4 ms·

The plural anecdote is not data. At the end of the day it comes down to statistics and reproducibility. A lot of things can happen for a lot a reasons. It’s why

The plural anecdote is not data. At the end of the day it comes down to statistics and reproducibility. A lot of things can happen for a lot a reasons. It’s why things like homeopathy still exist.

walterbell 5y ago
The site posted at the top of this thread has 800 scientific studies of various therapeutics, https://c19early.com/faq.html https://c19early.com/faq.html > Whether it be treatments or interventions, the plethora of seemingly similar quality research supporting contradictory positions, and the use of digital echo chambers, facilitates widespread confirmation bias - evidence supporting any initial position can be easily found. Self-serving bias and cognitive dissonance further make it challenging to alter positions, especially for those taking very public and strong positions with serious implications. We felt that attempts to organize and make the research and data easier to navigate, highlighting areas of applicability, and analyzing limitations, could potentially be beneficial. > What is the search strategy for papers? All significant original contributions related to the use of the treatments we cover, including searches of the typical sources augmented by contributions from the community. Covering all research is important because it is easy to choose a search strategy that results in a subset of papers reporting a desired conclusion. This is especially so if one takes the conclusions reported in papers at face value without examining the actual data, methods, or regions of applicability
- lolspace 5y ago
  That website is fake: - It is impossible to contact the owners of the website - The website has gotten direct critique of spreading misinformation in two papers - They draw their own conclusions from the paper. - A lot of the papers in its meta-analysis doesn’t have control groups or randomized trials. Some studies only compare ivermectin+X vs ivermectin+Y - The total sample size of all studies are only 2000 people. But let's ignore all these red flags and focus on this. They claim that each of the following individual medicines works, Fluvoxamine, Proxalutamide, Iota-carrageenan, Molnupiravir, Quercetin, Povidone-Iodine, Curcumin, Casirivimab, Sotrovimab, Bamlanivimab, Nitazoxanide, Budesonide, Zinc, Bromhexine, Colchicine, Vitamin D, Aspirin, Favipiravir, Hydroxychloroquine, Remdesivir, and Vitamin C. I find it unlikely that ALL those claims are true which in turn makes me question their scientific approach.
  - walterbell 5y ago
    > A lot of the papers in its meta-analysis doesn’t have control groups or randomized trials. Some studies only compare ivermectin+X vs ivermectin+Y If you have the time to share an illustrative example paper, that would be helpful. > I find it unlikely that ALL those claims are true which in turn makes me question their scientific approach. Response posted here: https://news.ycombinator.com/edit?id=28335175 https://news.ycombinator.com/edit?id=28335175

The plural anecdote is not data. At the end of the day it comes down to statistics and reproducibility. A lot of things can happen for a lot a reasons. It’s why things like homeopathy still exist.

walterbell 5y ago

The site posted at the top of this thread has 800 scientific studies of various therapeutics, https://c19early.com/faq.html https://c19early.com/faq.html > Whether it be treatments or interventions, the plethora of seemingly similar quality research supporting contradictory positions, and the use of digital echo chambers, facilitates widespread confirmation bias - evidence supporting any initial position can be easily found. Self-serving bias and cognitive dissonance further make it challenging to alter positions, especially for those taking very public and strong positions with serious implications. We felt that attempts to organize and make the research and data easier to navigate, highlighting areas of applicability, and analyzing limitations, could potentially be beneficial. > What is the search strategy for papers? All significant original contributions related to the use of the treatments we cover, including searches of the typical sources augmented by contributions from the community. Covering all research is important because it is easy to choose a search strategy that results in a subset of papers reporting a desired conclusion. This is especially so if one takes the conclusions reported in papers at face value without examining the actual data, methods, or regions of applicability

lolspace 5y ago
That website is fake: - It is impossible to contact the owners of the website - The website has gotten direct critique of spreading misinformation in two papers - They draw their own conclusions from the paper. - A lot of the papers in its meta-analysis doesn’t have control groups or randomized trials. Some studies only compare ivermectin+X vs ivermectin+Y - The total sample size of all studies are only 2000 people. But let's ignore all these red flags and focus on this. They claim that each of the following individual medicines works, Fluvoxamine, Proxalutamide, Iota-carrageenan, Molnupiravir, Quercetin, Povidone-Iodine, Curcumin, Casirivimab, Sotrovimab, Bamlanivimab, Nitazoxanide, Budesonide, Zinc, Bromhexine, Colchicine, Vitamin D, Aspirin, Favipiravir, Hydroxychloroquine, Remdesivir, and Vitamin C. I find it unlikely that ALL those claims are true which in turn makes me question their scientific approach.
- walterbell 5y ago
  > A lot of the papers in its meta-analysis doesn’t have control groups or randomized trials. Some studies only compare ivermectin+X vs ivermectin+Y If you have the time to share an illustrative example paper, that would be helpful. > I find it unlikely that ALL those claims are true which in turn makes me question their scientific approach. Response posted here: https://news.ycombinator.com/edit?id=28335175 https://news.ycombinator.com/edit?id=28335175