But if you had read any of the studies that were done about second hand smoke they were done that way. Just like the primary studies done on smoking. They scoured death certificates and even if a person had died of pneumonia or heart disease they were thrown into the smoking caused it category. You have to have been paying attention to the longitudinal studies
The problem there is not the proximate cause of death - pneumonia, heart disease - but the attribution of those proximate causes to second-hand smoke. In the current study, opioid overdose is the clear-cut proximate cause of death, which both the article and the research paper are careful to NOT say was caused by availability of legal medical marijuana.