I agree, it seems very strange, almost as if a second tier of...

I agree, it seems very strange, almost as if a second tier of investigation was layered over the original research project - that of effectiveness of prescribed opioids in non-cancer pain.

If one was to investigate 'use of cannabis', one would think standard measures of drug use: formulation, dosage, method of administration etc. would be integral to the study.

Seems their only measure was dosage - itself categorised into "not", "less frequent" and "daily or near daily", which is pretty weak imho . We are left to assume how these unquantifiable doses were administered to participants.

Clearly this was not a clinical trial, yet it leans toward clinical trial outcomes measures. Odd. Stewart was right to point out the value a clinical study can have.

imho, this paper adds little to the literature on therapeutic effectiveness of cannabis.