3 ms·

It's interesting to me how different people come to different conclusions even on technical issues like "is this meta analysis sound after removing that study".

It's interesting to me how different people come to different conclusions even on technical issues like "is this meta analysis sound after removing that study". Here's someone saying the exact opposite: https://twitter.com/GidMK/status/1416290819565244424?s=20 https://twitter.com/GidMK/status/1416290819565244424?s=20 I'm no statistician, and am left rather unsure what to think.

bb88 5y ago
Interesting. He is an epidemiologist, and it looks like two papers should not have been added to the meta-analysis. He writes more about it here. The Elgazzar paper were cited in 30 other papers. https://gidmk.medium.com/is-ivermectin-for-covid-19-based-on-fraudulent-research-5cc079278602 https://gidmk.medium.com/is-ivermectin-for-covid-19-based-on...
- MilnerRoute 5y ago
  The Los Angeles Times had a Pulitzer-prize winning writer explore one of the recent meta-analyses - who found this key detail: All the members of the research team are associated with a British pro-invermectin organization and the money for the research was raised by a Gofundme campaign headlined "Help us get a life-saving drug approved for COVID-19." https://www.msn.com/en-us/health/medical/column-ivermectin-another-bogus-covid-treatment-becomes-a-darling-of-conspiracy-mongers/ar-AAMr54U?ocid=BingNewsSearch https://www.msn.com/en-us/health/medical/column-ivermectin-a... The Times also points out that Elgazzar's dubious data "was not peer-reviewed before publication, so it probably should never have been incorporated in the meta-analysis at all."
  - native_samples 5y ago
    Peer reviews do not involve reviews of actual collected data, just a review of the paper that draws conclusions from it. I don't know if there's such a thing as peer reviewed data in the medical world. If there is, it can't be a consistent standard, as a lot of studies are published despite appearing to have made up data: https://blogs.bmj.com/bmj/2021/07/05/time-to-assume-that-health-research-is-fraudulent-until-proved-otherwise/ https://blogs.bmj.com/bmj/2021/07/05/time-to-assume-that-hea...
lettergram 5y ago
In that analysis, he appears to remove multiple papers (2?) not just the single paper. The problem with all research at the moment is it's moving so fast and it's not peer reviewed. Luckily, Ivermectin carries very little risks, is well studied and has minimal drug interactions (and is cheap @ ~$1 per pill). https://www.webmd.com/drugs/2/drug-1122/ivermectin-oral/details/list-sideeffects https://www.webmd.com/drugs/2/drug-1122/ivermectin-oral/deta... For anyone interested, there's a database of papers and results: https://c19early.com/ https://c19early.com/ I recommend actually diving into the papers, many are not very good, but that's the nature of the research all around (drugs, vaccine, covid19, etc) at the moment.
- MilnerRoute 5y ago
  There's another analysis (and a better explanation) on this blog - including a section which looks solely at what happens "If you remove the Elgazzar paper from their model..." https://gidmk.medium.com/is-ivermectin-for-covid-19-based-on-fraudulent-research-5cc079278602 https://gidmk.medium.com/is-ivermectin-for-covid-19-based-on... "There’s no benefit seen whatsoever for people who have severe COVID-19, and the confidence intervals for people with mild and moderate disease become extremely wide." Later there's another section in the blog post - but it's not that he removes a second paper. Since it's supposed to be a meta-analysis of all the available studies, "If you include another study that was published after the Bryant meta-analysis came out, which found no benefit for ivermectin on death, the benefits seen in the model entirely disappear."

bb88 5y ago

Interesting. He is an epidemiologist, and it looks like two papers should not have been added to the meta-analysis. He writes more about it here. The Elgazzar paper were cited in 30 other papers. https://gidmk.medium.com/is-ivermectin-for-covid-19-based-on-fraudulent-research-5cc079278602 https://gidmk.medium.com/is-ivermectin-for-covid-19-based-on...

MilnerRoute 5y ago
The Los Angeles Times had a Pulitzer-prize winning writer explore one of the recent meta-analyses - who found this key detail: All the members of the research team are associated with a British pro-invermectin organization and the money for the research was raised by a Gofundme campaign headlined "Help us get a life-saving drug approved for COVID-19." https://www.msn.com/en-us/health/medical/column-ivermectin-another-bogus-covid-treatment-becomes-a-darling-of-conspiracy-mongers/ar-AAMr54U?ocid=BingNewsSearch https://www.msn.com/en-us/health/medical/column-ivermectin-a... The Times also points out that Elgazzar's dubious data "was not peer-reviewed before publication, so it probably should never have been incorporated in the meta-analysis at all."
- native_samples 5y ago
  Peer reviews do not involve reviews of actual collected data, just a review of the paper that draws conclusions from it. I don't know if there's such a thing as peer reviewed data in the medical world. If there is, it can't be a consistent standard, as a lot of studies are published despite appearing to have made up data: https://blogs.bmj.com/bmj/2021/07/05/time-to-assume-that-health-research-is-fraudulent-until-proved-otherwise/ https://blogs.bmj.com/bmj/2021/07/05/time-to-assume-that-hea...

lettergram 5y ago

In that analysis, he appears to remove multiple papers (2?) not just the single paper. The problem with all research at the moment is it's moving so fast and it's not peer reviewed. Luckily, Ivermectin carries very little risks, is well studied and has minimal drug interactions (and is cheap @ ~$1 per pill). https://www.webmd.com/drugs/2/drug-1122/ivermectin-oral/details/list-sideeffects https://www.webmd.com/drugs/2/drug-1122/ivermectin-oral/deta... For anyone interested, there's a database of papers and results: https://c19early.com/ https://c19early.com/ I recommend actually diving into the papers, many are not very good, but that's the nature of the research all around (drugs, vaccine, covid19, etc) at the moment.

MilnerRoute 5y ago
There's another analysis (and a better explanation) on this blog - including a section which looks solely at what happens "If you remove the Elgazzar paper from their model..." https://gidmk.medium.com/is-ivermectin-for-covid-19-based-on-fraudulent-research-5cc079278602 https://gidmk.medium.com/is-ivermectin-for-covid-19-based-on... "There’s no benefit seen whatsoever for people who have severe COVID-19, and the confidence intervals for people with mild and moderate disease become extremely wide." Later there's another section in the blog post - but it's not that he removes a second paper. Since it's supposed to be a meta-analysis of all the available studies, "If you include another study that was published after the Bryant meta-analysis came out, which found no benefit for ivermectin on death, the benefits seen in the model entirely disappear."