A recent Atlantic essay, “The Cities That Said Yes to Drugs,” by Michael Powell, offers a disturbing portrait of Seattle, Burlington, and Vancouver. He describes open drug markets, fentanyl use in parks and on sidewalks, discarded needles, families retreating from public spaces, and governments struggling to reconcile compassion for people with addiction with their responsibility to the broader public.
Powell correctly identifies a central failure: a drug policy cannot claim compassion if it protects people who use drugs while leaving neighborhoods, parks, transit stops, and sidewalks dangerous or unusable. Addiction is a medical disorder; drug dealing is not. People living with addiction deserve dignity and care, but so do children, families, workers, business owners, and residents.
Yet Powell’s reporting demonstrates the failure of partial execution rather than harm reduction itself, an approach supported by substantial evidence for specific interventions. According to a Massachusetts study published in the BMJ by Alexander Walley and colleagues, communities with high implementation of overdose education and naloxone distribution experienced opioid-overdose death rates approximately 46 percent lower than communities without such programs. According to the Centers for Disease Control and Prevention, comprehensive syringe-service programs reduce HIV and hepatitis C incidence, with decades of research showing they do not increase illegal drug use or crime. The crucial issue is scope: harm reduction saves lives immediately, but it cannot carry the weight of an entire public health infrastructure alone.
Drug policy functions like a fixed menu rather than an à la carte selection. A complete system demands prevention, readily accessible treatment, harm reduction, mental-health care, housing, social reintegration, recovery support, and targeted law enforcement. Each addresses a different part of the problem. While lifesaving tools like naloxone must be deployed immediately, expecting single interventions to resolve a systemic crisis guarantees failure. One item on the menu should never be mistaken for the entire meal.
Switzerland recognized this dynamic over three decades ago. Confronted by open heroin scenes in the early 1990s, the Swiss government established its “Four Pillars” strategy: prevention, therapy, harm reduction, and enforcement. The Swiss framework aimed to reduce the consequences of drug use for both users and society, embedding this four-part balance into national law in 2008.
Switzerland embraced syringe exchanges, supervised consumption, and opioid-agonist treatments. Crucially, authorities also closed open drug scenes and retained enforcement against dealing and disorder. The outcomes reflect this shift. Federal health data show Swiss drug-related deaths dropped from more than 350 annually in the 1990s to 121 by 2012. Even with an uptick to 188 in 2024, mortality rates remain well below the peak of the open drug era. Swiss authorities also identify a major reduction in HIV infection among people who use drugs as one of the achievements of the four-pillar policy. Decades later, the Swiss model endures because maintaining public order preserved the political will required to sustain lifesaving health programs.
Vancouver offers a revealing contrast. In 2001, Vancouver adopted a similar Four Pillars framework designed to balance public health with public order, explicitly stating that open drug markets had to end. Public support was overwhelming, with nearly 90 percent of surveyed residents backing the integrated plan.
Given that history, the crucial question is why the balance among Vancouver’s four pillars broke down. Harm-reduction services expanded substantially, but prevention, treatment capacity, and enforcement of laws governing public disorder and drug dealing did not keep pace. The result illustrates what can happen when one pillar is expected to compensate for weaknesses in the others.
Supervised consumption facilities illustrate this balance in practice. As European drug authorities explicitly outline, consumption rooms are intended to prevent overdose and infectious disease while reducing public drug use and discarded equipment and connecting marginalized users with health and social services. They work best when accompanied by measures addressing open dealing, violence, and public disorder. The compact with the community is clear: drug use moves off the sidewalk and into a controlled environment; overdose interventions and treatment referrals become immediate; used needles are safely contained; and public dealing, violence, intimidation, and continued public consumption remain subject to law enforcement.
Portugal reached a similar balance through administrative reform. In 2001, Portugal replaced criminal penalties for personal drug possession with referrals to Commissions for the Dissuasion of Drug Addiction, while keeping trafficking illegal. Portugal’s reforms were accompanied by investments in prevention, treatment, harm reduction, social reintegration, and supply reduction. The Portuguese experience therefore cannot be attributed to decriminalization alone.
Portugal experienced substantial improvements in several health indicators after the reforms. Writing in the BMJ, Paula Vale de Andrade and Ludmila Carapinha reported that newly diagnosed HIV infections among people who injected drugs fell from 1,430 in 2000 to 352 in 2008. The same report, as subsequently corrected, placed officially recorded drug-related deaths at 76 in 2001 and 20 in 2008. That comparison must be interpreted cautiously, however, because Portuguese mortality systems have used different definitions and methods for identifying drug-related deaths. More recent national reporting recorded 80 overdose deaths in 2023, demonstrating that Portugal reduced important harms but did not eliminate drug mortality.
The vital takeaway, however, remains political and structural durability. Portugal’s integrated model has sustained public support for over two decades while keeping injection-associated HIV rates far below crisis levels. Decriminalization was never a magic bullet or a sign of governmental disengagement; it was the foundation of an active, lasting public health system.
Oregon highlights the danger of an unbalanced approach. Measure 110 removed criminal penalties for possession as Oregon was also confronting fentanyl, homelessness, visible public drug use, and longstanding deficiencies in its treatment system. Delays and weaknesses in implementation contributed to a political backlash against the measure. While research published in JAMA Psychiatry and JAMA Network Open indicates the rapid arrival of fentanyl—rather than the law itself—drove Oregon’s overdose surge, state audits confirmed the core failure: Oregon implemented decriminalization before the promised public-health and treatment infrastructure was fully operational and adequately coordinated.
A sustainable drug policy requires simultaneously building health systems, maintaining public order, and enforcing laws against drug dealing, violence, theft, and persistent public drug use. Readers reviewing the crises in Seattle, Burlington, and Vancouver should recognize that the failure lies not in providing harm reduction, but in expecting a single component of drug policy to substitute for the whole.
Lynn Webster, MD
Lynn R. Webster, MD, is a pain and addiction medicine specialist and serves as Executive Vice President of Scientific Affairs at Dr. Vince Clinical Research, where he consults with pharmaceutical companies.
