But see that is exactly what I am talking about. You cannot — cannot — attribute a change in overdose rate to state policy without examining and controlling for the factors that we know influence overdoses: personal income, homelessness, etc. This article completely fails to examine whether Oregon's changes can be due to a shift in the income among its population.
I'm pretty sure being on the streets as a drug addict also causes loss of income, so you really can't take that into account without taking in cyclic effects.
