Psychiatric Contacts Are Not Outcomes: Interpreting Evidence in Gender-Referred Youth.

It is widely agreed that, in order for research to be ethical, the risks it imposes on its participants must be justified by proportionate social value. However, there is no existing account of what proportionality between risk and social value consists in, and, relatedly, no accepted method for determining when it obtains. This paper demonstrates this gap, argues for a particular way of closing it, and sketches implications. The project is motivated by recent debates over the permissibility of controlled human infection trials (CHIs) using the novel coronavirus, that is, trials which would intentionally expose healthy volunteers to SARS‐CoV‐2. Early in the pandemic, some proposed such CHIs as a method for accelerating vaccine development [, ], leading to a robust scholarly discussion [, , ]. While early proposals were criticized on a number of grounds, the most compelling among them was that CHIs were, for various logistical reasons, very unlikely to succeed in accelerating vaccine licensure, and that they therefore lacked proportionate social value. In other words, the claim was that even if such CHIs' risks to participants were low enough to be in‐principle justifiable, their actual value was insufficient to do that justifying. Unfortunately, the undertheorized nature of the proportionality relationship left it unclear how to determine whether the claim of insufficient value was true. This paper attempts to advance our understanding of proportionality. It distinguishes two different kinds of proportionality, internal and external, and argues that it is internal proportionality that should be the primary object of research ethics review. It proposes understanding internal proportionality as a requirement that at least a certain fixed numeric ratio hold between a study's risks and benefits and argues that, so understood, the most plausibly required ratio is “1 or less.” It illustrates how, due to the vast imbalance between the numbers of people in a trial and those who stand to benefit, scientifically rigorous trials addressing widespread health conditions have a strong tendency to meet this standard, and that his can be true even when they are very unlikely to succeed. This account is defended from objections and alternatives are criticized.