Why we do not know enough yet about hydroxychloroquine and azithromycin
Posted by John T. Reed on
I said that the media was overhyping chloroquinine. Today they are hyping it about five times as much. Trump talked like a fool on this and so did Laura Ingraham.
.
1. They said it worked great on some patients. Dr. Fauci said it is ANECDOTAL evidence that means relatively few patients and no controls—same type of patients who unwittingly take placebos instead of the medicine being tested.
2. They said we know it’s not dangerous because it has been used for 75 years for malaria. The more correct version would be we know it does not hurt or rarely hurts people who do not have covid 19. We have very little data on whether there are any harmful effects on Covid 19 patients, like our having little data on the effect of thalidomide on pregnant women in the early 1960s.
3. We need to use hydroxychloroquine and azithromycin, which has shown some promise and placebos with nearly identical covid 19 patients to learn the efficacy, safety, and correct dosage.
.
Ingraham put down Dr. Fauci and others who called the current data anecdotal as being anti-cure, anti-hope, mindless bureaucrats who oppose mass instant adoption of those two medicines for stupid reasons.
.
I think Dr. Fauci needs to explain better why anecdotal evidence is insufficient to order immediate mass use of these promising-based-on-anecdotal evidence medicines. But I am not going to suggest he is some sort of malevolent dope compared to Trump or Ingraham. Ingraham considered going to medical school before she went to law school. Had she graduated from medical school, she would not now be putting Dr. Fauci down as some sort of uncaring, dopey, paper pusher.
.
There are a number of basic principles involved here. One is the LAW OF LARGE NUMBERS which says the more tests you do, the greater the probability that the actual experience will match the expected experience. In this case percentage of cures and expected experience means the percentage expected from the percentage cures in the anecdotal tests.
.
In order to draw big conclusions from relatively few observations, you have to take great care to make sure the few patients were TYPICAL of the vast population you want to administer these medicines to.
.
Another human nature principle is called PLACEBO EFFECT. That is an observed behavior in which if you tell a bunch of sick people that you are giving them a pill that will cure them, and above random percentage of them will confirm that the pill DID make them feel better even when it was a mere sugar pill that could not possibly have made them feel better.
.
Another problem with small samples that are no controlled by having a parallel test going on with placebos and another getting no treatment at all is CONFIRMATION BIAS on the part of the persons who think the treatment might work. They tend to see their theories confirmed even when they are not.
.
The encouraging data on these two medicines is at least partly lab results. In theory, that should eliminate placebo effect. But there are issues about lab procedures. Basically, the anecdotal results need to be confirmed by scientists who have no dog in the fight. That is it is not THEIR theory that is being tested or proven or disproven. When the guys who dreamed up the notion that hydroxychloroquine and azithromycin are effective covid 19 treatments are also the ones “grading the test,” you have to wonder if there is any conscious or subconscious cheating.
.
The best test is performed by neutral scientists who are only interested in producing a conclusion that will not be found to be invalid by yet another group of scientists. They want to get it RIGHT, not prove the medicine works or does not work.
.
1. They said it worked great on some patients. Dr. Fauci said it is ANECDOTAL evidence that means relatively few patients and no controls—same type of patients who unwittingly take placebos instead of the medicine being tested.
2. They said we know it’s not dangerous because it has been used for 75 years for malaria. The more correct version would be we know it does not hurt or rarely hurts people who do not have covid 19. We have very little data on whether there are any harmful effects on Covid 19 patients, like our having little data on the effect of thalidomide on pregnant women in the early 1960s.
3. We need to use hydroxychloroquine and azithromycin, which has shown some promise and placebos with nearly identical covid 19 patients to learn the efficacy, safety, and correct dosage.
.
Ingraham put down Dr. Fauci and others who called the current data anecdotal as being anti-cure, anti-hope, mindless bureaucrats who oppose mass instant adoption of those two medicines for stupid reasons.
.
I think Dr. Fauci needs to explain better why anecdotal evidence is insufficient to order immediate mass use of these promising-based-on-anecdotal evidence medicines. But I am not going to suggest he is some sort of malevolent dope compared to Trump or Ingraham. Ingraham considered going to medical school before she went to law school. Had she graduated from medical school, she would not now be putting Dr. Fauci down as some sort of uncaring, dopey, paper pusher.
.
There are a number of basic principles involved here. One is the LAW OF LARGE NUMBERS which says the more tests you do, the greater the probability that the actual experience will match the expected experience. In this case percentage of cures and expected experience means the percentage expected from the percentage cures in the anecdotal tests.
.
In order to draw big conclusions from relatively few observations, you have to take great care to make sure the few patients were TYPICAL of the vast population you want to administer these medicines to.
.
Another human nature principle is called PLACEBO EFFECT. That is an observed behavior in which if you tell a bunch of sick people that you are giving them a pill that will cure them, and above random percentage of them will confirm that the pill DID make them feel better even when it was a mere sugar pill that could not possibly have made them feel better.
.
Another problem with small samples that are no controlled by having a parallel test going on with placebos and another getting no treatment at all is CONFIRMATION BIAS on the part of the persons who think the treatment might work. They tend to see their theories confirmed even when they are not.
.
The encouraging data on these two medicines is at least partly lab results. In theory, that should eliminate placebo effect. But there are issues about lab procedures. Basically, the anecdotal results need to be confirmed by scientists who have no dog in the fight. That is it is not THEIR theory that is being tested or proven or disproven. When the guys who dreamed up the notion that hydroxychloroquine and azithromycin are effective covid 19 treatments are also the ones “grading the test,” you have to wonder if there is any conscious or subconscious cheating.
.
The best test is performed by neutral scientists who are only interested in producing a conclusion that will not be found to be invalid by yet another group of scientists. They want to get it RIGHT, not prove the medicine works or does not work.
Share this post
0 comment