[walterbell]

> The website has gotten direct critique of spreading misinformation in two papers - They draw their own conclusions from the paper.

Good to know, is that critique public?

> They claim that each of the following individual medicines works

The site text says:

> Analysis of the efficacy of early treatments for COVID-19. Treatments do not replace vaccines and other measures. All practical, effective, and safe means should be used. Elimination of COVID-19 is a race against viral evolution. No treatment, vaccine, or intervention is 100% available and effective for all current and future variants. Denying the efficacy of any method increases the risk of COVID-19 becoming endemic; and increases mortality, morbidity, and collateral damage ... We focus on the most effective/promising early treatments. There are many treatments that are helpful for late stage patients, however we do not have the resources to cover everything and currently focus on early treatment.

That text doesn't say that each of the listed therapeutics "works", it says they are presenting an analysis/summary of the efficacy of each treatment, with data from referenced studies. If there's a critique of a referenced study, or an error in the summary which is derived from references, that would be good to identify publicly.

This type of summary of public data and research papers should be on github or other public source repo (the site says the content is CC0 licensed), where issues/PRs could be used to correct errors. Maybe the site can be forked, errors removed and additional studies added. Early treatment is an important and neglected topic, which can supplement vaccines in breakthrough cases.

[rsfern]

This HN comment [0] links to a Twitter comment by an epidemiologist criticizing a related site for cherry-picking data, among other concerns.

It seems like there are several sites that all have similar layout/content and very distinctive plot styling. I think they are related, but I’m on a phone and haven’t made time for dig queries and such. Seems dodgy though.

Anyway, there’s been a feeding frenzy of questionable research since the pandemic started (well especially so anyway) so it’s more important than ever to practice informed skepticism.

[0]: https://news.ycombinator.com/item?id=28177264

[walterbell]

Thanks, this is helpful, will review.

In theory, those opposed to early treatments could publish a competing analysis of studies on early treatments. Maybe that has already been done?

On some contested topics, there's almost a need for rules of evidence, so there's an agreed upon list of papers/data for discussion. Then different stakeholders/teams can debate on their (likely biased) interpretations of a pool of evidence that is clearly bounded at any point in time, but expanding as new data/studies become available.

[rsfern]

Here is Derek Lowe on the matter back in June, with an update after the big Egyptian study was retracted: [0]. Not specifically on this website you linked, but on the body of evidence for ivermectin as a COVID therapeutic.

I was curious so I ran Whois queries on three sites: c19early.com which you linked, and ivmmeta.com and c19ivermectin.com. They all seem to be hosted through google sites, and use the same DNS privacy service. I don’t know if google sites by default uses this service or something, but just looking at the similarity of the sites I am wary.

[0]: https://blogs.sciencemag.org/pipeline/archives/2021/06/07/iv...

[walterbell]

Any idea why India is reporting success in the real world that isn't predicted by available studies? E.g. could someone design a study in Uttar Pradesh, with their huge population, to dig deeper into conflicts between the studies available for Lowe's analysis, and the claimed success in UP? There are states in India not using Ivermectin, which could be compared with the effects of UP's official state policy of making Ivermectin widely available to their population.

When experiments don't match our predictions, there is often either an error in the experiment, or an opportunity to improve the models behind the prediction. In this case, one of the experiments (UP with 200 million people) apparently has an unexpectedly good, positive, cheap outcome with minimal supply chain dependencies or safety risks.

If the observations are incorrect, that should be easy to establish.

If we are unable to explain this observed, positive, practical outcome via existing theory, that would be scientifically exciting!

[sorethescore]

India is lying and systematically undercounting their cases, hospitalizations and deaths by as much as 10 times. India does not have a robust public health system or effective record keeping. They can't even record every birth let alone keep track of deaths in a pandemic.

[belltaco]

India isn't reporting success in the real world, what evidence do you have except some shady video? Covid went down in a similar fashion in other states in India too at the same time.