What anabolic-androgenic steroids reveal about the limits of current harm reduction models [Addiction opinion and debate] (2026)

BACKGROUND: Harm reduction has largely been shaped by responses to psychoactive drug use where the most urgent harms are acute. These models focus on overdose, blood-borne viruses, and rapid-onset toxicity related harms. When applied wholesale to anabolic-androgenic steroids (AAS), they obscure the distinctive pharmacology, consumer typologies, and slow-developing physiological risks that define people who use AAS.
ARGUMENT: AAS use is often chronic, patterned, and long-term. Harms are often cumulative and organ-based rather than event-based. Routes of administration carry different risk profiles, with oral formulations being more hepatotoxic and commonly falsified through mislabelling or adulteration than injectable products. Despite this, most health services position injecting as inherently higher risk, applying paradigms developed for opioids and stimulants that have not been adapted for AAS use. Using evidence across pharmacology, delivery routes, dependence trajectories, and consumer types, this paper argues for expanding harm reduction models to consider the unique needs of different populations of people who use drugs, including AAS. We draw critically on lessons from other substances as well as multiple existing approaches including structured dosing frameworks, and supply-checking infrastructures, provide practical templates for expanding adaptation to AAS.
CONCLUSIONS: To remain evidence-based, harm reduction must evolve to be contextually relevant. We have used AAS as an example of this evolution, attempting to highlight cross-substance learnings which integrate consumer-focused tools, workforce development, and peer-led support. [Author's abstract]