Well I will fully admit that I didn't read the article and only glanced at the headline on Medscape. But like Vinay, I immediately realized that the study had to be complete BS for exactly the same reasons so eloquently put by Vinay. What is more surprising is how this paper ever got through the peer reviewed process in a serious medical journal.
On another note, I'm also of the view that the current Shindrix vaccine is highly reactogenic and quite frankly bad news (with something like 50% having a sufficiently bad adverse reaction that they cannot or don't feel like going to work the next day). There was no need to replace the original chicken pox vaccine for shingles which was more than sufficiently effective for a vaccine that conferred a minimal if any advantage (other than to make money).
Thanks, Vinay, for bringing attention to this paper. In reading through it, I encountered a few pretty egregious abuses of statistics. To wit:
1. I really don't like their use of the term "causal evidence." It seems very redundant, and every use of the term could just be replaced with "good evidence" or even "evidence." When you're trying to establish that a virus causes dementia, all evidence for the hypothesis is, by definition, causal evidence. I think what they're really trying to do is convince you that a regression discontinuity is just as good as prospective randomized data. But it's not, and no amount of repeated assertion will change that.
2. The paragraph on line 214 begins this way: "Using our regression discontinuity approach, we find that being eligible for the zoster vaccine caused a 1.3 (95% CI: 0.2 – 2.7; p=0.022) percentage point absolute, and 8.5% relative, reduction in the probability of a new dementia diagnosis over our seven-year follow-up period (Fig. 3, Panel A). Scaled to account for the fact that not all those who were eligible received the vaccine, we find that actually receiving the zoster vaccine reduced the probability of a new dementia diagnosis by 3.5 (95% CI: 0.6 – 7.1; p=0.019) percentage points, corresponding to a relative reduction of 19.9%." Am I missing something, or did they just give up the whole ballgame here? The whole point of this study is that they have found a pseudorandom comparator in the form of the arbitrary cutoff for vaccine eligibility. But from this point on in the paper, their relative risk reduction is extrapolated from the subset of people who chose to get the vaccine. You simply can't do that. It completely invalidates the claim of being a pseudorandom ("causal") dataset. People who choose the vaccine are different in important, nonrandom ways from those who don't. Granted, there is still an effect size in the population as a whole, but it's less than half as large (an 8.5% relative reduction and a 1.3% absolute one). I think it's much easier to chalk this smaller benefit up to noise or unmeasured confounding than the much more impressive 19.9% relative risk reduction they trumpet.
3. Last, and worst of all, is this attempt to explain the huge problem with their theory presented by the total lack of effect in men on line 313: "In fact, among men, the point estimates were close to zero across all specifications. Nonetheless, the 95% confidence interval for the effect among men included the possibility of a relative protective effect on dementia over a seven-year follow-up period of up to 23.9%. We can, therefore, not exclude the possibility that the vaccine also had a protective effect on dementia among men." What???? So your point estimate of the effect is zero, but the confidence interval includes some territory where there's a benefit? What kind of claim is that? That is true of ANY confidence interval centered on zero. The fact that it reaches as high as a 24% benefit just tells us that your study is under-powered, and nothing more. If I flip a coin 10 times and get 5 heads, the point estimate is a fair coin, but the confidence interval includes the possibility of getting 6 heads, which would be consistent with a coin biased towards heads. This is technically true but it's an utterly silly point.
Having off target effects that are beneficial is not of itself disqualifying. GLPs are diabetes drugs that result in quite a bit of weight loss. SGLTs are diabetes drugs that are effective in secondary prevention and HFrEF. A vaccine helps dementia? Sure, why not? VZV hangs out in the nervous system (peripheral, mind you)….And Alzheimer’s is amyloid plaques (if you buy that theory) in the nervous system (central, mind you)….but…could happen…we just need to see some proof.
But the galling aspect of that tweet is to suggest they had “CAUSAL” proof of such an effect. How on earth have they done that?
Point #3, "The size of the effect is bigger than the effect on preventing Shingles itself, in absolute terms!" does not strike me as a problem. If the baseline risk of dementia is higher than that of shingles (which I would expect), then the absolute risk reduction for dementia could be bigger even with a lower relative risk reduction and hence weaker evidence for causality.
It seems today everyone has an agenda or a narrative. It didn’t seem to be this way 30 plus years ago but maybe I was more naive in my younger days, however not critically reading something and accepting the abstract conclusions is mind boggling. No wonder the general public thinks PHDs and MDs are full of it
Hopefully there is some small subset of those who retweeted it and a much larger subset of those who read the preprint who didn't buy into the b*******.
This subject is a sore spot for me, after the second Pfizer dose, my mouth erupted in varicella sores, my headache was unbearable. I had to go on Valacyclovir, but it didn't help. Every time I get my titers tested, the virus is off the charts, running rampant in my system ever since 😢
Vinay, This is spot on. You get "smarter" every day -- or at least more eloquent in transmitting the truth and more honest in dropping your own, once obvious, political biases. As an independent, it is clear that the Democrats have fallen down a very deep hole with regard to Scientism...there are not many besides you trying to point this out.
Good commentary. This case of scientists retweeting an implausible association without the slightest effort at critical thinking is telling and gives strong support for point 4.
At 44, I got shingles this past Easter - in my head/eye. The worst headache I’ve had in my entire life. Wanted to blow my brains out by day 2, assumed I had an aggressive brain tumor which would kill me in days. No doctors could figure out what was happening... and then the rash appeared on day 4. Mystery solved.
What was interesting- and I don’t know if just BS, was my ophthalmologist suggested the varicella vaccine was responsible for so many young healthy people now getting shingles.
He said when he was in med school decades ago, a young person getting shingles was practically unheard of. But now that adults aren’t constantly exposed to children with chicken pox anymore, they aren’t getting naturally boosted.
Wonder if true? 3 more friends all in 30’s and 40’s also got shingles within the month I did.
I will tell you in the 90's when varicella vaccine was rolling out, Infectious disease experts at academic institutions were quite certain that we would see more shingles and in younger ages. When Varicella was circulating in the community, the exposure kept the virus dormant. Only older and/ or stress/ immune compromised would reactivate. This was not a theory. It was predicted and understood to be the consequence of universal varicella vaccination. And the reason we were putting Varicella into the childhood immunization schedule was to save money. We had mostly 2 parent working families, and it cost too much money (for businesses) or parents would lose money (if no paid time off) if 2-3 kids were home for 5+ days each with chicken pox. Chicken pox was thought to be one of beneficial childhood diseases in pediatrics. The vaccine could have been given just to those who had not got chicken pox before puberty.
I got it at 35, V3 type distribution (my lower jaw up to the edge of my ear canal). The stress I was under was extreme and sustained, but I didn't believe the diagnosis at first. I'm not old enough, I said.
I had zero stress. Was on vacation enjoying life. BAM. Was certain I was dying of stroke or hemorrhage. But then I kept living... assumed must be aggressive tumor. So incredibly weird.
Yes. This is true. Great Britain even considered not recommending it to kids because it would cause an increase in shingles in the elderly, but they did anyways.
Thanks for posting that! I looked into it maybe 10+ years ago and at the time thought that they had gone ahead and pushed for mass vaccination of kids. I'm glad they've backed off of that.
Interesting question. I got chickenpox in my 20’s from my son who brought it home in kindergarten. That was in California long before the vaccine was created. I’ve never been so sick as that (covid was nothing in comparison) but in my titre remains sky-high to this day. I’ll never take the shingles vaccine.
Saying that there couldn't possibly be a link is also not letting science do its thing. I can think of two explanations - one is that the vaccine crosses the blood brain barrier and stimulates the brain's immune system which somehow clears out plaques or something that is related to dementia.
A second possibility is that those with dementia are being killed by the vaccine.
I'm not a fan of the shingles vax - it could be avoided if we let kids have natural chicken pox and adults got reinforced immunity. It pisses me off that I will never have that opportunity because California mandates the vaccine for school. But I don't think that you can assume there is no plausible explanation and you need to start by seeing if the statistics do turn out to be significant and then asking "Why?"
PS: Why *shouldn't* there be differences between men and women?
Isn't it just rather ... bloody ... depressing?! (rhetorical).
Well I will fully admit that I didn't read the article and only glanced at the headline on Medscape. But like Vinay, I immediately realized that the study had to be complete BS for exactly the same reasons so eloquently put by Vinay. What is more surprising is how this paper ever got through the peer reviewed process in a serious medical journal.
On another note, I'm also of the view that the current Shindrix vaccine is highly reactogenic and quite frankly bad news (with something like 50% having a sufficiently bad adverse reaction that they cannot or don't feel like going to work the next day). There was no need to replace the original chicken pox vaccine for shingles which was more than sufficiently effective for a vaccine that conferred a minimal if any advantage (other than to make money).
Thanks, Vinay, for bringing attention to this paper. In reading through it, I encountered a few pretty egregious abuses of statistics. To wit:
1. I really don't like their use of the term "causal evidence." It seems very redundant, and every use of the term could just be replaced with "good evidence" or even "evidence." When you're trying to establish that a virus causes dementia, all evidence for the hypothesis is, by definition, causal evidence. I think what they're really trying to do is convince you that a regression discontinuity is just as good as prospective randomized data. But it's not, and no amount of repeated assertion will change that.
2. The paragraph on line 214 begins this way: "Using our regression discontinuity approach, we find that being eligible for the zoster vaccine caused a 1.3 (95% CI: 0.2 – 2.7; p=0.022) percentage point absolute, and 8.5% relative, reduction in the probability of a new dementia diagnosis over our seven-year follow-up period (Fig. 3, Panel A). Scaled to account for the fact that not all those who were eligible received the vaccine, we find that actually receiving the zoster vaccine reduced the probability of a new dementia diagnosis by 3.5 (95% CI: 0.6 – 7.1; p=0.019) percentage points, corresponding to a relative reduction of 19.9%." Am I missing something, or did they just give up the whole ballgame here? The whole point of this study is that they have found a pseudorandom comparator in the form of the arbitrary cutoff for vaccine eligibility. But from this point on in the paper, their relative risk reduction is extrapolated from the subset of people who chose to get the vaccine. You simply can't do that. It completely invalidates the claim of being a pseudorandom ("causal") dataset. People who choose the vaccine are different in important, nonrandom ways from those who don't. Granted, there is still an effect size in the population as a whole, but it's less than half as large (an 8.5% relative reduction and a 1.3% absolute one). I think it's much easier to chalk this smaller benefit up to noise or unmeasured confounding than the much more impressive 19.9% relative risk reduction they trumpet.
3. Last, and worst of all, is this attempt to explain the huge problem with their theory presented by the total lack of effect in men on line 313: "In fact, among men, the point estimates were close to zero across all specifications. Nonetheless, the 95% confidence interval for the effect among men included the possibility of a relative protective effect on dementia over a seven-year follow-up period of up to 23.9%. We can, therefore, not exclude the possibility that the vaccine also had a protective effect on dementia among men." What???? So your point estimate of the effect is zero, but the confidence interval includes some territory where there's a benefit? What kind of claim is that? That is true of ANY confidence interval centered on zero. The fact that it reaches as high as a 24% benefit just tells us that your study is under-powered, and nothing more. If I flip a coin 10 times and get 5 heads, the point estimate is a fair coin, but the confidence interval includes the possibility of getting 6 heads, which would be consistent with a coin biased towards heads. This is technically true but it's an utterly silly point.
Having off target effects that are beneficial is not of itself disqualifying. GLPs are diabetes drugs that result in quite a bit of weight loss. SGLTs are diabetes drugs that are effective in secondary prevention and HFrEF. A vaccine helps dementia? Sure, why not? VZV hangs out in the nervous system (peripheral, mind you)….And Alzheimer’s is amyloid plaques (if you buy that theory) in the nervous system (central, mind you)….but…could happen…we just need to see some proof.
But the galling aspect of that tweet is to suggest they had “CAUSAL” proof of such an effect. How on earth have they done that?
How is this even biologically plausible?
We are already infected with varicella the vaccine doesn't change that fact.
Point #3, "The size of the effect is bigger than the effect on preventing Shingles itself, in absolute terms!" does not strike me as a problem. If the baseline risk of dementia is higher than that of shingles (which I would expect), then the absolute risk reduction for dementia could be bigger even with a lower relative risk reduction and hence weaker evidence for causality.
Does the shingles vaccine not work to prevent shingles?
It seems today everyone has an agenda or a narrative. It didn’t seem to be this way 30 plus years ago but maybe I was more naive in my younger days, however not critically reading something and accepting the abstract conclusions is mind boggling. No wonder the general public thinks PHDs and MDs are full of it
Hopefully there is some small subset of those who retweeted it and a much larger subset of those who read the preprint who didn't buy into the b*******.
I didn't read Vinay's article, but I'm sending to friends.
hahahahahahahah
Sure doesn't seem to have the noted positive effect for Senator Diane Feinstein. Just saying.
This subject is a sore spot for me, after the second Pfizer dose, my mouth erupted in varicella sores, my headache was unbearable. I had to go on Valacyclovir, but it didn't help. Every time I get my titers tested, the virus is off the charts, running rampant in my system ever since 😢
Ugh. Try high-dose vitamin D. (10000IU - 50K IU). Looks for Dr. Eric Berg's videos on Shingles. For example: https://www.youtube.com/watch?v=E4ghFFE2Czc
I love Dr. Berg, yes I’m trying that :)
Vinay, This is spot on. You get "smarter" every day -- or at least more eloquent in transmitting the truth and more honest in dropping your own, once obvious, political biases. As an independent, it is clear that the Democrats have fallen down a very deep hole with regard to Scientism...there are not many besides you trying to point this out.
Kudos.
Good commentary. This case of scientists retweeting an implausible association without the slightest effort at critical thinking is telling and gives strong support for point 4.
At 44, I got shingles this past Easter - in my head/eye. The worst headache I’ve had in my entire life. Wanted to blow my brains out by day 2, assumed I had an aggressive brain tumor which would kill me in days. No doctors could figure out what was happening... and then the rash appeared on day 4. Mystery solved.
What was interesting- and I don’t know if just BS, was my ophthalmologist suggested the varicella vaccine was responsible for so many young healthy people now getting shingles.
He said when he was in med school decades ago, a young person getting shingles was practically unheard of. But now that adults aren’t constantly exposed to children with chicken pox anymore, they aren’t getting naturally boosted.
Wonder if true? 3 more friends all in 30’s and 40’s also got shingles within the month I did.
I will tell you in the 90's when varicella vaccine was rolling out, Infectious disease experts at academic institutions were quite certain that we would see more shingles and in younger ages. When Varicella was circulating in the community, the exposure kept the virus dormant. Only older and/ or stress/ immune compromised would reactivate. This was not a theory. It was predicted and understood to be the consequence of universal varicella vaccination. And the reason we were putting Varicella into the childhood immunization schedule was to save money. We had mostly 2 parent working families, and it cost too much money (for businesses) or parents would lose money (if no paid time off) if 2-3 kids were home for 5+ days each with chicken pox. Chicken pox was thought to be one of beneficial childhood diseases in pediatrics. The vaccine could have been given just to those who had not got chicken pox before puberty.
The vaccine could have been give to those who did not get it by the time they hit puberty, after which there are more complications.
I got it at 35, V3 type distribution (my lower jaw up to the edge of my ear canal). The stress I was under was extreme and sustained, but I didn't believe the diagnosis at first. I'm not old enough, I said.
I had zero stress. Was on vacation enjoying life. BAM. Was certain I was dying of stroke or hemorrhage. But then I kept living... assumed must be aggressive tumor. So incredibly weird.
Yes. This is true. Great Britain even considered not recommending it to kids because it would cause an increase in shingles in the elderly, but they did anyways.
If any chance you could toss a source on that would be fascinating to read about.
EDIT
Found it!
https://patient.info/news-and-features/should-your-child-have-the-chickenpox-vaccine
This is 2019
Thanks for posting that! I looked into it maybe 10+ years ago and at the time thought that they had gone ahead and pushed for mass vaccination of kids. I'm glad they've backed off of that.
It is likely true imo. Many smart people warned of it before the vaccines were approved
Interesting question. I got chickenpox in my 20’s from my son who brought it home in kindergarten. That was in California long before the vaccine was created. I’ve never been so sick as that (covid was nothing in comparison) but in my titre remains sky-high to this day. I’ll never take the shingles vaccine.
Saying that there couldn't possibly be a link is also not letting science do its thing. I can think of two explanations - one is that the vaccine crosses the blood brain barrier and stimulates the brain's immune system which somehow clears out plaques or something that is related to dementia.
A second possibility is that those with dementia are being killed by the vaccine.
I'm not a fan of the shingles vax - it could be avoided if we let kids have natural chicken pox and adults got reinforced immunity. It pisses me off that I will never have that opportunity because California mandates the vaccine for school. But I don't think that you can assume there is no plausible explanation and you need to start by seeing if the statistics do turn out to be significant and then asking "Why?"
PS: Why *shouldn't* there be differences between men and women?