The field of mental health continues to struggle with the question of how best to structure its diagnostic systems. This issue is of considerable ethical importance, but the implications for public health approaches to mental health have yet to be explored in any detail. In this article I offer a preliminary treatment, drawing out several core issues while sounding a note of caution. A central strand of the debates over diagnosis has been the contrast between categorical and dimensional models, with (...) renewed attention due to recent publication of the DSM-5, launch of the RDoC, and ongoing work on the ICD-11. This dispute involves an interesting assortment of ethical and empirical considerations, many with direct relevance for public health. It has been proposed, for example, that dimensional diagnosis may be morally preferable because it can help reduce the stigma associated with mental disorder. This is a pressing concern, as preventive strategies are expanded in mental health, often operating under dimensional assumptions that target various risk factors. But this type of proposal relies upon an empirical claim and the scientific status of dimensionality remains unresolved, including its relation to stigma. I suggest, then, that the current state of the evidence does not yet warrant clear adjudication between competing frameworks, and thus any implications for public mental health remain highly provisional. More research is needed to help resolve these issues, including ethical analysis. (shrink)