> How did you come to the conclusions that, life for life, a drug user's death is any better than someone getting shot in Afganistan?

Because getting shot in Afghanistan (especially during a war) is a random event, and the risk is spread evenly among society. The risk of drug overdose is concentrated. You don't accidentally trip on heroine, it takes a set of (largely predictable) circumstances to get there.

Society correctly assesses this risk, which is why solving the opioid epidemic polls a lot less favorably than policing and defense. The idea that society (writ large) cares more about preventing genocide and war than preventing isolated drug deaths isn't really that controversial.