Given the current legal environment around marijuana use (I'm not sure what it's like in the UK), I doubt you'd be able to adequately control for other confound

Given the current legal environment around marijuana use (I'm not sure what it's like in the UK), I doubt you'd be able to adequately control for other confounding variables in a study like this. For example, is there a skew in demographics between users and non users? Socio-economic status? Other health indicators? Diet? Pre-existing hypertension? (Were people self-medicating?) Lifestyle? What about delivery? Is this a smoking-in-general related effect or would other forms of consumption also have the same effect? 1200 people is a pretty large study, but from this writeup, it isn't clear that they would be able to completely single out cannabis as the sole factor. I'd have to read the fully paper to know more. That said -- it is a worrying concern and certainly a good enough reason to continue studying this potential effect. Hopefully they will be able to continue with lab based studies to try and find an underlying mechanism.