Kids are much less likely to die from COVID than older people, and vaccines have side-effects. So is it worth vaccinating them? Here are the costs and benefits.
The UK JCVI statement you discuss was about vaccination for ages 12-15. The JCVI recognized a benefit, so it is unlikely that the additional considerations you give would change the calculations much, particularly as it had not been demonstrated that the vaccine reduces the risk of PMIS, if infected. However, there is not necessarily a big difference between the JCVI's and your views. The JCVI recognized that there may be other considerations that strengthen the case for vaccination, and explicitly suggested the Chief Medical Officers consider those in making their final recommendation. The decision of the CMOs at that time was to make one dose of vaccine available, recognize that there should be no stigmatization of parents who choose not to vaccinate due to the fine balance of arguments, and state that the issue of a second dose should be revisited in a few months. Those decisions were before Omicron emerged. When Omicron emerged, the UK decided to give second doses for ages 12-15.
Thank you for your detailed analysis. I've not seen anything done as thoroughly anywhere. Although I agree with your findings and methods, I don't come to the same conclusion and this is why --- based on your final summary graph, about 650 unvaccinated kids per 100k are likely to admitted to an ICU, or have some side effect... vs 60/70 per 100k with vaccine. This sounds like a no brainer but for me even 650 per 100k is not a compelling case when you weigh in the fact that this is still new, and who knows if other worse side effects will surface over time. So for me, it's still a NO for vaccinating my kids. But thank you for all the detailed analysis.
My take on this is you need to make a guess on what’s most likely: additional side-effects of vaccines, or of COVID.
What do you think can leave you with more long-term, unknown side effects: a piece of RNA that only produces a bunch of spike proteins, or billions of entire live viruses in your bloodstream, infecting cells throughout your body, replicating, mutating?
If anything, this is an underestimate of comparative harm.
Yes that's the million dollar question. I guess I have more faith in nature producing a vaccine that is better than what humans can produce and I'd rather that vaccine make it to my daughter naturally even if it could potentially 0.065% harm her... because it would give her much stronger immunity and prevent future harm from the same virus and possibly more variants without needing more shots. I wish there were more studies on natural immunity vs mRNA immunity. I assume my 3 year old daughter has already been exposed to COVID and has immunity by now. The fact they are giving anyone a COVID shot prior to knowing if they are already immune from a previous exposure is what bothers me the most. I believe nature has the best immunity to offer and to discount it and push the shot is what makes me not wish to participate for my children.
I personally have taken it because I'm old and have pre-existing conditions. If it ends up stunting my growth, fogging my memory or reducing my fertility, I've accepted that possibility and take the vaccine anyways because of it's benefits and my time to need those (except maybe the memory) has passed. But I don't wish to subject my daughter to a bunch of unknowns, especially if they're not even willing to factor natural immunity into the equation.
The question is the long-term effects of vaccines vs. long-term effects of the virus. The latter are much more probable.
I don't agree with this last point you make. That's like cutting a forest so that it doesn't burn. The point of the vaccine is so that you don't get an infection, and if you do, you're much more protected and the infection is mild. Suffering the infection to be protected for future infections defeats the purpose.
mRNA vaccines are not "experimental" in any meaningful sense. mRNA produces an antigen which produces an immune response. Neither process is new and both are well understood. mRNA produces all our proteins. The immune reaction to the antigen is no different than that from any other. The mRNA in the vaccine is rapidly destroyed just like all other mRNAs - in fact the biggest technical hurdle with the vaccines has been developing a capsule to protect them for long enough so that they can produce enough antigen. The antigen itself is only part of the spike. It is not infectious. All this is basic immunology. The long term risk is insignificant and we have way more data on vaccine harm vs benefit that a knowledgeable person needs to justify the shot.
Hello Tomas, Thank you so much for your informative article. I live in the UK and have 2 children boy age 10 and girl age 9. They both currently have covid 19 (my husband and I are currently triple vaccinated and negative) so I want to know if you think giving our kids a first Pizer vaccine in time would benefit them since they have caught omnicron and also if there is any data yet on reproductive issues from the vaccine? Kind regards and many thanks
Thanks, Tomas. Super clear and well explained, like all your articles. I have one question, what about babies <1 year old? Currently pregnant, I got my three vaccine shots (2 pfeizer and the third, moderna) so my baby should have some kind of immunity when he’s born, but to what extent will he be protected? Is there any data/report on protection on babies? Or plans if vaccines will be tested and available for them in the near future? Thanks a lot
Great article. Thanks for posting. Quick question about the final bar chart: for apple to apple comparison, shouldn't you consider weighing the odds of getting covid? In other word, if you get the vaccine, you have 50/100K chances of getting side effects. If you do not get the vaccine, you have 600/100K chances of getting side effects conditioning on getting covid. Let's say that 20% of kids will get covid (I don't know what the right number is). Then, shouldn't you divide the covid bar by 5? Would the two bars be comparable then? Maybe I am missing something...
Yes. I mention it in the article. Based on Omicron and the evolution of variants, I put the likelihood of getting exposed to COVID in the next few months / years high enough that I didn't think it was useful to discount it. If you disagree, feel free to do exactly what you say!
Thank you Tomas for another elegant summary. As I read it I kept thinking of writing a sister article called “Should You Face Your Fear of Vaccinating Your Kids?” Evolution has left us poorly equipped to parse nuance and uncertainty without feeling helpless and vulnerable. In a way the cognitive dissonance of vaccine deniers is akin to an overactive immune response to foreign information that threatens their worldview. I’d love to live in a binary black and white safe/unsafe world but this messy one is home. When things get tough we really do seem like emotional beings who think more than thinking beings who feel.
Hi Tomas - Love your work, but I was surprised by the really small CFS percentages as they were markedly lower than the ~2-3% (across all ages) you reported in your Long Covid article. I went to the source of the data in the graph you presented here (https://www.ncbi.nlm.nih.gov/labs/pmc/articles/PMC6331450/) and it is actually NOT from covid patients. It was published in 2018 before covid. Please update or clarify when you get a chance.
When the risk of serious harm AND major benefit is low, then minor harm and benefit become more important. For many parents the inconvenience of arranging the jab and pain from the needle outweigh the benefits. The indigenous population I work with solved this by vaccinating at schools. Everybody gets it, which is convenient and children see that it is no big deal. Of course the tribe really cares about its elders, unlike many Westerners.
Thanks for the analysis. What would be interesting to known to better weight the risk-benefit balance of COVID-19/vaccines in kids would be to compare the data you report with those from other respiratory infections of kids/vaccines. The flu for example, what is the incidence (or better what was in the pre-covid era) of complications/hospitalizations/death in kids between 5-12? In my country (Italy) the flu-vaccine in the pre-covid era was indicated only for the elderly (over 65 years of age) and my kid's pediatrician had never considered to suggest the flu-vaccine. If hospitalization and death incidence of the flu are somewhat similar to those of COVID-19, it would be a nice piece of extra information to add on the balance. There's still the protect the others reason that might be more relevant in this pandemic from a national health perspective, but when the (healthy) kids are your own the protect the others inevitably comes second (especially when you still have a relevant percentage of unvaccinated elderly that should protect themself).
Actually, flu numbers are based per 100k symptomatic ilnessess not infections so numbers are probably worse than COVID, which is often asymptomatic in this age range and unrecognized.
Unfortunately, flu statistics numbers pull together 5-11 and 12-17 years of age.
If I did the maths right, this supports the conclusion that vaccinating kids against COVID is primarily a matter of public health. I am not saying it is not right, I am just saying it is not really about the kids (as lots, at least in Italy, are saying).
Yeah so I’m not sure we’re far away from the ball park. But there’s a structural issue with comps with the flu: the denominator.
The flu’s underascertainment is very unlikely to be at the “heavy symptomatic” level, because ppl in those cases do go to the doctor. So if you have 100k cases and 600 deaths, what you really have is 100k heavy symptomatic cases and 600 deaths, but, say, 500k infections and 600 deaths. Your fatality rate is going to be, say, 5x lower.
This doesn’t happen with COVID because we have massive monitoring and population-wide analyses. So we have a much better sense of the denominator.
Mmmm, I am not sure about that. Statistical models which take into account the underdetection of cases are used by the CDC for influenza surveillance (see methods on the same web page). In Italy the epidemiologic surveillance is based every year on the numbers communicated by every GP, pediatrician etc. and parents always take kids to the pediatrician for a cold and a fever, a flu in a child would not go unmissed (adults is surely a different thing). COVID instead go largely missed because it is asymptomatic at this age range in more than 50% of the cases (Italian health institute data) and we are completely unable at tracking and testing :-( (by we I mean in Italy). Anyway, I am positive that those number are not really comparable, they should be studied in depth, but I would not expect a huge difference in favor of one or the other. I am still inclined to think that the institutional communication in favor of the vaccination of children is mainly driven by the public health hence socio-economic burden of the pandemic.
Oh I absolutely agree with you that the institutional communication is for public health purposes. In fact, that's how I lead the article, right? You can't trust what they say because they have a public health interest in mind and have a track record of lying to achieve it. That's why I wanted to independently look at the numbers.
Agree also on the stats of flu vs. COVID. A deep dive would be valuable here. It's unclear where the data lies—and whether the conclusions you or I draw could be drawn!
Your article is great but you should really take into account the fact that that the vaccine isn't based on the spike of the variants currently circulating. It protects very well against severe disease - and likely will keep doing so - because there are many epitopes on the prefusion spike it produces + cellular immunity adapts itself; but for kids, the focus should probably be more on long covid and/or potential concentration of spike protein in some organs (which means potential long term side effects, from the vaccine as well as from the infection) than on severe disease.
If they get infected easily after the vaccine (and they likely will with Omnicron, as you mention) they might develop long covid, as well as other potential post-covid long term effects, vaccinated or not. The odds of getting long covid might be lower, we don't know yet, nor do we know how lower it's gonna be, but we know that immune escape variants will get more and more likely. And that vaccinated kids who never met the virus (and have no cross immunity with an hCoV) will get infected at some point in the future.
After vaccine + infection (or infection + vaccine, or even infection only probably) the situation is not the same as there is some local immunity in the upper respiratory track and immunity, not focusing only on S is wider; with the vaccine only the only thing that will happen is that kids might recover quicker, and that's not even sure for those who would have recovered quicker anyway. Of course, for those who would have needed to go to the ICU the vaccine would have been a game changer. But we don't vaccinate kids against the flu. For those who go to the ICU, the vaccine would have been nice there as well...
Of course, for kids at risk of severe disease the answer is easy: get the vaccine asap. For the healthy ones the vaccine is probably a good choice, but it might be not that useful if it only provides protection against severe disease. In my opinion the question remains open because the risk of severe disease is pretty low, and the vaccine doesn't necessarily protects against infection side effects from the current - and next - variants.
- Most kids have been exposed to omicron if they go to school, so the debate is even less important today than it was before. With natural immunity, you don’t need vaccines as much
- this effect is likely due to dosage more than anything else. Extremely unlikely that it works for all ages except for one small range
The UK JCVI statement you discuss was about vaccination for ages 12-15. The JCVI recognized a benefit, so it is unlikely that the additional considerations you give would change the calculations much, particularly as it had not been demonstrated that the vaccine reduces the risk of PMIS, if infected. However, there is not necessarily a big difference between the JCVI's and your views. The JCVI recognized that there may be other considerations that strengthen the case for vaccination, and explicitly suggested the Chief Medical Officers consider those in making their final recommendation. The decision of the CMOs at that time was to make one dose of vaccine available, recognize that there should be no stigmatization of parents who choose not to vaccinate due to the fine balance of arguments, and state that the issue of a second dose should be revisited in a few months. Those decisions were before Omicron emerged. When Omicron emerged, the UK decided to give second doses for ages 12-15.
That makes sense. Thank you for the added context.
Excellent post. Can you draft an essay on the efficacy of booster shots?
Maybe in the future, not planned right now
Thank you very much for your excellent work.
Thank you for your detailed analysis. I've not seen anything done as thoroughly anywhere. Although I agree with your findings and methods, I don't come to the same conclusion and this is why --- based on your final summary graph, about 650 unvaccinated kids per 100k are likely to admitted to an ICU, or have some side effect... vs 60/70 per 100k with vaccine. This sounds like a no brainer but for me even 650 per 100k is not a compelling case when you weigh in the fact that this is still new, and who knows if other worse side effects will surface over time. So for me, it's still a NO for vaccinating my kids. But thank you for all the detailed analysis.
Thx for your thoughtful and introspective reply.
My take on this is you need to make a guess on what’s most likely: additional side-effects of vaccines, or of COVID.
What do you think can leave you with more long-term, unknown side effects: a piece of RNA that only produces a bunch of spike proteins, or billions of entire live viruses in your bloodstream, infecting cells throughout your body, replicating, mutating?
If anything, this is an underestimate of comparative harm.
Yes that's the million dollar question. I guess I have more faith in nature producing a vaccine that is better than what humans can produce and I'd rather that vaccine make it to my daughter naturally even if it could potentially 0.065% harm her... because it would give her much stronger immunity and prevent future harm from the same virus and possibly more variants without needing more shots. I wish there were more studies on natural immunity vs mRNA immunity. I assume my 3 year old daughter has already been exposed to COVID and has immunity by now. The fact they are giving anyone a COVID shot prior to knowing if they are already immune from a previous exposure is what bothers me the most. I believe nature has the best immunity to offer and to discount it and push the shot is what makes me not wish to participate for my children.
I personally have taken it because I'm old and have pre-existing conditions. If it ends up stunting my growth, fogging my memory or reducing my fertility, I've accepted that possibility and take the vaccine anyways because of it's benefits and my time to need those (except maybe the memory) has passed. But I don't wish to subject my daughter to a bunch of unknowns, especially if they're not even willing to factor natural immunity into the equation.
This is not what I said.
The question is the long-term effects of vaccines vs. long-term effects of the virus. The latter are much more probable.
I don't agree with this last point you make. That's like cutting a forest so that it doesn't burn. The point of the vaccine is so that you don't get an infection, and if you do, you're much more protected and the infection is mild. Suffering the infection to be protected for future infections defeats the purpose.
Well the problem is Omnicron: kids will get infected, vaccinated or not!
mRNA vaccines are not "experimental" in any meaningful sense. mRNA produces an antigen which produces an immune response. Neither process is new and both are well understood. mRNA produces all our proteins. The immune reaction to the antigen is no different than that from any other. The mRNA in the vaccine is rapidly destroyed just like all other mRNAs - in fact the biggest technical hurdle with the vaccines has been developing a capsule to protect them for long enough so that they can produce enough antigen. The antigen itself is only part of the spike. It is not infectious. All this is basic immunology. The long term risk is insignificant and we have way more data on vaccine harm vs benefit that a knowledgeable person needs to justify the shot.
Hello Tomas, Thank you so much for your informative article. I live in the UK and have 2 children boy age 10 and girl age 9. They both currently have covid 19 (my husband and I are currently triple vaccinated and negative) so I want to know if you think giving our kids a first Pizer vaccine in time would benefit them since they have caught omnicron and also if there is any data yet on reproductive issues from the vaccine? Kind regards and many thanks
I unfortunately don’t know that, sorry.
I hope they get better soon!
Fantastic article! It is thorough, factual, prescriptive.
TY for your contributions to public health!
Thanks, Tomas. Super clear and well explained, like all your articles. I have one question, what about babies <1 year old? Currently pregnant, I got my three vaccine shots (2 pfeizer and the third, moderna) so my baby should have some kind of immunity when he’s born, but to what extent will he be protected? Is there any data/report on protection on babies? Or plans if vaccines will be tested and available for them in the near future? Thanks a lot
No data on your specific case I’m afraid. Sorry :(
Great article. Thanks for posting. Quick question about the final bar chart: for apple to apple comparison, shouldn't you consider weighing the odds of getting covid? In other word, if you get the vaccine, you have 50/100K chances of getting side effects. If you do not get the vaccine, you have 600/100K chances of getting side effects conditioning on getting covid. Let's say that 20% of kids will get covid (I don't know what the right number is). Then, shouldn't you divide the covid bar by 5? Would the two bars be comparable then? Maybe I am missing something...
Yes. I mention it in the article. Based on Omicron and the evolution of variants, I put the likelihood of getting exposed to COVID in the next few months / years high enough that I didn't think it was useful to discount it. If you disagree, feel free to do exactly what you say!
ok. thanks!
I can't find your premium article on anti-vaxx arguments analysis. I am a subscriber
Coming later this week!
Have you written that article yet as I’m keen to subscribe for a month to read it (I can’t afford annual membership)
Here!
https://unchartedterritories.tomaspueyo.com/p/corrected-anti-vaxxing-conspiracy
is there a Spanish language version of this article coming?
No. If you translate it though I will circulate it
Thank you Tomas for another elegant summary. As I read it I kept thinking of writing a sister article called “Should You Face Your Fear of Vaccinating Your Kids?” Evolution has left us poorly equipped to parse nuance and uncertainty without feeling helpless and vulnerable. In a way the cognitive dissonance of vaccine deniers is akin to an overactive immune response to foreign information that threatens their worldview. I’d love to live in a binary black and white safe/unsafe world but this messy one is home. When things get tough we really do seem like emotional beings who think more than thinking beings who feel.
Very intelligent comment. You’re right, I think
Hi Tomas - Love your work, but I was surprised by the really small CFS percentages as they were markedly lower than the ~2-3% (across all ages) you reported in your Long Covid article. I went to the source of the data in the graph you presented here (https://www.ncbi.nlm.nih.gov/labs/pmc/articles/PMC6331450/) and it is actually NOT from covid patients. It was published in 2018 before covid. Please update or clarify when you get a chance.
Unfortunately AFAIK there’s no data on MECFS for kids from COVID so we have to make do with proxies
When the risk of serious harm AND major benefit is low, then minor harm and benefit become more important. For many parents the inconvenience of arranging the jab and pain from the needle outweigh the benefits. The indigenous population I work with solved this by vaccinating at schools. Everybody gets it, which is convenient and children see that it is no big deal. Of course the tribe really cares about its elders, unlike many Westerners.
Thanks for the analysis. What would be interesting to known to better weight the risk-benefit balance of COVID-19/vaccines in kids would be to compare the data you report with those from other respiratory infections of kids/vaccines. The flu for example, what is the incidence (or better what was in the pre-covid era) of complications/hospitalizations/death in kids between 5-12? In my country (Italy) the flu-vaccine in the pre-covid era was indicated only for the elderly (over 65 years of age) and my kid's pediatrician had never considered to suggest the flu-vaccine. If hospitalization and death incidence of the flu are somewhat similar to those of COVID-19, it would be a nice piece of extra information to add on the balance. There's still the protect the others reason that might be more relevant in this pandemic from a national health perspective, but when the (healthy) kids are your own the protect the others inevitably comes second (especially when you still have a relevant percentage of unvaccinated elderly that should protect themself).
So much work...
From a quick look I found this CDC page (https://www.cdc.gov/flu/about/burden/past-seasons.html).
Using influenza data from 2016-2017, the most recent year with validated data (https://www.cdc.gov/flu/about/burden/2016-2017.html), it seems that the figures are similar (per 100,000 infected):
- 0-4 yrs, hospitalization: 697 flu vs 746 covid
- 0-4 yrs, death: 5.3 flu vs 1.9 covid
- 5-17 yrs, hospitalization: 274 flu vs 421 covid
- 5-17 yrs, death: 1.9 flu vs 1.1 covid.
Actually, flu numbers are based per 100k symptomatic ilnessess not infections so numbers are probably worse than COVID, which is often asymptomatic in this age range and unrecognized.
Unfortunately, flu statistics numbers pull together 5-11 and 12-17 years of age.
If I did the maths right, this supports the conclusion that vaccinating kids against COVID is primarily a matter of public health. I am not saying it is not right, I am just saying it is not really about the kids (as lots, at least in Italy, are saying).
Thx.
Yeah so I’m not sure we’re far away from the ball park. But there’s a structural issue with comps with the flu: the denominator.
The flu’s underascertainment is very unlikely to be at the “heavy symptomatic” level, because ppl in those cases do go to the doctor. So if you have 100k cases and 600 deaths, what you really have is 100k heavy symptomatic cases and 600 deaths, but, say, 500k infections and 600 deaths. Your fatality rate is going to be, say, 5x lower.
This doesn’t happen with COVID because we have massive monitoring and population-wide analyses. So we have a much better sense of the denominator.
Mmmm, I am not sure about that. Statistical models which take into account the underdetection of cases are used by the CDC for influenza surveillance (see methods on the same web page). In Italy the epidemiologic surveillance is based every year on the numbers communicated by every GP, pediatrician etc. and parents always take kids to the pediatrician for a cold and a fever, a flu in a child would not go unmissed (adults is surely a different thing). COVID instead go largely missed because it is asymptomatic at this age range in more than 50% of the cases (Italian health institute data) and we are completely unable at tracking and testing :-( (by we I mean in Italy). Anyway, I am positive that those number are not really comparable, they should be studied in depth, but I would not expect a huge difference in favor of one or the other. I am still inclined to think that the institutional communication in favor of the vaccination of children is mainly driven by the public health hence socio-economic burden of the pandemic.
Oh I absolutely agree with you that the institutional communication is for public health purposes. In fact, that's how I lead the article, right? You can't trust what they say because they have a public health interest in mind and have a track record of lying to achieve it. That's why I wanted to independently look at the numbers.
Agree also on the stats of flu vs. COVID. A deep dive would be valuable here. It's unclear where the data lies—and whether the conclusions you or I draw could be drawn!
Thx for your comments!
Your article is great but you should really take into account the fact that that the vaccine isn't based on the spike of the variants currently circulating. It protects very well against severe disease - and likely will keep doing so - because there are many epitopes on the prefusion spike it produces + cellular immunity adapts itself; but for kids, the focus should probably be more on long covid and/or potential concentration of spike protein in some organs (which means potential long term side effects, from the vaccine as well as from the infection) than on severe disease.
If they get infected easily after the vaccine (and they likely will with Omnicron, as you mention) they might develop long covid, as well as other potential post-covid long term effects, vaccinated or not. The odds of getting long covid might be lower, we don't know yet, nor do we know how lower it's gonna be, but we know that immune escape variants will get more and more likely. And that vaccinated kids who never met the virus (and have no cross immunity with an hCoV) will get infected at some point in the future.
After vaccine + infection (or infection + vaccine, or even infection only probably) the situation is not the same as there is some local immunity in the upper respiratory track and immunity, not focusing only on S is wider; with the vaccine only the only thing that will happen is that kids might recover quicker, and that's not even sure for those who would have recovered quicker anyway. Of course, for those who would have needed to go to the ICU the vaccine would have been a game changer. But we don't vaccinate kids against the flu. For those who go to the ICU, the vaccine would have been nice there as well...
Of course, for kids at risk of severe disease the answer is easy: get the vaccine asap. For the healthy ones the vaccine is probably a good choice, but it might be not that useful if it only provides protection against severe disease. In my opinion the question remains open because the risk of severe disease is pretty low, and the vaccine doesn't necessarily protects against infection side effects from the current - and next - variants.
Also, I don't know if a flu vaccine is offered to children in the US and how many get it! In Italy it was not until 2020
Thanks for the article. Would you change your mind looking at this latest research? (Nearly no protection for kids in the group age 5-11years after 1 month): https://www.cnn.com/2021/12/17/health/pfizer-vaccine-children/index.html
- Most kids have been exposed to omicron if they go to school, so the debate is even less important today than it was before. With natural immunity, you don’t need vaccines as much
- this effect is likely due to dosage more than anything else. Extremely unlikely that it works for all ages except for one small range