I can't find a firm answer to the question "Among the homeless, what fraction have opioid use disorder; a history that contains a legitimate medical visit conne

I can't find a firm answer to the question "Among the homeless, what fraction have opioid use disorder; a history that contains a legitimate medical visit connected to the prescription of opioids; and no history of substance abuse prior to that visit?" We'd need a Danish style medical records system to answer queries like that, or you can study it with VA data. Opioid use disorder is crazy common, nearly 5% of Americans (https://www.nih.gov/news-events/rates-nonmedical-prescription-opioid-use-opioid-use-disorder-double-10-years); https://www.nih.gov/news-events/rates-nonmedical-prescriptio... but also 9% of the population uses cannabis (https://www.cdc.gov/drugoverdose/deaths/index.html https://www.cdc.gov/drugoverdose/deaths/index.html). It's unclear how many overdose deaths happen to homeless people every year generally, and you can conclude that since homelessness is rare, so too will be "homeless given overdose death." However, the flip side - overdose death as a fraction of all deaths given homeless - can be 17% in one study (https://drugfree.org/drug-and-alcohol-news/more-homeless-die-from-drug-overdoses-than-from-aids/ https://drugfree.org/drug-and-alcohol-news/more-homeless-die...). So overall, I think you're right. Opioid disorder is so much more common than homelessness generally, homelessness is not going to possibly explain opioid use disorder in the general case. Of course it can interfere with treatment. However, when so many people use cannabis, and I haven't even investigated alcohol, the right idea is that substance abuse is so prevalent in America, you'd need more serious interventions earlier on. The issue that can be acted by just medical professionals, which is the prescription of painkillers due to legitimate medical visits that lead to substance abuse, I think there is action on this but I'm not sure what it is.