So, in our view, a harm reduction programme or policy is one in which (1) the primary goal is to reduce net health, social and/or economic harm without necessarily seeking to reduce use and (2) it can be directly demonstrated, against broadly agreed criteria, that net harm across these dimensions has been reduced, rather than claiming or inferring that harm has been reduced from changes in other indices. This definition requires that to claim the mantle of harm reduction, a policy or programme must provide more than simple intention or self-evident worth. It requires a demonstration that harm reduction has occurred in an overall sense. The corollary is that policies and programmes that aim to reduce indirect or proxy outcome measures, such as risk, must illustrate how reduction in these is associated with a reduction of harm. Furthermore, those who claim that use reduction constitutes harm reduction need to establish that such an approach does actually result in an overall reduction in harm.