I think for the comparison it would be better to use what percentage of users are killed. For example even though fentanyl kills something like 50% fewer people

I think for the comparison it would be better to use what percentage of users are killed. For example even though fentanyl kills something like 50% fewer people than alcohol, alcohol probably has several orders of magnitude more users. If fentanyl was as common I suspect it would kill more people than alcohol