@Seethlaw
"So the risk assessment must be conducted on transwomen in general. It's not our decision: it's the law and ethics both."
I think there is a crucial distinction between how a scientific risk assessment is calculated and how a legal policy is implemented.
Even if a legal or ethical framework dictates that a government cannot discriminate based on medical status when applying a policy, that does not change the laws of physics, biology, or statistics.
A statistician calculating literal physical risk must separate the cohorts based on the variables that actually cause the risk (like testosterone levels).
If a researcher artificially lumps them together because of a legal policy, the resulting data is scientifically meaningless for determining the actual risk profile of the medically transitioned group.
If the law forces policymakers to lump everyone into one group, then the resulting policy is being built strictly on legal compliance and ethical categories, not on the actual, isolated physical risk of the transitioned cohort.
Which shows the essay's point:
The author was asking a statistical question about physical harm reduction.
By answering that we are legally forced to ignore the specific statistical variables in favor of a broad legal category, you are confirming that the boundary being drawn isn't actually about the actuarial risk data, but about a categorical rule.