Flu vaccinations associated with reduced T3D risk

(Joey) #1

An interesting epidemiological study finds statistically meaningful association between older adults getting annual flu vaccines with reduced risk of Alzheimer’s/dementia (aka Type 3 Diabetes).

The presumed mechanism for this finding is the reduced incidence of inflammation from microbial attack of neurological tissues as a result of the protective effect of flu vaccines.

Inasmuch as many carb-restricting eaters are intent on reducing various forms of inflammation, I thought this paper might be of some interest around here…

https://content.iospress.com/articles/journal-of-alzheimers-disease/jad220361

(Robin) #2

Thanks for sharing this! While dealing with covid brain and doing all the labs, mri’s etc, my doc said the best thing I have going for me is an almost non-existent level of inflammation. .05 on a scale of 1-30.
My joints will attest!

(Bob M) #3

Could be solely due to the healthy user effect.

(Joey) #4

@ctviggen That was my first thought too (being a healthy skeptic). The study asserts that it has normalized for comorbidities, medication usage, and demographics.

Presumably that was in an effort to eliminate the confounding effects of higher vaccine acceptance by the more health-conscious subjects.

(Bacon is a many-splendoured thing) #5

It’s virtually impossible to eliminate the possibility of confounding. The only real way to be sure would be to conduct a randomised controled trial.

(Joey) #6

No disagreement here. Epidemiology is an inherently limited path toward science. Yet, while no match for RCT-based research, it can still point to interesting relationships worth further exploration.

For example, correlation and association reveals this relationship, featured in Forbes magazine…

10%20PM
Screen Shot 2022-06-26 at 12.47.10 PM.jpg1620×1160 336 KB

(Bob M) #7

I can’t find the study, but there was a theory that the flu vaccine saved 50% lives than without the flu vaccine. That is, people who get the flu shot versus people who don’t. Someone did an epidemiological study, but also looked at deaths in summer, when there was no flu. What they found that getting a flu shot was really a marker of being healthy: the sick folk who got ill/died in the summer didn’t get a flu shot in the winter/fall.

I do think it’s an interesting study, and I saw one where people who got the flu shot had less heart disease. They had theories about why, but it’s always hard to disentangle.

(Bacon is a many-splendoured thing) #8

My favourite correlation is between the divorce rate and margarine consumption in the state of Maine. If only those Mainers would stop getting divorced, they could go back to eating butter again!

My next favourite is the correlation between shoe size and reading comprehension scores. If you want to read better, get bigger shoes!

#9

Definitely worth a follow up trial. Did the research group register a follow up investigation?

#10

The kinds of bias (and ability to control for them) are entirely different in “ecological studies” (where the unit of observation is the group) compared with individual level studies (where measurement of exposure and outcome is on the individual). If this community could get that right, the arguments I read on this site would be much stronger.

(Joey) #11

Several endnotes cited the need for further investigation, but no mention of intent by the authors to pursue any particular study.

However, this note acknowledging the need to further reduce confounding influences was offered:

“Several lines of future investigation are warranted. A follow-up study using a prospective cohort design would provide more robust reduction of confounding.”

(Joey) #12

Fascinating point. Would this help folks draw clearer distinctions between whether low-carb is generally a healthier WOE than high-carb, while allowing for the many n=1 variations and metabolic sensitivities within each group?

#13

It would help us fairly evaluate the epidemiologic evidence that people share here. No respectable epidemiologist would consider ecologic studies strong evidence of anything (it’s the “ecological fallacy”), so bringing up examples of those is not helpful. On the other hand, giving examples of specific uncontrolled or incompletely controlled confounding in an individual-level study, or the likelihood of selection bias (mentioned above), is very helpful.

(Joey) #14

This does make me wonder … Maybe I don’t quite understand what you mean by “ecologic” studies.

If no respectable epidemiologist would consider such studies as meaningful evidence of anything, how do epidemiologists get such work through the peer review process?

#15

The flu vaccination study is a cohort study (follows individuals with known exposure status for the disease outcome in the individual), while the pirate example is ecological. Any ecological study that gets through peer review these days (in a reputable journal) would be considered “hypothesis-generating”. There’s nothing necessarily wrong with that. Science can get ideas to study from patterns in people (many n’s of 1) or variation across groups. Then we test those patterns to see if they hold true. RCTs are the paradigm for observational epidemiology. They are (supposed to be) what observational epidemiology studies aim to emulate (through appropriate selection of comparison groups and control for confounding).

(Joey) #16

@Wendy198 Thanks for the helpful clarification.

Decades after enjoying courses in statistics, econ modeling and biometrics, I recently came across a delightful ditty by Alex Reinhart. Might be a fun read if you haven’t already seen it…

https://www.statisticsdonewrong.com/

(Bob M) #17

You’re never get RCTs for this. No one will pay.

(Joey) #18

Interestingly enough, my wife and I were paid with retail gift cards to get the annual flu vaccine by our health insurance company. No match for lab rats, but it was still fun to go shopping at Target with our band-aids and bounty.

(Bob M) #19

Though if you got it this year (and I got my first flu vaccine – in decades – this past season), it was supposedly a terrible match with circulating virus.

I just think that unless the government gets involved, there’s no benefit to a flu vaccine manufacturer to doing these studies. I think the idea that the flu could affect heart disease, for instance, is really interesting and could be true. I just think there’s not a lot of benefit for flu manufacturers.

Though I could be wrong:

Some benefit there.

Not enough information here:

Hmm…I think I am wrong:

Actually an RCT, with a beneficial result that doesn’t appear to be made up (where they go in and look at subgroups or combined endpoints).

(Joey) #20

Wow, a double blind placebo vs treatment study on live human beings. This is pretty good stuff. :+1: