It would be interesting to see numbers on how the proportion of overdoses has changed based on consumption medium: have prevention sites led to a significant reduction in overdose deaths for the mediums they permit, or is smoking just more dangerous?