The article mentioned that there was a specific type of blood clot of concern called CVST. This was new to me. It says there were 6 incidents reported to VAERS, with 9 million doses administered. That’s a rate of 0.067 cases per 100,000 vaccine recipients. I found this paper https://iovs.arvojournals.org/article.aspx?articleid=2690377 that surveys the US population and finds an adjusted rate per year of 1.16/100K for men and 1.78/100K for women. (They control for age—I’m not sure if this helps or hurts the comparison to the rate of J&J clots). The mean afflicted person was 40 years old, so this rate isn’t being pumped up by elderly people. Let’s shoddily combine those gendered rates by eye and say it’s 1.4/100K CVST cases per year. The J&J had only been being administered for a little bit when the 6 cases were reported, so let’s call the 0.067/100K rate the CVST rate for one month. Multiplying by 12 to match the yearly rate gives a rate of 0.8/100K.
So if we assume the cases reported to VAERS are all the cases, then the rate for J&J recipients is lower than the base population rate. VAERS is under-reported, however, and I’m not sure by how much. If assuming only half of people with this fairly serious clot report, we get a rate of 1.6/100K for J&J receivers—about the same as the population rate. But maybe VAERS is more under-reported than that, and a factor of 10 is more fair? Then it’s 8/100K CVSTs per person-year for J&J recipients, which is higher than the base rate. I’m not sure about the right factor here.
A different useful cost-benefit analysis would compare the rate of blood clots in people who get COVID to the rate of blood clots with people who get J&J. Blood clots (not just CVST though) in hospitalized COVID patients is something like 1 in 5 (!): https://www.thelancet.com/journals/eclinm/article/PIIS2589-5370(20)30383-7/fulltext. Most people who catch COVID, especially those who have been infected before, will not be hospitalized, so this factor as it applies to your case is much much lower than 1 in 5, but is it as low as the 8/100K number above? I’d guess not, but I haven’t done the calculations to back it up.
Sorry Lithuania has those wacky combination of rules! Sounds like a difficult situation.
But maybe VAERS is more under-reported than that, and a factor of 10 is more fair?
It may be a negative slope: the more serious a side effect is, the more it gets reported. Going by that, a factor of two or three for blood clots sounds pessimistic enough. Then we’d still be somewhere around the base rate.
This study estimates the risk of cerebral venous thrombosis (CVT) and portal vein thrombosis (PVT) following Covid-19 diagnosis vs vaccination (mRNA).
Excluding those with prior history of CVT or PVT, we get 3.53/100K and 17.5/100K cases, respectively. This includes both hospitalized and non-hospitalized patients.
Also, for CVT
Among the 23 events, 7 were observed in patients under the age of 30, 4 between 30 and 39, 2 between 40 and 49, 3 between 50 and 59, 2 between 60 and 69, and 5 between 70 and 79.
I am afraid if we only look at the relevant group (e.g. under 30, hon-hospitalized), we will end up with too small a sample to draw any conclusions, so I’m not going to do so. But excluding those with prior history seems to be a good choice.
Supposing nobody gets both CVT and PVT, we end up with a total of 21.03/100K cases. This is still 2.6 times more than your worst-case scenario. Of course, the probabilities are not adjusted for those recovered from Covid-19 and having natural immunity, in which case the risk probably goes down drastically.
The study itself concludes you’re about 10 times more likely to get thrombosis from Covid-19 than from Pfizer/Moderna. I wonder how Jannsen compares to that.
Side note: my mother had a contact with an infected individual. Coincidentally, she tested for antibodies soon after that, and found out they shot up quite drastically (about 7-fold) from almost not existing to a decent level. This gets me thinking just how pointless it is to race after constantly high levels. Haven’t we known for a good deal of time that the immune system produces what’s necessary at the moment?
The article mentioned that there was a specific type of blood clot of concern called CVST. This was new to me. It says there were 6 incidents reported to VAERS, with 9 million doses administered. That’s a rate of 0.067 cases per 100,000 vaccine recipients. I found this paper https://iovs.arvojournals.org/article.aspx?articleid=2690377 that surveys the US population and finds an adjusted rate per year of 1.16/100K for men and 1.78/100K for women. (They control for age—I’m not sure if this helps or hurts the comparison to the rate of J&J clots). The mean afflicted person was 40 years old, so this rate isn’t being pumped up by elderly people. Let’s shoddily combine those gendered rates by eye and say it’s 1.4/100K CVST cases per year. The J&J had only been being administered for a little bit when the 6 cases were reported, so let’s call the 0.067/100K rate the CVST rate for one month. Multiplying by 12 to match the yearly rate gives a rate of 0.8/100K.
So if we assume the cases reported to VAERS are all the cases, then the rate for J&J recipients is lower than the base population rate. VAERS is under-reported, however, and I’m not sure by how much. If assuming only half of people with this fairly serious clot report, we get a rate of 1.6/100K for J&J receivers—about the same as the population rate. But maybe VAERS is more under-reported than that, and a factor of 10 is more fair? Then it’s 8/100K CVSTs per person-year for J&J recipients, which is higher than the base rate. I’m not sure about the right factor here.
A different useful cost-benefit analysis would compare the rate of blood clots in people who get COVID to the rate of blood clots with people who get J&J. Blood clots (not just CVST though) in hospitalized COVID patients is something like 1 in 5 (!): https://www.thelancet.com/journals/eclinm/article/PIIS2589-5370(20)30383-7/fulltext. Most people who catch COVID, especially those who have been infected before, will not be hospitalized, so this factor as it applies to your case is much much lower than 1 in 5, but is it as low as the 8/100K number above? I’d guess not, but I haven’t done the calculations to back it up.
Sorry Lithuania has those wacky combination of rules! Sounds like a difficult situation.
Hm,
It may be a negative slope: the more serious a side effect is, the more it gets reported. Going by that, a factor of two or three for blood clots sounds pessimistic enough. Then we’d still be somewhere around the base rate.
This study estimates the risk of cerebral venous thrombosis (CVT) and portal vein thrombosis (PVT) following Covid-19 diagnosis vs vaccination (mRNA).
Excluding those with prior history of CVT or PVT, we get 3.53/100K and 17.5/100K cases, respectively. This includes both hospitalized and non-hospitalized patients.
Also, for CVT
I am afraid if we only look at the relevant group (e.g. under 30, hon-hospitalized), we will end up with too small a sample to draw any conclusions, so I’m not going to do so. But excluding those with prior history seems to be a good choice.
Supposing nobody gets both CVT and PVT, we end up with a total of 21.03/100K cases. This is still 2.6 times more than your worst-case scenario. Of course, the probabilities are not adjusted for those recovered from Covid-19 and having natural immunity, in which case the risk probably goes down drastically.
The study itself concludes you’re about 10 times more likely to get thrombosis from Covid-19 than from Pfizer/Moderna. I wonder how Jannsen compares to that.
Side note: my mother had a contact with an infected individual. Coincidentally, she tested for antibodies soon after that, and found out they shot up quite drastically (about 7-fold) from almost not existing to a decent level. This gets me thinking just how pointless it is to race after constantly high levels. Haven’t we known for a good deal of time that the immune system produces what’s necessary at the moment?
A lower ratio than I expected! Thanks for doing the analysis. Cheers