January 2026. Drawing on the whole lesson, write a brief that (1) defines what such an exemption would and would not change, (2) presents the strongest evidence for and against from Portugal, Oregon and British Columbia, naming conflicting evaluations as conflicting, (3) explains what cannabis legalization and alcohol regulation in Canada suggest about legal regulation as a further alternative, and (4) states which criteria you would weight most heavily and why, acknowledging that others might weight them differently.
An exemption would remove criminal penalties for possessing small amounts of specified drugs, as British Columbia’s did below 2.5 grams. It would not change the supply: production and sale would remain crimes, so people would still buy drugs of unknown content from the unregulated market.
The strongest evidence in favour is that decriminalization reduces criminal justice contact. British Columbia’s police-recorded possession offences fell from 509 to 165 a month under the original exemption, Gaudreault and colleagues (2025) found a 57 percent fall in possession incidents, and Portugal’s share of prisoners with drug offences fell from 44 to 21 percent between 1999 and 2008 without a major rise in use. The strongest evidence against is that the evaluations in this lesson have not shown health benefits: none of the three British Columbia studies found a significant change in deaths, a 2026 systematic review judged the pilot’s longer-term impacts uncertain (Ali et al., 2026), Nguyen and colleagues (2025) found opioid hospitalizations 58 percent higher than a pre-2020 baseline, and evaluations of Oregon conflict, with Spencer (2023) estimating 182 extra deaths and Zoorob and colleagues (2024) finding no association after adjusting for fentanyl. British Columbia also narrowed its exemption over public drug use and ended it in 2026.
Cannabis legalization shows that legal regulation can move most consumers to a quality-controlled supply and cut possession charges, while also bringing harms such as child poisonings from edibles; alcohol shows that a commercial legal market can coexist with very large harms. I would weight deaths and criminal justice harms most heavily, because both fall hardest on people already marginalized. A committee member who weights public order more heavily could reasonably reach a different conclusion from the same evidence.
Minimum 30 characters required.
Final Knowledge Assessment
Question 1: Which pairing of a policy and an approach from Figure 12.2 is accurate?
Question 2: Why does the box on why the supply column matters (Section 1, 2.2 Classifying Policies on a Spectrum of Control) suggest that decriminalization alone would not be expected to change what is in the drugs people use?
Question 3: Oregon courts recorded 10,028 Class E possession cases from February 2021 to August 2024. All charges were dismissed in 704, and 7,925 ended in conviction after the person failed to appear. What does this suggest?
Question 4: A study finds opioid hospitalizations in British Columbia 58 percent higher in 2023 than in a baseline set before safer supply began in 2020. Which conclusion is most defensible?
Question 5: Two accounts of Portugal’s reform reported drug-related deaths falling from close to 400 to 290 (1999 to 2006) and rising from 216 to 314 (2006 to 2007). What do these figures illustrate?
Question 6: Which statement about the Cannabis Act and criminal charges is supported by the expert panel’s 2024 report?
Question 7: Why did the comparison of Quebec with Ontario, British Columbia and Alberta strengthen the link between legal edibles and poisonings in young children?
Question 8: Which evaluation design builds a weighted combination of untreated places whose past outcomes closely match the place with the policy?
Question 9: What does Canada’s experience with legal alcohol and tobacco suggest for debates about the legal regulation of other drugs?
Question 10: In Interactive 12.4, one student gives “deaths and acute harms” a weight of 3 and other criteria a weight of 1; another gives the weight of 3 to “criminal justice harms.” If their performance judgements are identical, which is true?
Question 11: Which pairing of a syndemic strand with the series course that develops it is correct?
Question 12: Which statement about Vancouver’s Four Pillars framework and Insite is accurate?
Question 13: Uruguay and Canada both legalized cannabis. Which difference matters most when comparing their outcomes?
Question 14: The Canadian Cannabis Survey found that 72 percent of past-year consumers used a legal usual source in 2024, while the expert panel judged the illegal market “entrenched.” How can both be accurate?
Question 15: Which explanation best accounts for the conflicting evaluations of Oregon’s Measure 110 and overdose deaths?
Glossary: Key Terms, People & Frameworks
📚 Reference page, available throughout the lesson
Definitions of the key terms, frameworks and people in Lesson 12, with dates where a policy has changed.