Commenter Kevin writes:
Consider this article on the CDC website:
“New CDC Study: Vaccination Offers Higher Protection than Previous COVID-19 Infection”
From that headline, you would think that those who have been vaccinated (but have never had Covid) have better immunity than those who have recovered from Covid (but have never been vaccinated). After all, it says right there that vaccination offers higher protection. And in fact this is how the media, government officials, corporate policy makers, university administrations, etc. have interpreted it. It’s the reason why vaccination mandates don’t have an exclusion for those who have previously recovered from Covid.
Yet if one reads the body of the article carefully, the study in question does not support the headline. BOTH groups of subjects in the study were people who had previously been infected and recovered from Covid. The study merely indicates that vaccination can provide ADDITIONAL short-term protection above and beyond the immunity acquired by recovering from Covid. It does NOT compare those who have been vaccinated (only) to those who have recovered from Covid (only).
Given that there are various studies showing that previous infection (alone) provides as good or better protection than vaccination (alone), and better in the case of the Delta variant, this article seems to be a deliberate attempt to mislead the public.
I have no idea if this is a deliberate attempt to mislead—people make lots of errors by accident—but . . . the press release is dated August 6! I’d think that someone during the past two months would’ve notified the problem and informed the CDC press office.
It says, “Page last reviewed: August 6, 2021,” and maybe they have a policy of not correcting old press releases, and I can kind of understand why such a policy would generally make sense, but in this case I hope they can fix the headline.
On the plus side, the body of the press release seems just fine, and the study in question (with the accurate title, “Reduced Risk of Reinfection . . .”) appears to support the CDC’s recommendation, “If you have had COVID-19 before, please still get vaccinated.” So I don’t know that the misleading headline has policy implications, but, yeah, don’t write misleading headlines. And correct them when you learn about the error.
P.S. OK, I guess the CDC isn’t quite so bad as Harvard‘s Jesus story, where the untruths were not just in the headline but in the entire webpage, which was rotten to the core. So, in the spirit of the above, I added “? . . . no, but they could still do better” to the above headline. Had I not made that addition, I’d be as bad as the CDC!
The body is also very misleading, where it says:
At best it shows you are twice as likely to report an infection. As mentioned in the actual CDC report:
The CDC itself told people they didn’t need to get tested if they were vaccinated. It is pretty obvious what is going on here, the only thing we are unsure about is the magnitude. Are vaccinated 1/2 as likely to get tested, 1/4th, 1/10th?
Yes, I think this is a valid point.
The Cleveland Clinic and Israeli studies show the opposite – namely, that previously infected people gain no benefit from the vaccine. Does not sound like scientific consensus to me.
Adrian –
> The Cleveland Clinic and Israeli studies show the opposite – namely, that previously infected people gain no benefit from the vaccine.
Are you sure? The Israeli studies I’ve seen did not show that at all, and pretty much all info I’ve seen says that getting a vaccine has a very large benefit to those previously infected.
Here’s some info:
https://www.nbcnews.com/health/health-news/hybrid-immunity-people-covid-still-get-vaccinated-rcna1974
And Crotty is interviewed here:
https://www.microbe.tv/twiv/twiv-802/
Joshua,
Well, the mainstream media I do not regard as reliable on anything at all. I do not accept it as an authority on anything, but particularly anything scientific or medical. They have all been highly unreliable on that for years, but their reporting on everything else has sunk to that level. If you quote NBC News to me saying that NBC says X, I’m going to interpret that as pretty good evidence that not X is true.
See my reply to Mendel for more. I am by no means the only person interpreting the Cleveland Clinic and Israeli studies as showing no vaccination benefit for the naturally immune.
Adrian –
As I recall, the Israeli study that got the most attention found three levels of protection, in particular against SEVERE DISEASE ABD DEATH.
1 infected + vaxed = most protection
2. Infected but not vaxed =2nd most protected
3. Vaxed but not infected = 3rd most protected.
Please note, even if previously infected aren’t more protected against infection by getting vaxed, they can still get a huge benefit from vaxing if they have stronger immunity against severe disease and death.
Your comment above and the one below seem to be focused only on the benefit regarding protection against reinfection.
And dismissing what Crotty has to say because he’s quoted in an “MSM” source seems foolish to me. But if that’s your inclination then listen to the podcast where he’s in discussion with highly qualified virologists.
Adrian –
Sorry. Looking at your comment below again I see the question of statistical significance is raised by that Israeli study. But that aspect is weirdly contradicted by their abstract.
At any rate, the one (or even two) study rule should apply, imo – as supported here:
https://www.factcheck.org/2021/09/scicheck-instagram-post-missing-context-about-israeli-study-on-covid-19-natural-immunity/
With reference to other findings.
As well, I strongly recommend listening to Crotty in the podcast – as his discussion of the mechanism by which vaxing would confer benefit to those previously infected is extremely important context for grounding any retrospective analysis.
Joshua,
Yes, I agree the Israeli study had some poor wording choices in there. Maybe it was a translation from Hebrew? I’m just guessing.
I don’t trust fact-checkers any more than I trust the mainstream media. It’s really quite simple: I reject the mainstream media because they have rejected the correspondence theory of truth. As a result, I can have no assurance that anything they say corresponds with reality at all.
There aren’t a lot of information sources I actually trust a whole lot. Some conservative news outlets seem reliable, and I can handle any bias they might have. Such conservative news outlets have not abandoned the correspondence theory of truth. Even scientific papers are often very shaky, for reasons that are apparent to any reader of this blog. I would bet that a majority of scientific papers have statistical methodology errors in them, including errors fatal to their conclusions.
I don’t know if I want to make time to listen to Crotty. I could get the same thing from listening to Gregory Poland, most likely. I would ask this question: I know Poland has patents in vaccines, has likely invested in them, etc. Is the same true of Crotty? Is there information from an epidemiologist or virologist who has no financial ties to the vaccine industry to corroborate?
Adrian –
> I don’t trust fact-checkers any more than I trust the mainstream media. It’s really quite simple:
It’s not a matter of “trusting” any source to me. That fact check makes arguments. They seem like good arguments to me and if you have counterarguments I’d be interested to read them.
But even that isn’t my point for providing that link. I provided the article because it links to other relevant studies.
As for Crotty – the Wikipedia page does a good job of providing some information with which to assess his credibility. He has been a co-author on some pretty important work on covid immunity and covid vaccines.
I find it a bit ironic that you dismiss information merely beciswe if the source and yet reference Makary – who has done some high public profile but absolutely terrible advocacy related to COVID.
Adrian, I believe you are mistaken.
Cleveland Clinic, Shresta/Burke/Nowacki/Terpeluk/Gordon, Necessity of COVID-19 vaccination in previsusly infected individuals. medRxiv June 5, 2021
“vaccination was associated a [not] significantly lower risk of SARS-CoV-2 infection [..] among those previously infected (HR 0.313, 95% CI 0 to Infinity).” A hazard ratio of 0.3 is a sizeable effect that comnpares well with teh other studies (CDC Kentucky 1/2.34=0.42, Israel see below)
Israel, Gazit/Shlezinger et al, “Comparing — reinfections versus breakthrough infections”, medRxiv August 25, 2021″
“Examining previously infected individuals to those who were both previously infected and received a single dose of the vaccine, we found that the latter group had a significant 0.53-fold (95% CI, 0.3 to 0.92) (Table 4a) decreased risk for reinfection”
All 3 studies (CDC, Cleveland Clinic, Israel) show a benefit of vaccination for previously infected individuals that is approximately 2-fold.
They are all consistent with the idea that you are less likely to get tested after vaccination.
I don’t understand how dozens of crappy studies repeating the same 2-3 errors are supposed to add up to reliable conclusions. Science doesn’t work that way.
Anoneuoid –
Just an FYI on your estimate a few days ago as to actual deaths in Florida once the lag has been filled in…
-snip-
In the past seven days, on average, the state has added 270 deaths and 4,962 cases per day, according to Herald calculations of CDC data.
-snip-
Read more at: https://www.miamiherald.com/news/coronavirus/article254709207.html#storylink=cpy
Mendel,
Dr. Marty Makary of Johns Hopkins has been quoted as saying, “Requiring the vaccine in people who are already immune with natural immunity has no scientific support. While vaccinating those people may be beneficial – and it’s a reasonable hypothesis that vaccination may bolster the longevity of their immunity – to argue dogmatically that they must get vaccinated has zero clinical outcome data to back it. As a matter of fact, we have data to the contrary: A Cleveland Clinic study found that vaccinating people with natural immunity did not add to their level of protection.” Dr. Makary does not interpret the Cleveland Clinic study the same way you do.
In addition, the Cleveland Clinic study itself says this: “Not one of the 1359 previously infected subjects who remained unvaccinated had a SARS-CoV-2 infection over the duration of the study.” So there were NO BREAKTHROUGH CASES among the previously infected but unvaccinated. I think that’s a very telling number, and I do not think my interpretation of the papers are mistaken.
Mendel,
The full quote from the Cleveland Clinic study is this: “In a Cox proportional hazards regression model, after adjusting for the phase of the epidemic, vaccination was associated with a significantly lower risk of SARS-CoV-2 infection among those not previously infected (HR 0.031, 95% CI 0.015 to 0.061) but not among those previously infected (HR 0.313, 95% CI 0 to Infinity).”
So that’s saying that among the not-previously-infected, vaccination helps. It’s also saying that among the previously infected, the vaccine DOES NOT HELP.
Here’s another direct quote from the Cleveland Clinic paper: “This study provides direct evidence that vaccination with the best available vaccines does not provide additional protection in previously infected individuals.”
So you can talk about hazard ratios all you want, but I think that you are mistaken about this paper.
In the Israeli paper, here is a direct quote: “Individuals who were previously infected with SARS-CoV-2 seem to gain additional protection from a subsequent single-dose vaccine regimen. Though this finding corresponds to previous reports, we could not demonstrate significance in our cohort.”
So that is a conclusion of “not enough evidence to support” the hypothesis that vaccination helps previously infected individuals. The Israeli study definitely shows that natural immunity is FAR BETTER than the vaccine when it comes to the Delta variant. Quote: “Although the results could suggest waning natural immunity against the Delta variant, those vaccinated are still at a 5.96-fold increased risk for breakthrough infection and at a 7.13-fold increased risk for symptomatic disease compared to those previously infected. SARS-CoV-2-naïve vaccinees were also at a greater risk for COVID-19-related-hospitalization compared to those who were previously infected.”
Adrian:
When the confidence interval is 0 to infinity, that doesn’t imply null effect; it implies not enough data to estimate the effect. Saying it wrong in ALL CAPS doesn’t make it right.
Andrew: Fair enough. But then why does the paper say “This study provides direct evidence that vaccination with the best available vaccines does not provide additional protection in previously infected individuals.”?
I’ve wondered why in the US there isn’t more talk about exclusions from mandates for people who have infection-induced immunity – as they have in Israel.
One reason could be that one shot significantly improves the immunity of people who have been infected and recovered. Additionally, apparently there is a great deal of variability in the immunity among people who have recovered from infection. So mandating that the get vaccinated as well does create a better outcome across the spectrum of public health. But still, that seems kind of weak to me.
I have also thought that the reason is that public health officials don’t want to encourage the view that getting infected nets better results than vaccinations, and therefore people are better off if they go out and get infected. The difference in public attitudes towards vaccination here and in Israel might help to explain why Israel has that different policy; in Israel there’s less likelihood that a significant number of people will seek out getting infected as a superior option to getting vaccinated.
Finally, the thought occurred to me that in the US, in contrast to Israel, there might be a much higher prevalence of people who would fake having an infection (proving vaccination seems easier to verify) – not the least because they seem to have a much better system of medical record-keeping.
Hmmmm.
I like to not think that there’s no actual sound thinking behind the policy in the US (even if I don’t necessarily agree with the thinking – as it seems to me an exception for people who have recovered from infection is a good idea).
Oh yeah – I forgot to mention.
I heard that antibody tests are easy to get and free in the UK. So I was wondering why we didn’t have them easily availiable here and why they couldn’t be a part of determining to whom vaccine mandates should apply.
But from what I’ve seen, the antibody testing we available doesn’t really translate that directly. I found that out when I looked into seeing whether I could get an antibody test to assess my own immunity before seeking a booster shot. From what I saw, the antibody tests available don’t give a comprehensive picture of vaccine-induced immunity. I assume it’s also true for infection-induced immunity?
on testing: https://marginalrevolution.com/marginalrevolution/2021/09/the-nytimes-on-the-fda-and-rapid-tests.html
Considering we (in the U.S.) see people lying about their vaccination status and getting fake vaccination cards, I agree that if previous infection ‘counted’ as being vaccinated then people would fake that too.
Even so, it doesn’t make sense to me that previous infection wouldn’t count as being effectively like a vaccination. OK, sure, people will cheat, but those people will cheat anyway. I don’t think there are a lot of people saying “I would never lie about being vaccinated, but sure, I’ll lie about having been infected with COVID.”
I think a bigger problem would be that a lot of people would assume they’ve had COVID but haven’t. Unless you require a blood test that confirms COVID antibodies, I’m not sure how you’d check whether someone has actually had COVID. I think if people are willing to get a test, and the result is positive, then they _should_ be able to qualify as if they’d been vaccinated. But if that is going to happen in large numbers then we’d need a huge expansion in blood testing. And without it, we’d be reduced to taking people’s word for it that they’ve had it. Lots of people who had the flu or the common cold or whatever other ailment would assume they’d had COVID, many of them in good faith, but they’d be wrong. And if you just take people’s word for it that they’ve had it, lots of those same people who are lying about being vaccinated will lie about having had COVID, and it’ll be much easier to get away with the lie because you don’t even have to get a fake vaccination card.
Telling everyone that they should get vaccinated makes sense to me. But if getting COVID provides as much or more protection against future infection as getting vaccinated, then telling people the contrary is immoral.
Phil –
> Unless you require a blood test that confirms COVID antibodies, I’m not sure how you’d check whether someone has actually had COVID.
See my comment above. Do you know that mere confirmation of having COVID antibodies ensures that someone has (at least
minimum degree of) immunity against infection? Would merely a dichotomous test suffice, or to be meaningful would you need a more comprehensive assessment?
Joshua,
I was told that the antibody test is unreliable. Have you read or heard that?
If the purpose is protecting others then its the mucosal immunity that matters. A blood test won’t tell you that at all.
Sameera –
That’s consistent with what I’ve seen. Mina answered a related DM and said he’s been doing a lot of advocating for antibody testing but when I followed up with questions about reliability I didn’t get an answer (I’m sure he’s too busy to answer a high percentage of DM’s from random people).
What Anoneouid said makes sense but testing to see if someone could get infected (and thus transmit) wouldn’t seem to me to be the only relevant question (although from an ethics standpoint it’s probably the main question).
Joshua- Are you sure? Mina has mainly been advocating rapid antigen tests, not antibody tests. They fill completely different functions.
Michael –
Yes. I’m sure. I’ve followed his advocacy for rapid antigen testing since way back in the day.
He said that he’s also been lobbying for antibody testing, (without much success).
Hi Everyone,
Thanks Joshua. I don’t think Michael Mina has addressed the question of the utility of the antibody test. In one of Mina’s tweets, I gleaned that he suggested antibody tests. I could have misread it. I’ll check with him.
Michael, yes Mina’s cause has been in home and at work COVID Rapid Antigen testing. He is a consultant advisor to DETECT, a rapid PCR test.
What I am wondering is whether there is a mucosal immunity test specifically?
Antibody tests that are widely available are highly reliable. Tests directed at detecting the nucleocapsid protein were developed early in the pandemic and were purposely designed with lower sensitivity in order to preserve higher specificity: there were concerns about false positives caused by similar antibodies to human endemic coronaviruses. This issue is perhaps where you’ve heard about low reliability. Most current antibody tests are directed at the spike protein and are highly sensitive AND specific. These antibodies will be detected in essentially all vaccinated individuals and in most naturally infected ones, both within two weeks of exposure. Of note, nucleocapsid antibodies will only be seen in those who had natural infection, though not in all, due to the consciously lower sensitivity. Here I’m speaking about major pharma products.
Antigen tests in general are for acute infection and are much less reliable.
As you say, commercial antibody tests are generally fairly accurate (specificity upper 90s, sensitivity in high 80s to mid 90s depending on the population studied, time since infection, and specific test). But they’re not good correlates of protection. There are three basic issues: 1) they’re not quantitative; 2) not all antibodies are created equal and the tests ignore the distinctions; 3) antibodies are likely not the whole story.
1) and 2) are routinely addressed in lab settings. For (2)Neutralizing antibodies are the preferred target for clinical trials, but they’re technically challenging and are not commercially available. They’re also clearly not the whole story. For (3) there’s growing evidence T cells are important for preventing severe disease, though lots more science to do there.
Basically, if you’re negative after a vaccine I wouldn’t read much into it. If you’re positive before a vaccine you probably had an infection in the past so have some protection, but the various studies trying to quantify that protection don’t give very consistent answers on how good that protection is…probably because protection varies and we don’t really understand why and aren’t measuring the right things in enough people to get good answers.
Fwiw, my sense of the various studies I’ve seen is the ordering by effectiveness is:
Natural infection < vaccination by mRNA < infection then vaccination
But we really only have good data on the middle one.
So yeah, just get a vaccine regardless. There’s very good data to back up effectiveness of those. Given the uncertainty around the immunity provided by past infections, and the certainty of the effectiveness of vaccine, I think the policy choice of not granting exceptions for past infections is a good choice. (The Merits of this particular press release notwithstanding)
The important thing is neutralizing antibodies in the mucosa.
What studies were these?
Len –
Thanks for that. This is what I was (loosely) remembering having read:
There is currently not enough data to support the use of serologic testing for determining protection against primary or re-infection. This is because the antibody response against SARS-CoV-2 represents just one aspect of a complex immune response that is highly dynamic and includes components of innate immunity and virus-specific T-cell activity (Dan, February 2021). While many studies have shown positive associations between the level and quality of neutralizing antibody responses and protection from primary infection and re-infection, others have shown antibody levels to be associated with worse outcomes (Lucas, July 2021; Garcia-Beltran, January 2021), suggesting that the relationship is complex and involves both the quality and dynamics of antibody production. Furthermore, while T-cells may play an important role in protection, FDA-approved assays evaluating SARS-CoV-2 specific T-cell responses are not in widespread use, as the clinical significance of testing is unknown. Additionally, the correlation between antibodies and T-cells generated in response to natural infection or immunization is likely to vary by patient population. A lack of antibodies does not necessarily mean that a patient is not protected, as a memory B-cell response generated from prior infection or vaccination may still confer protection upon re-exposure.
https://www.idsociety.org/covid-19-real-time-learning-network/diagnostics/antibody-testing/
So I guess when I was thinking about “reliability, ” I was (unintentionally) conflating a scientific distinction with a less scientific use of the word.
Do you have thoughts on why they aren’t more readily available here like in the UK, and potentially used as exclusion/confirmatory tests relative to accine mandates?
Not just Israel – much of europe is also vaccine or previous infection.
I don’t find this surprising whatsoever: vaccines have entered the hyper-polarized political blender where nuance goes to die. Admitting that prior infection is approx. as good as vaccination takes away a blunt object to hammer the red team with. This also explains why “mistakes” like this headline are so frequent.
“I like to not think that there’s no actual sound thinking behind the policy in the US.”
Haven’t ever worked in government but my experience in a wide range of organizations and business is that policies are rarely made solely on the basis of sound thinking. Always some politics and capriciousness. So my prior would be there’s going to be sound thinking and some of something else.
You’re suggesting that this isn’t a big deal since the body of the article is fine (maybe), but the headline is so stunningly misleading that the CDC’s posting of this is really quite unethical. I think things like this contribute to mistrust of agencies like the CDC, as well as nuttier conspiracy theories. After all, with this headline, the CDC *is* lying to us…
Raghu:
I agree that the misleading headline is a big deal, whether or not it’s on purpose. If it was a mistake, then I guess it’s not quite unethical to make a mistake, but it’s unethical not to fix it once it’s been noticed.
I agree. I sent an email to CDC media relations a few minutes ago, by the way.
After I learned last week how many people still cite retracted studies, my capacity for shock is diminished.
I certainly won’t spend energy getting upset over a press release headline when the title of the actual study gets it right; I trust the scientists more than the journalists anyway, and correcting a months-old press release won’t have much of an effect.
These aren’t journalists being incorrect and misleading. It’s the *CDC*’s press office! Also, the CDC press office, unlike my university’s press office, is presumably not staffed by people who have to write press releases related to the English department, the business school, the psychology department, the school of music, etc: they write press releases *about disease control and prevention.* Period.
I still expect the press office to be staffed by people with a journalism background and not a biology/epidemiology background.
Some mistakes ARE unethical. If I confuse the units on an engineering calculation and people die… But a simple routine check would have caught it. That’s unethical, and likely illegal. In my opinion, if you publish this press release but didn’t ask anyone to review it… It’s unethical. If they reviewed it and allowed it through then it’s likely deliberate.
A big part of what’s going on in society these days is that we have gone all post truth post ethics. cf Donald Trump as president
Daniel,
It’s not a post-truth world per se. It’s just that everyone seemingly has their own truth and are fond of speaking about it.
I agree that consequences matter and have some bearing on whether a particular action/inaction was ethical or not. Divorcing consequences from the matter entirely is a mistake.
The misleading headline aside, I take issue with the implied certainty of conclusion — originally by the CDC and later by Harvard — coming from a very small and obviously confounded observational study. Confidently concluding a >2 odds ratio, from “a few hundred” Kentuckians over a single month of observation is totally non-credible.
A more credible explanation: there is a correlation between KY residents who got vaccinated even after testing positive and those who were “out and about”.
KaiserFung has done a nice job documenting on his blog, for a lay audience, the pro-vaccine biases inherent in these observational studies. From the start he’s estimated the much-hyped 99%+ VE estimates were actually more like ~60%. Which to be clear would be amazing if it weren’t being oversold. I’d love to see that thread of analysis covered by this (non-lay) blog.
The headline wasn’t a mistake. It was spin, propaganda, advocacy. It’s telling people whatever the headline writer thought was most likely to make them get vaccinated. Like every other pronouncement from CDC since Day One of the pandemic.
Name:
I have no idea, but I agree with the general principle that making such errors is irresponsible at best, and leaving such errors uncorrected is mistakes and leaving them uncorrected is wrong.
Here are some data from an Oklahoma hospital that shows “vaccines to be more effective than natural immunity”:
https://oklahoma.gov/content/dam/ok/en/covid19/documents/weekly-epi-report/2021.09.29%20Weekly%20Epi%20Report.pdf
I can’t tell if there is a confounding factor that they didn’t control for their vaccine group to not have any record of previous infection though, so I suspect that biases the result. The relative rate of breakthrough vs reinfection ranged form 2x to 10x (with unvaccinated reinfection being more likely than breakthrough). If they didn’t control for those vaccinated individuals also having previous covid infections, we might assume that a significant double-digit percentage of them probably did already have some natural immunity which would work in conjunction with the vaccine to boost their overall immunity even more (I think that’s what the Israeli study found). Furthermore the Israeli study showed that vaccine immunity waned over time and that being a factor making natural immunity more effective. This Oklahoma data doesn’t seem to specify anything about time since vaccination or time delay between vaccination and breakthrough infection.
Overall, both vaccination and natural immunity seem to be quite effective though, and that seems to be a consistent finding among studies so far.