The CDC says: Even if you already had Covid, get vaccinated. But is that based on science? Good science? This is an important question! Many people, including myself, already had Covid before the vaccine came out. So I spent a week analyzing the studies themselves. Here’s what I learned.
Maxim- good article, and nice attempt to put things into context. It is important that we put these "relative" risk numbers into perspective, in absolute terms, and I applaud you for that effort.
I have done my own deep dives on the topic I agree, that vaccination in recovered COVID individuals roughly halves the risk (a recent study out of Kaiser says it may be lower for Moderna, maybe 8-33%). Nonetheless, if the risk reduction is less than 50%, it would not even pass the FDA threshold for vaccine efficacy (for COVID recovered individuals). But lets assume for a second, that vaccination in the previously infected leads to about a 50% risk reduction.
Your assertion that the baseline risk of the naturally immune to <reacquire> COVID is between 4-9% is wrong for several reasons: [1] not sure where in your citing article you see the 4-9% number, [2] the number you are referring to is the number of a likely never infected individuals getting infected over the course of the year (The Pfizer trial cites about 6%/year), and [3] the article you cite estimates (not measures) asymptomatic infection -- which none of the vaccine trials actually ever tested efficacy for. I would ask you to look at Lawandi et al. that found incidence of previously infected to be roughly 0.2%, Hanrath et. al, and Lumley et al. to be <1%.
The famous "KY" study only found 246 reinfections out of 275,000 individuals - a .09% over two months, so about 0.5% per year. So many studies found a symptomatic rate of <1% per year. So your estimate of 6% is much to high, and the absolute risk reduction much to high as well.
If you take an upper bound of reinfection as 1% (.01), then your risk reduction after vaccination would be 0.005. The number need to treat (NNT) would be closer to 200 (rather than the 25 you calculate) in the course of a year. Moreover, the vaccine efficacy studies were really only performed over the 3-5 months, so its not really fair to say that it continues to be 50% effective for a year as well (it could be less over the year). The NNT for a never infected person is closer to 25-30, based on the vaccine trials, however. So your calculation, is an order of magnitude off.
Not trying to be annoying about this, because, again I really applaud the effort to look at the question in absolute (rather than only relative figures). But I think it is important for your readers to understand this.
Thanks, interesting points. Off the bat: the UK study actually did do randomized testing, and so should have caught most asymptomatic cases, too. I need to look at what their estimated rate was...
The KY study is surely undercounting, and possibly way undercounting, due to make people not seeking testing.
The 4-9% is an estimate based on a study finding that general pop has a 33% chance of infection in 2020, and applying the protectivity ranges I'd already calculated earlier to that.
KY study observation was also not during a major "wave", and doesn't cover nearly a year. If you adjust their 0.5 to annualize it and account for the higher case numbers at other parts of the year, I bet you get close, or into, my range. Will do this calculation when I get time.
The KY study was over 2 months. 0.09*6 annualizes to 0.5%. Account for higher case numbers for greater phases? OK, but how do you do that if you do not know how well NI performs against those variants? (A 20x rise in overall cases, doesn't mean a 20x rise in reinfection in the recovered.). You can get to the number you want by making any assumption you want, but what good is that?
You do raise that the KY study has some problems. And you are right about that. If it "undercounted", then why are we accepting its 50% risk reduction value ( I do agree with this value, because other studies endorse it, and this just seemed to stumble on it). Also, the KY study was not a test-negative design -- meaning the non-reinfected, never got tested to prove no infection. Such a flawed study overall, but somehow, the CDC and media promoted it like crazy.
I would encourage you to look at studies that looked at actual incidence, not just a fuzzy estimation from one study in a news release. See below for peer reviewed studies. Lawandi et al. is most relevant. NIH study of 238 hospitals in the US, reinfection 0.2%/4-6 months (which included the second phase). There is no way the range for reinfection in the recovered is in the 6% range.
Abu-Raddad LJ, Chemaitelly H, Malek JA, Ahmed AA, Mohamoud YA, Younuskunju S, Ayoub HH, Al Kanaani Z, Al Khal A, Al Kuwari E, Butt AA, Coyle P, Jeremijenko A, Kaleeckal AH, Latif AN, Shaik RM, Rahim HFA, Yassine HM, Al Kuwari MG, Al Romaihi HE, Al-Thani MH, Bertollini R. Assessment of the risk of SARS-CoV-2 reinfection in an intense re-exposure setting. Clin Infect Dis. 2020 Dec 14:ciaa1846. doi: 10.1093/cid/ciaa1846. Epub ahead of print. PMID: 33315061; PMCID: PMC7799253.
Hanrath AT, Payne BAI, Duncan CJA. Prior SARS-CoV-2 infection is associated with protection against symptomatic reinfection. J Infect. 2021 Apr;82(4):e29-e30. doi: 10.1016/j.jinf.2020.12.023. Epub 2020 Dec 26. PMID: 33373652; PMCID: PMC7832116.
Sheehan MM, Reddy AJ, Rothberg MB. Reinfection Rates among Patients who Previously Tested Positive for COVID-19: a Retrospective Cohort Study. Clin Infect Dis. 2021 Mar 15:ciab234. doi: 10.1093/cid/ciab234. Epub ahead of print. PMID: 33718968; PMCID: PMC7989568.
Lawandi A, Warner S, Sun J, Demirkale CY, Danner RL, Klompas M, Gundlapalli A, Datta D, Harris AM, Morris SB, Natarajan P, Kadri SS. Suspected SARS-CoV-2 Reinfections: Incidence, Predictors, and Healthcare Use among Patients at 238 U.S. Healthcare Facilities, June 1, 2020- February 28, 2021. Clin Infect Dis. 2021 Aug 5:ciab671. doi: 10.1093/cid/ciab671. Epub ahead of print. PMID: 34351392.
There is no mention of side effects of the vaccines on "young" men and women as regards to the occurrence of myocarditis for the young men and the albeit sparse reporting of menstrual cycle disruptions in young women. In point of fact, all these "studies" are too small and the vaccines too new to proclaim any "long lasting" efficacy and safety.
If one finds themselves in a "high risk" group, the risk/benefit ratio validates taking the vaccine. A fifty-year-old woman isn't worried too much about menstrual disruptions, and, as far as I know, older men have yet to report myocarditis after taking the "jab." Of course, the deaths following the "jab" seem disturbingly high, so you might want to look into that as well.
I already sent you this in a tweet, however, incase it gets flagged & so others may get a chance to advance their research, I wanted to share this here as well: https://youtu.be/ihjNDf32_Ac .
Hello, you discuss vaccines decreasing effectiveness over time. Does this only affect chances of getting an infection? Or is the reduction of symptoms when vaccinated also decreasing with time? This is important because many say the main benefit of vaccines is much less hospitalization chance if you do get infected anyway. Does this also wear off with time?
Hey! No evidence that I'm aware of, that effectiveness vs symptoms wanes. May also want to take a close look at the Israeli graph in this article, which has some sort of relevant data.
Maxim- good article, and nice attempt to put things into context. It is important that we put these "relative" risk numbers into perspective, in absolute terms, and I applaud you for that effort.
I have done my own deep dives on the topic I agree, that vaccination in recovered COVID individuals roughly halves the risk (a recent study out of Kaiser says it may be lower for Moderna, maybe 8-33%). Nonetheless, if the risk reduction is less than 50%, it would not even pass the FDA threshold for vaccine efficacy (for COVID recovered individuals). But lets assume for a second, that vaccination in the previously infected leads to about a 50% risk reduction.
Your assertion that the baseline risk of the naturally immune to <reacquire> COVID is between 4-9% is wrong for several reasons: [1] not sure where in your citing article you see the 4-9% number, [2] the number you are referring to is the number of a likely never infected individuals getting infected over the course of the year (The Pfizer trial cites about 6%/year), and [3] the article you cite estimates (not measures) asymptomatic infection -- which none of the vaccine trials actually ever tested efficacy for. I would ask you to look at Lawandi et al. that found incidence of previously infected to be roughly 0.2%, Hanrath et. al, and Lumley et al. to be <1%.
The famous "KY" study only found 246 reinfections out of 275,000 individuals - a .09% over two months, so about 0.5% per year. So many studies found a symptomatic rate of <1% per year. So your estimate of 6% is much to high, and the absolute risk reduction much to high as well.
If you take an upper bound of reinfection as 1% (.01), then your risk reduction after vaccination would be 0.005. The number need to treat (NNT) would be closer to 200 (rather than the 25 you calculate) in the course of a year. Moreover, the vaccine efficacy studies were really only performed over the 3-5 months, so its not really fair to say that it continues to be 50% effective for a year as well (it could be less over the year). The NNT for a never infected person is closer to 25-30, based on the vaccine trials, however. So your calculation, is an order of magnitude off.
Not trying to be annoying about this, because, again I really applaud the effort to look at the question in absolute (rather than only relative figures). But I think it is important for your readers to understand this.
Thanks, interesting points. Off the bat: the UK study actually did do randomized testing, and so should have caught most asymptomatic cases, too. I need to look at what their estimated rate was...
The KY study is surely undercounting, and possibly way undercounting, due to make people not seeking testing.
The 4-9% is an estimate based on a study finding that general pop has a 33% chance of infection in 2020, and applying the protectivity ranges I'd already calculated earlier to that.
KY study observation was also not during a major "wave", and doesn't cover nearly a year. If you adjust their 0.5 to annualize it and account for the higher case numbers at other parts of the year, I bet you get close, or into, my range. Will do this calculation when I get time.
The KY study was over 2 months. 0.09*6 annualizes to 0.5%. Account for higher case numbers for greater phases? OK, but how do you do that if you do not know how well NI performs against those variants? (A 20x rise in overall cases, doesn't mean a 20x rise in reinfection in the recovered.). You can get to the number you want by making any assumption you want, but what good is that?
You do raise that the KY study has some problems. And you are right about that. If it "undercounted", then why are we accepting its 50% risk reduction value ( I do agree with this value, because other studies endorse it, and this just seemed to stumble on it). Also, the KY study was not a test-negative design -- meaning the non-reinfected, never got tested to prove no infection. Such a flawed study overall, but somehow, the CDC and media promoted it like crazy.
I would encourage you to look at studies that looked at actual incidence, not just a fuzzy estimation from one study in a news release. See below for peer reviewed studies. Lawandi et al. is most relevant. NIH study of 238 hospitals in the US, reinfection 0.2%/4-6 months (which included the second phase). There is no way the range for reinfection in the recovered is in the 6% range.
Abu-Raddad LJ, Chemaitelly H, Malek JA, Ahmed AA, Mohamoud YA, Younuskunju S, Ayoub HH, Al Kanaani Z, Al Khal A, Al Kuwari E, Butt AA, Coyle P, Jeremijenko A, Kaleeckal AH, Latif AN, Shaik RM, Rahim HFA, Yassine HM, Al Kuwari MG, Al Romaihi HE, Al-Thani MH, Bertollini R. Assessment of the risk of SARS-CoV-2 reinfection in an intense re-exposure setting. Clin Infect Dis. 2020 Dec 14:ciaa1846. doi: 10.1093/cid/ciaa1846. Epub ahead of print. PMID: 33315061; PMCID: PMC7799253.
Hanrath AT, Payne BAI, Duncan CJA. Prior SARS-CoV-2 infection is associated with protection against symptomatic reinfection. J Infect. 2021 Apr;82(4):e29-e30. doi: 10.1016/j.jinf.2020.12.023. Epub 2020 Dec 26. PMID: 33373652; PMCID: PMC7832116.
Sheehan MM, Reddy AJ, Rothberg MB. Reinfection Rates among Patients who Previously Tested Positive for COVID-19: a Retrospective Cohort Study. Clin Infect Dis. 2021 Mar 15:ciab234. doi: 10.1093/cid/ciab234. Epub ahead of print. PMID: 33718968; PMCID: PMC7989568.
Lawandi A, Warner S, Sun J, Demirkale CY, Danner RL, Klompas M, Gundlapalli A, Datta D, Harris AM, Morris SB, Natarajan P, Kadri SS. Suspected SARS-CoV-2 Reinfections: Incidence, Predictors, and Healthcare Use among Patients at 238 U.S. Healthcare Facilities, June 1, 2020- February 28, 2021. Clin Infect Dis. 2021 Aug 5:ciab671. doi: 10.1093/cid/ciab671. Epub ahead of print. PMID: 34351392.
There is no mention of side effects of the vaccines on "young" men and women as regards to the occurrence of myocarditis for the young men and the albeit sparse reporting of menstrual cycle disruptions in young women. In point of fact, all these "studies" are too small and the vaccines too new to proclaim any "long lasting" efficacy and safety.
If one finds themselves in a "high risk" group, the risk/benefit ratio validates taking the vaccine. A fifty-year-old woman isn't worried too much about menstrual disruptions, and, as far as I know, older men have yet to report myocarditis after taking the "jab." Of course, the deaths following the "jab" seem disturbingly high, so you might want to look into that as well.
Thanks for the effort.
I already sent you this in a tweet, however, incase it gets flagged & so others may get a chance to advance their research, I wanted to share this here as well: https://youtu.be/ihjNDf32_Ac .
Interesting perspective in The Lancet. https://www.thelancet.com/pb-assets/Lancet/pdfs/S0140673621020468.pdf
Hello, you discuss vaccines decreasing effectiveness over time. Does this only affect chances of getting an infection? Or is the reduction of symptoms when vaccinated also decreasing with time? This is important because many say the main benefit of vaccines is much less hospitalization chance if you do get infected anyway. Does this also wear off with time?
Hey! No evidence that I'm aware of, that effectiveness vs symptoms wanes. May also want to take a close look at the Israeli graph in this article, which has some sort of relevant data.