4 ms·

Looking at the two papers (original, and updated) they only removed the Elgazzer paper but not the contested Shakhsi Niaee also mentioned in the Nature article.

Looking at the two papers (original, and updated) they only removed the Elgazzer paper but not the contested Shakhsi Niaee also mentioned in the Nature article. This is noteworthy because it's also the only other RCT to show strong effect on mortality (see Fig3 in the two papers). Original Marik, Kory paper Fig3 is here: https://pubmed.ncbi.nlm.nih.gov/34375047/#&gid=article-figures&pid=figure-3-uid-2 https://pubmed.ncbi.nlm.nih.gov/34375047/#&gid=article-figur... Also, the Nature article is concerned that the underlying data isn't sound due to bad randomization and thus naive meta-analysis is inherently flawed in these cases.

nradov 5y ago
Sure that's a good point. But is mortality the appropriate clinical trial endpoint? Mortality rates are already low when following the NIH treatment guidelines, so it would take a large study population to discern any statistically significant effects. https://www.covid19treatmentguidelines.nih.gov/ https://www.covid19treatmentguidelines.nih.gov/ In addition to mortality it probably makes sense to look at other endpoints such as patient reported symptoms, RT PCR test cycle count, and time to hospital discharge. Those should allow us to tell if there is a real effect (or not) with smaller subject groups.
- scoopertrooper 5y ago
  Case fatality has been declining for sure, but it's still at about 3%. So it shouldn't be that hard to find a signal. https://www.abs.gov.au/articles/covid-19-mortality-0 https://www.abs.gov.au/articles/covid-19-mortality-0 https://www.ajtmh.org/view/journals/tpmd/104/6/article-p2176.xml https://www.ajtmh.org/view/journals/tpmd/104/6/article-p2176...
  - alvah 5y ago
    The reported CFR may be in that region, but most credible sources estimate a much lower real CFR, due both to many undetected / asymptomatic infections, and confusion between deaths with & deaths caused by Covid-19. For a statistical reporting agency such as the ABS to claim to know the actual CFR with any level of certainty seems careless at the very least.
    - jjeaff 5y ago
      Case fatality ratio is only counting diagnosed cases. Unknown, asymptomatic cases are irrelevant.
    - robbiep 5y ago
      The term you are looking for is IFR
    - scoopertrooper 5y ago
      The ABS clearly sets out their method for calculating CFR, which is covid deaths divided by confirmed infections. Yes, that is only a proxy for the overall all infection fatality rate. However, given the contact tracing, testing, and medical systems have been quite robust in Australia, it shouldn't be too far off the mark. Furthermore, the 3% figure only applies to the population as a whole. According to the ABS, CFR climbs to over 30% for people 80 and over. Similarly, a recent American study found it to be about 21% [1]. So, back to the original point, it should be very easy to find a mortality signal in a RCT, which included that cohort. [1] https://www.medrxiv.org/content/10.1101/2021.04.09.21255193v1.full https://www.medrxiv.org/content/10.1101/2021.04.09.21255193v...
      - nradov 5y ago
        The actual infection fatality ratio is nowhere near 3%. The US CDC estimated it at 0.6%. https://www.cdc.gov/coronavirus/2019-ncov/cases-updates/burden.html https://www.cdc.gov/coronavirus/2019-ncov/cases-updates/burd...

nradov 5y ago

Sure that's a good point. But is mortality the appropriate clinical trial endpoint? Mortality rates are already low when following the NIH treatment guidelines, so it would take a large study population to discern any statistically significant effects. https://www.covid19treatmentguidelines.nih.gov/ https://www.covid19treatmentguidelines.nih.gov/ In addition to mortality it probably makes sense to look at other endpoints such as patient reported symptoms, RT PCR test cycle count, and time to hospital discharge. Those should allow us to tell if there is a real effect (or not) with smaller subject groups.

scoopertrooper 5y ago
Case fatality has been declining for sure, but it's still at about 3%. So it shouldn't be that hard to find a signal. https://www.abs.gov.au/articles/covid-19-mortality-0 https://www.abs.gov.au/articles/covid-19-mortality-0 https://www.ajtmh.org/view/journals/tpmd/104/6/article-p2176.xml https://www.ajtmh.org/view/journals/tpmd/104/6/article-p2176...
- alvah 5y ago
  The reported CFR may be in that region, but most credible sources estimate a much lower real CFR, due both to many undetected / asymptomatic infections, and confusion between deaths with & deaths caused by Covid-19. For a statistical reporting agency such as the ABS to claim to know the actual CFR with any level of certainty seems careless at the very least.
  - jjeaff 5y ago
    Case fatality ratio is only counting diagnosed cases. Unknown, asymptomatic cases are irrelevant.
  - robbiep 5y ago
    The term you are looking for is IFR
  - scoopertrooper 5y ago
    The ABS clearly sets out their method for calculating CFR, which is covid deaths divided by confirmed infections. Yes, that is only a proxy for the overall all infection fatality rate. However, given the contact tracing, testing, and medical systems have been quite robust in Australia, it shouldn't be too far off the mark. Furthermore, the 3% figure only applies to the population as a whole. According to the ABS, CFR climbs to over 30% for people 80 and over. Similarly, a recent American study found it to be about 21% [1]. So, back to the original point, it should be very easy to find a mortality signal in a RCT, which included that cohort. [1] https://www.medrxiv.org/content/10.1101/2021.04.09.21255193v1.full https://www.medrxiv.org/content/10.1101/2021.04.09.21255193v...
    - nradov 5y ago
      The actual infection fatality ratio is nowhere near 3%. The US CDC estimated it at 0.6%. https://www.cdc.gov/coronavirus/2019-ncov/cases-updates/burden.html https://www.cdc.gov/coronavirus/2019-ncov/cases-updates/burd...