Hacker News Reader

Comment by prostheticvamp | Hacker News Reader

Medical doctor and statistician weighing in just to say: I agree fully with what you wrote. No miracles observed.

If the control group still is positive for corona after 7 days in 90% of cases and we anticipate the intervention to get that number to 50%, with an alpha of 0.05 and a size of 24 in the intervention group, the study has a power of 0.957 and answers the question whether chloroquine stops viral shedding by day 7. This isn't cardiology where we need 50000 patients to find the significance.

I don't mean and disrespect to you as a physician, but there appear to be many experts who disagree with you to the point of acting out their disagreement at scale. Chloroquine has become a standard part of treatment for this disease in several countries, and its continued use seems to imply its efficacy.

>there appear to be many experts who disagree with you to the point of acting out their disagreement at scale. Chloroquine has become a standard part of treatment for this disease in several countries, and its continued use seems to imply its efficacy.

The fact that experts agree that this drug should be used doesn't mean they agree that it is miraculous. And this certainly doesn't suggest they disagree with the parent MD/statistician.

It seems like such a strange nit to pick here over my use of the term miraculous. I don’t know what else you would call the results of the study I linked (supported by the reaction of many national health orgs). Yeah, maybe this isn’t the cure to this disease, and of course we need more testing, but the result as reported are incredibly hopeful. It seems to have hit a nerve with some people that I used the term “miraculous”, but I will continue to defend that term. We are looking at a virus that has disrupted the global order. Here is a study claiming incredible efficacy. That isn’t a miracle in the divine sense, but if definitely reports as being extremely helpful during a dire need.

You were wrong when you presented alleged evidence/argument that experts must necessarily disagree with the MD/statistician, above.

If doctors are using this, and not seeing an effect, why do they continue using it? Doctors are humans, and prone to irrationality of course, but they're also scientists.

Maybe my faith in the collective ability of the world medical professionals to evaluate their treatment techniques is misplaced, but I still have faith that these people are acting in accordance with their scientific training.

Maybe that's wrong, and I shouldn't be trusting these people. But if I'm not trusting doctors (and especially infectious disease experts who are writing papers about their findings), then who should I trust?

I think a lot of this distrust in institutions and doctors is going to get us into a lot of trouble. I wish people would start listening to these orgs.

Your post does not address my point. That's okay, if you simply wanted to say the things you said. There are things you've said that I agree with, and things which I strongly disagree with. In a conversation, I like to be careful about going down too many side tangents if certain fundamentals are not already mutually understood.

It seems like you disagree with my implication that use of something (in this case, chloroquine to treat covid-19, the use of which seems to be increasing daily) implies its efficacy.

Why? I'm modeling the doctors and health officials who are making these decisions as agents which are presumably continuously evaluating their decisions. So basically:

A lot of experts are taking this action -> assume the action has a desirable outcome -> believe what the experts are saying (in this case both their words and their actions seem to agree with each other, and point to the idea that chloroquine is an effective treatment of covid-19).

I am genuinely at a loss as to why this stance (my stance) is anything other than completely obvious.

No. Here, as in several places, you make unwarranted assumptions about other's thought processes. Remember that "efficacy" is a matter of degree. I try not to spend too much time and effort on strawmen.

I'm trying to understand you here. Is the assumption that doctors are taking actions based on previous experience with those sorts of actions unwarranted in your opinion? Or a strawman?

I'm sorry if but I'm having a difficult time figuring out what I have said that you disagreed with.

You seem sincere, and unlike prior comments to be approaching this in good faith. Maybe you even welcome criticism as a means of improving your thought process. I don't mean to be cryptic, and certainly don't mean to be rude, I just try to conserve energy and not get caught up in non-constructive arguments.

You don't seem to be taking people's statements at face value, but rather projecting additional meaning onto them, almost as if you conceive of fewer possible positions than there actually are, and pigeonhole us.

Consider my prior two statements in this thread which start with "the fact" and "you were". Do you agree that these statements were correct, on their own terms, in their own context, and independent of additional complexities involving degree if efficacy or various arguments for or against use, testing, or hopefulness?

Also, do you see that the people who disagree with your framing are not 'picking nits' nor are they having 'struck nerves'? Can you see other reasons why they might respond as they did?

Its use in the presence of a plausible mechanism, and in the absence of any good supporting data, does not indicate efficacy. It indicates that when a patient is circling the drain you throw things at the wall that could plausibly have some benefit.

Haha. Continued use seems to imply it's efficacy. This might be the most illogical post on Hacker News in some time.

I think we could have a whole website of things that were continued to be used without a shred of evidence of effectiveness.

We could also have a whole website of things people strongly believed were effective that RCTs absolutely demolished.

RCT? Sorry don't know that abbreviation.

Randomized controlled trial

Thanks

I don't think he said no one is using (or should use) it, just that the results aren't "miraculous".

Miracle (noun): an extraordinary and welcome event that is not explicable by natural or scientific laws and is therefore attributed to a divine agency.

Reply on news.ycombinator.com