[groovy2shoes]

Even still, it's rather difficult to accidentally fatally overdose on just prescription opioids. The majority of prescription opioid-related deaths involve a mixture with some other depressant -- typically, as you say, alcohol or benzos.

Heroin is a different story, since there's no telling how potent a particular batch will be until you do some -- if it winds up being significantly stronger than what you're used to, it can be easy to do way too much. You also have various adulterants to take into account, including fentanyl which is often added with "worthless" adulterants as a cheap way to keep the qualitative potency of the effects up while significantly cutting the product. The thing about heroin is that you really don't know what you're getting -- this is a very different story when you're getting pills, unless you're happening upon counterfeits.

[martinald]

Other big problem is people going to rehab/getting clean, then relapsing after a while. Tolerance is much lower, but out of habit they are likely to do a dose similar to before.

This would be equally as possible with oxycotin.

[oldmanjay]

It would be less likely with either if addicts didn't have to hide the addiction for fear of authorities

[rhizome]

In another world street-drug users might have cheap and easily available chromatography or reagent tests.

[grkvlt]

Sure, but the reagent tests are really just qualitative, there's no good way of telling that the heroin I just bought from the nice man on the street corner is twice as pure as last weeks. I suppose there may be some tests that can tell if there is fentanyl present, which might help...?