Iranian clinicians inform a reversible pilot proposal, not proven policy…

Global debate on cannabis policy has been dominated by jurisdictions moving toward legalization or regulated markets, leaving prohibition contexts analytically underserved. Iran maintains full criminalization of cannabis alongside rising use among young people, no approved pharmacotherapy for cannabis use disorder (CUD) - although off-label options are sometimes used, no medication is approved for CUD in Iran - and a treatment system that clinicians themselves describe as fragmented. Drawing on the first qualitative study of Iranian psychiatrists' views on the demedicalization of CUD, this commentary translates those findings into a testable policy hypothesis for Iranian health authorities: a phased pilot model of conditional demedicalization. By 'conditional demedicalization' we mean a partial and reversible withdrawal of the medical frame from mild-to-moderate CUD, coupled with de-penalization of the person who uses, expansion of psychosocial services, and pre-registered stopping rules that allow the reform to be paused if harms emerge. The model comprises five elements: reframing mild-to-moderate CUD as a psychosocial rather than a primarily biomedical problem while preserving medical pathways for severe and comorbid presentations; incrementally substituting therapeutic-supportive referral for criminal sanction against people who use cannabis, informed by the Portuguese dissuasion experience but adapted to existing Iranian harm-reduction infrastructure; redirecting resources from pharmacological management toward evidence-based psychosocial services; an evidence-based prevention shield for adolescents as a precondition of any reclassification; and an intersectoral governance mechanism with explicit monitoring indicators and pre-registered stopping rules. We argue that, in a prohibition context, conditional demedicalization is best understood not as a retreat from care but as a testable harm-reduction hypothesis that, if pilot data support it, could lower barriers to help-seeking while guarding against premature normalization.