This paper is very much a case of read past the abstract, especially the limitations of the study. As always it’s important for a clinician to explain the risks and current evidence when prescribing, no matter the substance. A lot of medicines have limited evidence, but they still work for some people.
Personally I use prescribed pharmacutical cannabis oils as I have much lower levels of a couple of important enzymes than most people which renders opioids mostly ineffective, even intravenous morphine as I recently found out after surgery. High CBD cannabis oil works, as does paracetamol but that’s way more dangerous.
I've had this discussion with American friends quite recently, it's very much an American English thing to not use those constructions. Certainly in British, New Zealand, and Australian English we do all the time.
> And then, from the perspective of ethics, once you know it's just placebo, you kinda shouldn't keep giving it to people, even if it helps?
That's a very big ethical question in the medical field. Placebos _do_ help, but only if people believe they will. So is it ethical to lie to a patient and give them a placebo knowing it's likely to help?
It Depends. It's going to depend on your location, how your health system works, and a bunch of luck.
Even in the most well-resourced system if your high-priority call comes in just after a bunch of other high-priority calls you may not get an ambulance in time as everyone's already helping someone else. Also in our current economic system there's a whole bunch of pressures that mean we can't base our medical care availability on the worst case, so sometimes people don't get the care they need due to lack of staff.
However I do think in a good system dispatchers would have visibility to know if an ambulance can be dispatched or retasked and how long it will take to get there. You can't make good recommendations without the information to do so.
Yes, that's why you engage with the people doing the work first and run it on a staging environment to see what would be overwritten. You test until it's working well enough to enhance the effort done by the translators.
Well, in this era Im not entirely sure the quality aspect is even considered. CEO wants AI? Then he will get it, so that the next earnings call can be bombastic!
Saving zero dollars and making the product worse is not important, only that there doesn’t seem to be a browser monopoly is.
What, you mean that US companies should ask their local branches before pushing changes in every countries? /s
This happens all the time, in every US company I know. It's as if the Americans where entirely oblivious to the fact that the rest of the world exists.
It might be more helpful to point out which guidelines it did follow. Humans are expected to read and obey these things - so presumably whoever deployed them will be aware and can demonstrate that they were followed.
It has an interesting conclusion that says more research in to CBD rather than THC is needed and cites some papers looking in to that.
reply