I only recently found your site, it's got a lot of great analysis.
I wondered if you have looked at the more recent UK vaccine surveillance files?
Starting I think around week 36 they began reporting rates per 100K population (there are also time series of vaccination rates broken down by age).
Anyway, a strange phenomenon has arisen where case rates are higher among the vaccinated population, although hospitalization and death still appears at lower rates.
Week over week, the difference in case rates has continued to get worse for some months now.
I'm really not sure what to make of it, and wondered if you had seen it or written about it?
One possible theory is that the delta variant finds more fertile ground in vaccinated populations due to the fact that it evades the NTD antibodies produced by the vaccine.. but perhaps delta did not seem generally more adaptive (and therefore did not spread widely) in largely unvaccinated populations?
See these for more discussion of evolution of delta and vaccine evasion:
Using the data source “SARS-CoV-2 variants of concern and variants under investigation in England, Technical Briefing 17”, UK government document dated June 25th, found here https://www.gov.uk/coronavirus (use the search), looking at tables 3, 4 and 8, you can conclude three things; (a) vaccinations are effective against COVID; (b) the COVID Delta variant is about 4x less deadly than the COVID Alpha variant; (c) being old sucks. Here is more detail:
1) 50-and-overs are a hundred times less likely to die once exposed to COVID if they are vaccinated, compared with not being vaccinated. This is a combination of ~20x less likely to be infected and ~4.3x less likely to die once infected.
2) The under-50s are ~24x less likely to die of COVID if they are vaccinated than not just because they are ~20x less likely to be infected and ~20% less likely to die once infected. (But they are not very likely to die, esp. if the are young.)
3) Under 50s have a 1 in 1800 chance of being killed by COVID (Alpha variant) once they catch it.
4) 50-and-overs have a 1 in 21 chance of being killed by COVID (Alpha variant) once they catch it.
5) Under 50s have a 1 in 10,000 chance of being killed by COVID (Delta variant) once they catch it.
6) 50-and-overs have a 1 in 88 chance of being killed by COVID (Delta variant) once they catch it.
My data came from that source. Many of your statements here are not well defined incorrect, but I would know where to correct them given that you haven't shown work or explained methodology. How about you write up a blog post somewhere providing warrants for you conclusions. Then we will talk.
Hiya thanks for replying. I did not do anything complicated but you don't have to to get a good quality result. In table 4 the have cases and deaths for 50+ and <50. I just divided deaths by cases to get the death rate for each category. Its not exact because there are cases ongoing, but the error can't be too large. And table 8 tells you the effectiveness against symptomatic disease, but I see now I incorrectly used the values from table 9 not table 8 so the result is too large. Should be 10x not 20x for Alpha for example.
From table 8 of the Source: "Vaccine effectiveness against symptomatic disease" for dose 1 is 49% (Alpha variant) and 32% (Delta variant); dose 2 is 89% (Alpha) and 79% (Delta). This is in reasonable agreement with the "Case" values you calculated, if they are the same thing.
Regarding this the vaccine's "effectiveness against death", table 4 gives the number of deaths for individuals testing positive for Delta in the vaccinated <50 and >50 and also the unvaccinated <50 and >50. The quick glance at the data shows that the survival rates are essentially the same for vaccinated and unvaccinated <50s. That is not the case for the >50s, where the death rate for unvaccinated individuals (3.89%) is much higher than the death rate for vaccinated individuals (0.91%). This is in my opinion a misleading number because it averages the whole 50+ category which is a mix of very different survival rates. (As you noted in your article.)
Looking at table 4 again the data for the different doses is very odd. It shows very different survival rates immediately after the first dose (<21 days), later after the first dose (>21 days) and after the second dose. For about the same case numbers (3865 and 5546) for one dose, <21 days later and >21 days later, the number of deaths is 17 and 50. The only explanation I can think of concurs with your analysis that 21 days after the 1st dose, individuals are already protected from death, but not yet maximally? protected from catching COVID.
Nefarious has a ring to it.
Hi Michael,
I only recently found your site, it's got a lot of great analysis.
I wondered if you have looked at the more recent UK vaccine surveillance files?
Starting I think around week 36 they began reporting rates per 100K population (there are also time series of vaccination rates broken down by age).
Anyway, a strange phenomenon has arisen where case rates are higher among the vaccinated population, although hospitalization and death still appears at lower rates.
Week over week, the difference in case rates has continued to get worse for some months now.
I'm really not sure what to make of it, and wondered if you had seen it or written about it?
One possible theory is that the delta variant finds more fertile ground in vaccinated populations due to the fact that it evades the NTD antibodies produced by the vaccine.. but perhaps delta did not seem generally more adaptive (and therefore did not spread widely) in largely unvaccinated populations?
See these for more discussion of evolution of delta and vaccine evasion:
https://eugyppius.substack.com/p/the-mystery-of-the-variants
https://eugyppius.substack.com/p/preprint-the-sars-cov-2-delta-variant
Thank you so much for all your hard work!
You are keeping many people slightly more sane (by some metrics).
Using the data source “SARS-CoV-2 variants of concern and variants under investigation in England, Technical Briefing 17”, UK government document dated June 25th, found here https://www.gov.uk/coronavirus (use the search), looking at tables 3, 4 and 8, you can conclude three things; (a) vaccinations are effective against COVID; (b) the COVID Delta variant is about 4x less deadly than the COVID Alpha variant; (c) being old sucks. Here is more detail:
1) 50-and-overs are a hundred times less likely to die once exposed to COVID if they are vaccinated, compared with not being vaccinated. This is a combination of ~20x less likely to be infected and ~4.3x less likely to die once infected.
2) The under-50s are ~24x less likely to die of COVID if they are vaccinated than not just because they are ~20x less likely to be infected and ~20% less likely to die once infected. (But they are not very likely to die, esp. if the are young.)
3) Under 50s have a 1 in 1800 chance of being killed by COVID (Alpha variant) once they catch it.
4) 50-and-overs have a 1 in 21 chance of being killed by COVID (Alpha variant) once they catch it.
5) Under 50s have a 1 in 10,000 chance of being killed by COVID (Delta variant) once they catch it.
6) 50-and-overs have a 1 in 88 chance of being killed by COVID (Delta variant) once they catch it.
My data came from that source. Many of your statements here are not well defined incorrect, but I would know where to correct them given that you haven't shown work or explained methodology. How about you write up a blog post somewhere providing warrants for you conclusions. Then we will talk.
Hiya thanks for replying. I did not do anything complicated but you don't have to to get a good quality result. In table 4 the have cases and deaths for 50+ and <50. I just divided deaths by cases to get the death rate for each category. Its not exact because there are cases ongoing, but the error can't be too large. And table 8 tells you the effectiveness against symptomatic disease, but I see now I incorrectly used the values from table 9 not table 8 so the result is too large. Should be 10x not 20x for Alpha for example.
Please read my two articles before posting.
Hi, I have read them.
From table 8 of the Source: "Vaccine effectiveness against symptomatic disease" for dose 1 is 49% (Alpha variant) and 32% (Delta variant); dose 2 is 89% (Alpha) and 79% (Delta). This is in reasonable agreement with the "Case" values you calculated, if they are the same thing.
Regarding this the vaccine's "effectiveness against death", table 4 gives the number of deaths for individuals testing positive for Delta in the vaccinated <50 and >50 and also the unvaccinated <50 and >50. The quick glance at the data shows that the survival rates are essentially the same for vaccinated and unvaccinated <50s. That is not the case for the >50s, where the death rate for unvaccinated individuals (3.89%) is much higher than the death rate for vaccinated individuals (0.91%). This is in my opinion a misleading number because it averages the whole 50+ category which is a mix of very different survival rates. (As you noted in your article.)
Looking at table 4 again the data for the different doses is very odd. It shows very different survival rates immediately after the first dose (<21 days), later after the first dose (>21 days) and after the second dose. For about the same case numbers (3865 and 5546) for one dose, <21 days later and >21 days later, the number of deaths is 17 and 50. The only explanation I can think of concurs with your analysis that 21 days after the 1st dose, individuals are already protected from death, but not yet maximally? protected from catching COVID.
https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/997418/Variants_of_Concern_VOC_Technical_Briefing_17.pdf