Hacker News Reader

Comment by gdudeman | Hacker News Reader

This really isn't the case in the Shingles vaccine unless the UK study is flawed a way that isn't clear to me.

The study looked at the effect of being eligible for a vaccine and the results were clear. (see chart below the fold here: https://erictopol.substack.com/p/the-shingles-vaccine-and-re...)

There was a hard age cutoff in the UK study. Above a certain age, you weren't eligible. Below it, you were. People who were born in the "can get the vaccine" group have markedly lower rates of dementia. People in the "too old" group have higher rates. It's one of those studies where you don't even need to look at the p-value to see the difference.

I'm very open to being wrong about this!

View on HN

Thanks very much for linking that substack, very informative.

But in your summary:

> People who were born in the "can get the vaccine" group have markedly lower rates of dementia. People in the "too old" group have higher rates.

I would change "people" to "women". I thought it was very interesting that the benefit of the Shingles vaccine eligibility for Alzheimer's was largely confined to women - men showed no such benefit per the graph in that substack article.

The video explains why the analysis is wrong and gives a very clear graph of how when you look at the proper data, there’s no benefit from the shingles vaccine for dementia. The signal is clear as day.

FWIW Eric Topol is an extremely unreliable source. He has “fame” but most of his stories end up being wrong because of his poor analysis like the review above. I subscribed to him during the pandemic when he migrated to substack but ended my subscription after countless bad articles.

Right. And RCTs not showing things than can be seen in larger observational studies seems plausibly related to the problems with RCTs, for instance statistical power being generally lower because n sizes are lower, attrition is an issue, etc… I want to believe the research though, especially since I like most smart people are going to get the Shingrix shot anyway.

It’s the opposite. Observational studies are very biased by the selection process and the only way to get signal from them is rct.

So you have a 30 person RCT, half the treatment bails midway through, you include the assigned but not treated (ITT) in the analysis. That’s going to give you the power to make a definitive statement while a large scale longitudinal with n=50,000 will not?

Reply on news.ycombinator.com