Flu shots and useless journalists
Published paper + scientist + useless CNN writer = a game of telephone
One big theme of this Substack is that what we are told “the science” says and what it actually says are often very different. Finding this difference often requires nothing more than reading the studies that ostensibly support the claim being made.
Today’s post is a great example of this. So if you want a better understanding of how scientific research gets misrepresented, dig in. This might be long but it’s worth the read if you want to see the progression from scientific research to mainstream journalism, and how the research gets distorted along the way.
Before we get started I have to share something - while typing this up I had Spotify playing Tchaikovsky, Violin Concerto In D, Op. 35 (with Janine Jansen, the best violinist in the world of violinists whom I can name) and when it finished, the autoplay moved on to an album called ...perchance to dream: A Lullaby Album for Children and Adults.
Do they not know the Shakespeare reference? Do they know what a lullaby is?
To sleep, perchance to Dream; aye, there's the rub,
For in that sleep of death, what dreams may come,
When we have shuffled off this mortal coil,
Must give us pause.
The line is about the fear of death and the unknowns that await us all. Totally the thing you should use when putting the toddler to bed. I love it when a writer says one thing, and someone quotes it without understanding it.
The bit about not understanding the subject matter may have some relevance today. Spin up1 the Tchaikovsky and let’s take get started.
The dumb Daily Beast article
There’s a useless little online publication called The Daily Beast (a name which the founders intended as a clever reference, by the way), where pretend journalists copy the work of other pretend journalists, who themselves mostly just copy press releases or conduct five minute phone interviews.
They put in the extra effort for their readers and wrote a whole paragraph2 summarizing a CNN story about the anticipated failure of the 2021/2022 seasonal flu vaccine to do anything besides generate revenue for the manufacturers:
This year’s flu vaccine is a “major mismatch,” meaning it may not prevent against infection. But experts say people should still get the shot because it will keep severe disease at bay. CNN reports that the main strain of the flu virus that’s circulating has mutated and the vaccine prepped well before flu season does not appear to create antibodies against the variations. But the good news is that the mutations don’t affect another part of the immune system—T-cells—and those can stop you from getting very, very sick if you catch the flu. The mismatch has happened before, and people who were vaccinated fared better than those who skipped the shot.
So this year’s flu shot might not stimulate the production of antibodies for the main expected flu strain, but “experts” say we should get it anyway, or some Pfizer exec won’t make that extra billion this year.
According to the story, even if you don’t come into contact with one of the strains for which the shot was formulated, it will still improve your T-cell response and this is still a benefit – so get the possibly useless shot anyway.
The lazy CNN story that inspired the dumb Daily Beast paragraph
In the journalistic tradition of reporting news by quoting other journalists, they are just summarizing an article3 from CNN. The warm body CNN hired for this didn’t actually interview “experts,” the writer interviewed one – by the name of Hensley.
[Rather than call out the Daily Beast writer by name, from here on that writer will be referred to as Warm Body.]
Anyone who reads papers on influenza vaccines already knows they like to include the caveat that the annual flu shot only works if the vaccine is matched to the strain in circulation. And this year there’s an issue:
"From our lab-based studies it looks like a major mismatch," Scott Hensley, a professor of microbiology at the University of Pennsylvania who led the study, told CNN.
Why is this significant? Crack CNN journalist Warm Body tells us:
The vaccine mismatch may help explain an outbreak of flu at the University of Michigan last month that affected more than 700 people. More than 26% of those who tested positive had been vaccinated against flu -- the same percentage as those who tested negative. That indicates the vaccine was not effective in preventing infection.
So in this case getting vaccinated had no impact on whether students tested positive or negative for the influenza virus. Bummer, dude (in collegiate speak).
Luckily, the changes are unlikely to affect the second line of defense offered by immune system -- cells called T-cells, so even if the vaccines don't protect against infection, they are likely to protect people against severe disease and death, Hensley said.
That sentence by itself is a little confusing (“unlikely to affect”?) but I assume what Warm Body really meant was that those who do get infected will be less likely to have severe illness (or die) if they have been vaccinated despite the mismatch. And per Hensley, this is because the vaccine will still result in a better T-cell response.
[We’re just going to skip past the part where the example given by Warm Body about the university students comes close to contradicting what Hensley is saying. The same percentage of kids got sick regardless of vaccination status. I suppose we’re to believe the vaccinated ones got better faster, or something.]
Before discussing that claim, let’s take a quick look at these antigenic mismatches. CNN tells us:
Flu vaccine efficacy changes from year to year. Part of the problem is the lead time needed to make influenza vaccines. Most are made using old technology that requires the use of chicken eggs incubated for weeks. So vaccine strains are chosen six months ahead of when the vaccines will be delivered.
The CNN article doesn’t really explain much about influenza vaccines, but lucky for us some people at the FDA wrote an overview.
An overview of the regulation of influenza vaccines in the United States (Weir and Gruber 20164)
Developing annual flu shots is even more complicated than Warm Body’s Starbucks order:
This is a complex, lengthy process that requires extensive collaboration among influenza manufacturers, vaccine regulators, and global public health laboratories.
Manufacturers and regulators collaborating? What could possibly go wrong?
They include a nice chart that shows the various steps and when those steps occur. The strain selection occurs in March and the shots start being administered in September, six months later. The original shots contained only one subtype (so they were monovalent) but current flu shots contain strains from several influenza viruses.
Prior to 1978, inactivated seasonal influenza vaccines were monovalent and bivalent; since 1978, most US licensed influenza vaccines have been trivalent incorporating two influenza A subtype viruses (H1N1 and H3N2) and an influenza B virus. More recently, quadrivalent inactivated influenza vaccines containing an additional influenza B virus antigen have been licensed.
Okay, so there are four commonly circulating influenza virus subtypes that could be included in a typical flu shot:
influenza A(H1N1)
Influenza A(H3N2)
influenza B (Victoria)
influenza B (Yamagata)
(I looked up the names for the two influenza B subtypes just to be more complete.5 Looking things up is apparently not something Warm Body is paid to do.)
But each subtype mutates occasionally, creating new strains. So each year the flu shot, which might contain one to four of these, is updated for any changes to the viruses. And as we learned already, sometimes there are mismatches.
Adjuvanted influenza vaccines (Tregoning, Russell, and Kinnear 20186)
This paper gives us some specific numbers on efficacy for a few years:
Influenza vaccines have highly variable rates of efficacy, ranging from 10% in 2004–5 to 60% in 2010–11; the biggest factor being the match or mismatch between the vaccine strains and the circulating strains. Between 2000 and 2011, influenza B vaccine strains did not match circulating strains in six influenza seasons.
2000 to 2011 is twelve flu seasons, and the influenza B vaccine was wrong in half of them. But of course it wasn’t the only flu virus circulating, so anyone receiving a trivalent vaccination might still have been vaccinated for the subtype they actually encountered.
Note that even in good years the shots are not 100% effective, there are always breakthrough infections (people who were vaccinated but became infected anyway).
The paper being cited in the CNN article (the one with Hensley as co-author) is about one of the mismatches, for the current strain of influenza A(H3N2).
Antigenic and virological properties of an H3N2 variant that will likely dominate the 2021-2022 Northern Hemisphere influenza season. (Bolton et al. 20217)
Warm Body told us Hensley led the study, but the lead author is actually Bolton - something we can learn by reading the first page of the study. Hensley is the corresponding author (the one you contact if you’re either a journalist or someone who works at CNN), so he’s the person quoted in the article.
We found that antibodies elicited by the 2021-2022 Northern Hemisphere influenza vaccine poorly neutralize the new H3N2 strain. Together, these data indicate that 3C.2a1b.2a2 H3N2 viruses efficiently replicate in human cells and could potentially cause an antigenic mismatch if they continue to circulate at high levels during the 2021-2022 influenza season.
That’s the antigenic mismatch mentioned in the CNN article. The full designation of the new strain of the H3N2 subtype is 3C.2a1b.2a2.
Our studies suggest that the new 2a2 H3N2 replicates efficiently in human airway cells and can partially circumvent antibodies elicited by egg-adapted 2021-2022 Northern Hemisphere influenza vaccines.
The new strain (2a2) reproduces well in humans so it could cause infection despite vaccination. Here is the quote from the paper that Warm Body was stressing:
Studies have clearly shown that seasonal influenza vaccines consistently prevent hospitalizations and deaths even in years where there are large antigenic mismatches.
The magic words! “Studies have shown,” and apparently in this case, “clearly.”
This is the magic phrase that proves you’re right. The usual way to trump the “studies have shown” card is to play the “those studies have been debunked” card. We’ll take a different approach today.
The paper gives one reference for the statement, which is a review of studies on influenza vaccines and breakthrough infections (people who became infected even though they had been vaccinated).
Reading comprehension isn’t a requirement to be a staff writer at CNN, so it’s not surprising that Warm Body didn’t bother to check the claim. But we completed elementary school so let’s see what that reference says.
Does influenza vaccination attenuate the severity of breakthrough infections? A narrative review and recommendations for further research. (Ferdinands et al. 20218)
First, from the intro:
The possibility that influenza vaccines can attenuate the severity of influenza disease among those who become infected despite vaccination has been an intermittent topic of research and debate for decades.
So this thing we are going to discuss has been debated for decades. It sounds like it might not exactly be a settled question.
We reviewed the literature for evidence to inform the question of whether influenza illness is less severe among individuals who received influenza vaccination compared with individuals with influenza illness who were unvaccinated prior to their illnesses.
The authors are combing through the available published studies and comparing the severity of influenza illness for people were vaccinated but still had the flu, with the severity for people who were not vaccinated and had the flu.
Overall, our findings support the hypothesis that influenza vaccination may attenuate the course of disease among individuals with breakthrough influenza virus infection.
Whoa, “may attenuate.” That’s pretty strong language there. Just as a reminder, “Studies have clearly shown that seasonal influenza vaccines consistently prevent hospitalizations and deaths…” was the quote from the CNN article.
More than one person involved in this story needs to look up the words “may” and “attenuate.” Not that Merriam-Webster can be trusted either, considering all of their updates to the meaning of “vaccine.”9
At this point we need to make a note of something else: nowhere in this paper do the words “antigenic” or “mismatch” occur. As we have already learned, there are breakthrough infections without antigenic mismatch – which appears to be what this paper is reviewing.
But there’s more:
Although the studies we reviewed are intriguing and the trends certainly merit further investigation, none of the studies we reviewed were designed in a way that could reasonably clarify how and why influenza vaccine attenuates the manifestation, severity, and/or duration of influenza illness.
This seems to fall quite far short of “studies have clearly shown” something about influenza cases in years with antigenic mismatches. The studies aren’t specifically about antigenic mismatches, and are “intriguing” but there’s no way to determine how or why they are seeing less severe illness in some people.
And what can we say about T-cells? Nothing for certain, apparently.
More studies with methodological rigor to examine the association between vaccination and the spectrum of pathogenesis from infection to illness and to short- and longer-term impacts on health and functioning would provide additional evidence and could be particularly beneficial for evaluating age- or (sub)type-specific effects. Layering such studies with assessments of humoral and cell-mediated immune responses to vaccination and infections is necessary to disentangle the underlying causal mechanisms by which influenza vaccination may attenuate influenza severity.
[Quick aside: What are humoral and cell-mediated responses? For those new to the study of viruses, this is from Principles of Virology (Flint et al. 201510):
The adaptive response comprises two complex actions, the humoral response (antibody) and the cell-mediated (helper and effector cells).
Okay, so antibodies are a humoral response and T-cells are a cell-mediated response.]
To summarize some points from the Ferdinands paper:
This is the reference given in the original paper, but it doesn’t discuss antigenic mismatches.
This paper is about “breakthrough infections,” which are people who still got the flu despite being vaccinated.
The actual wording says that vaccination may attenuate the severity of influenza infection. Maybe. The available studies are “intriguing” and “merit further investigation.”
If vaccination does have an impact, the mechanism by which this happens cannot be determined from the available studies.
The authors don’t address things like the healthy user effect or the healthy adherer effect (which could be part of that causal mechanism the authors are talking about). What are those, you ask?
Healthy user and related biases in observational studies of preventive interventions: a primer for physicians (Shrank 201111)
The healthy user effect is best described as the propensity for patients who receive one preventive therapy to also seek other preventive services or partake in other healthy behaviors.
Similarly, the healthy adherer effect arises when patients who adhere to preventive therapy are more likely to engage in other healthy behaviors than their non-adherent counterparts.
The Shrank paper goes into examples of therapies that looked good based on observational studies, but randomized controlled trials contradicted those results.
This may be significant in the case of influenza vaccines. The Ferdinands paper reviewed 38 studies - 34 observational and 4 randomized controlled trials. So the majority of their data was from the observational studies.
Getting back to the findings of the Ferdinands paper, the authors make no mention of antigenic mismatches at all. People can have matched antibodies but still get an infection (a breakthrough infection) and the causal mechanism for the (possibly) better outcomes can’t be determined using the available data.
The difference in outcomes they cite might be due to the subjects having the correct antibodies but still getting infected OR it might be due to an improved cell-mediated response (OR something else). But the paper is not specifically reviewing studies with antigenic mismatches and the authors specifically state they don’t know the causal mechanism.
Also note that they are describing an “association” between two things, leaving open the possibility that there’s no cause and effect relationship at all. This could be the case if there are other reasons for the outcomes of those people - such as the healthy user hypothesis.
Here’s a little thought experiment about the observational studies: people with health insurance generally have better health outcomes. Those same people are also more likely to get a flu shot because their insurance pays for it. If the vaccinated do better, is it because of the vaccine or because they also had access to other treatments that “attenuated the severity” of the breakthrough infection?
Oh, and by the way:
Since to our knowledge this is the first review of this topic, we believed basic descriptions of the topics, populations, and methods were needed and that more sophisticated analysis and critiques could follow as the field advances.
The authors work for the CDC and the paper was published in 2021, and there aren’t other papers (that they know of) covering this topic so this is the first.
In reality, there are still a lot of unknowns about influenza and influenza vaccines.
Influenza. (Krammar et al. 201812)
This paper contains lots of basic information about influenza, and according to the authors the efficacy of flu vaccines is still debated.
Despite the availability of seasonal and pandemic influenza vaccines, debate is ongoing as to the efficacy (as measured by randomized controlled trials) and effectiveness (as measured by observational studies involving vaccinated and unvaccinated individuals) of these vaccines.
Recall from a previous post that there are very few randomized controlled trials of these vaccines - most of the data comes from observational studies, which can have issues. However, the authors still see a benefit (although they don’t qualify how much):
Although different methodologies, the variability of the virulence of different seasonal strains, the fit or matching of the vaccine with the circulating strains and age differences of different cohorts can make interpretation of results between studies difficult, most studies find a positive effect of vaccination on the overall health of vaccinated individuals.
But recent years have seen ineffective vaccines, and some researchers have suggested that repeated annual vaccinations might reduce the efficacy.
Nevertheless, many challenges still remain, such as the documented lack of efficacy of LAIVs in the United States in the past years, the possible reduced vaccine efficacy with repeated annual immunizations and the problems associated with vaccine mismatches.
If repeated annual shots might reduce the efficacy, and some years are an antigenic mismatch so the efficacy might be a big fat donut, one has to wonder if maybe there’s a reason to just skip some years.
And there goes my future Pfizer grant.
Let’s review this whole mess
This quote from future Pulitzer Prize winner Warm Body:
Studies have clearly shown that seasonal influenza vaccines consistently prevent hospitalizations and deaths even in years where there are large antigenic mismatches.
- comes directly from the paper of which Hensley is a co-author. The reference cited by that paper doesn’t say anything about antigenic mismatches, and it’s the first (and only) review of this topic.
And this part:
Luckily, the changes are unlikely to affect the second line of defense offered by immune system -- cells called T-cells, so even if the vaccines don't protect against infection, they are likely to protect people against severe disease and death, Hensley said.
- sounds like the opinion of Hensley (and he is welcome to his opinion, it’s still a free country) but the reference being cited is about breakthrough infections, and the authors don’t know if the (small) differences in outcomes they found were due to T-cell response. We learned this by, well, reading the paper.
The authors of the referenced paper, Ferdinands and pals, work for the CDC and their paper (the first of its kind) about breakthrough infections was published in 2021, when hesitancy about vaccination is already high and the current flu shot might not work.
And anyone familiar with the research on influenza and influenza vaccines would already know how inconclusive the evidence is for those vaccines.
But Award Winning CNN Journalist Warm Body wrote up a story that ignores all the details in a very complicated subject and reduces it to: flu shot = always good. This is the reasoning capacity of a ten year old.
In fact, the CNN article reads like it was written by a ten year old whose mommy is Vice President of Sales/Yacht Parties at Pfizer.
So mission accomplished for CNN, and the folks at The Daily Beast are just “useful idiots.”13
One more thing…
Remember that paper about the healthy user? Here’s another quote from the final section, Advice for Practicing Physicians:
When interpreting epidemiologic studies of prevention in the scientific literature, we recommend a healthy skepticism when encountering what seem like surprising large beneficial effects of preventive therapies. Readers must first assess the plausibility of results. For example, one argument against the plausibility of the finding that flu shots reduce mortality by 50% is the observation that the flu season is only associated with a 10% increase in all-cause mortality.
I wonder if they’ve met a guy named Peter Doshi14? Because that’s almost exactly what he said.
21 June 2022 update:
The joke about Vice President of Sales/Yacht Parties turned out not to be a joke at all. From a recent story15 about political corruption in Colombia, partially in the form of a very swanky party on a yacht:
Cambio also revealed that the party was paid for by Pfizer, which has been lobbying for expanding its business in Colombia. In fact, the video released by Cambio shows several Pfizer executives present on the yacht, including Christopher Ariyan, manager of Pfizer for the Andean countries.
It’s getting harder every day to write parody that isn’t also fact.
That’s a reference to something called a vinyl record. Yes, I’m old.
Lazy Daily Beast story about CNN story:
https://www.thedailybeast.com/flu-shot-is-a-mismatch-for-dominant-strain-but-still-get-one
Archived on Wayback Machine:
https://web.archive.org/web/20211218142812/https://www.thedailybeast.com/flu-shot-is-a-mismatch-for-dominant-strain-but-still-get-one
Archived on archive.today:
https://archive.vn/ygLxy
Dumb CNN story about the annual flu shot:
https://www.cnn.com/2021/12/16/health/flu-vaccine-mismatch/index.html
Archived on Wayback Machine:
https://web.archive.org/web/20211218131453/https://www.cnn.com/2021/12/16/health/flu-vaccine-mismatch/index.html
Archived on archive.today
https://archive.vn/O7AZn
Weir JP, Gruber MF. An overview of the regulation of influenza vaccines in the United States. Influenza Other Respir Viruses. 2016 Sep;10(5):354-60. doi: 10.1111/irv.12383. Epub 2016 Mar 24. PMID: 27426005; PMCID: PMC4947948.
https://pubmed.ncbi.nlm.nih.gov/27426005/
Peteranderl C, Herold S, Schmoldt C. Human Influenza Virus Infections. Semin Respir Crit Care Med. 2016 Aug;37(4):487-500. doi: 10.1055/s-0036-1584801. Epub 2016 Aug 3. PMID: 27486731; PMCID: PMC7174870.
https://pubmed.ncbi.nlm.nih.gov/27486731/
Tregoning JS, Russell RF, Kinnear E. Adjuvanted influenza vaccines. Hum Vaccin Immunother. 2018 Mar 4;14(3):550-564. doi: 10.1080/21645515.2017.1415684. Epub 2018 Jan 25. PMID: 29232151; PMCID: PMC5861793.
https://pubmed.ncbi.nlm.nih.gov/29232151
Antigenic and virological properties of an H3N2 variant that will likely dominate the 2021-2022 Northern Hemisphere influenza season. Marcus J. Bolton, Jordan T. Ort, Ryan McBride, Nicholas J. Swanson, Jo Wilson, Moses Awofolaju, Allison R. Greenplate, Elizabeth M. Drapeau, Andrew Pekosz, James C. Paulson, Scott E. Hensley. MedRxiv 2021.12.15.21267857; doi: https://doi.org/10.1101/2021.12.15.21267857
https://www.medrxiv.org/content/10.1101/2021.12.15.21267857v1
Jill M. Ferdinands, Mark G. Thompson, Lenee Blanton, Sarah Spencer, Lauren Grant, Alicia M. Fry,
Does influenza vaccination attenuate the severity of breakthrough infections? A narrative review and recommendations for further research, Vaccine, Volume 39, Issue 28, 2021, Pages 3678-3695, ISSN 0264-410X, https://doi.org/10.1016/j.vaccine.2021.05.011.
https://www.sciencedirect.com/science/article/pii/S0264410X21005624
J. Flint, V.R. Racaniello, G.F. Rall, A.M. Skalka. Principles of Virology, Fourth Edition, Bundle. American Society of Microbiology (2015). doi:10.1128/9781555819521. ISBN 978-1-55581-952-1.
Shrank WH, Patrick AR, Brookhart MA. Healthy user and related biases in observational studies of preventive interventions: a primer for physicians. J Gen Intern Med. 2011 May;26(5):546-50. doi: 10.1007/s11606-010-1609-1. Epub 2011 Jan 4. PMID: 21203857; PMCID: PMC3077477.
https://pubmed.ncbi.nlm.nih.gov/21203857/
Krammer, F., Smith, G.J.D., Fouchier, R.A.M. et al. Influenza. Nat Rev Dis Primers 4, 3 (2018). https://doi.org/10.1038/s41572-018-0002-y.
https://www.nature.com/articles/s41572-018-0002-y
https://www.investors.com/politics/commentary/lenin-used-useful-idiots-to-spread-propaganda-to-the-west/
Story about yacht parties sponsored by Pfizer:
https://thegrayzone.com/2022/06/19/pfizer-miami-yacht-colombia-presidential-rodolfo-hernandez/
Archived copy:
https://web.archive.org/web/20220620043835/https://thegrayzone.com/2022/06/19/pfizer-miami-yacht-colombia-presidential-rodolfo-hernandez/