11 Comments
User's avatar
SkinShallow's avatar

I'm not a neurologist OR a psychiatrist (tho sort of adjacent to/dabbled in some ways) so this is a more "constructionist" position, but your take really makes sense. The reserve/redundancy (cognitive and otherwise) must make a massive difference, and OBVIOUSLY it will be bigger in people with let's say more education and lower with people with emotional disorders (depression, PTSD) or, for that matter, hearing loss. Hearing loss also makes sense as the most visible one: language and its processing has A LOT of redundancy (we know how many letters can be missing from a text for it to be entirely legible; the effect is similar for speech, and it's also demonstrated by the fact that when listening to a language or even accent that's foreign to us, we typically need higher volume to pick everything up) but hearing loss removes this buffer. It will also affect social interactions.

And incidentally, I have my own (completely speculative) hypothesis for oft-touted "social contact being so very highly protective of dementia", which is that in addition to reverse causation you mention, there's also the fact that for the vast majority of older people, especially the very old, social interaction is likely the most cognitively demanding task they do. So it provides the training/stretching and building up of the reserve. It's not that "love" is necessarily better than writing philosophy articles, playing Bach, reading poetry or doing calculus, it's that very few elderly people do those latter things.

Expand full comment
Thomas Reilly's avatar

Agree! Also, because social interaction is cognitively demanding, I imagine those experiencing early symptoms of dementia start to become isolated as a consequence of their illness, rather than a cause.

Expand full comment
S. Rudd's avatar

Have you read Charles Pillers new book on Alzheimer's research? https://www.simonandschuster.com/books/Doctored/Charles-Piller/9781668031247

Expand full comment
Thomas Reilly's avatar

I haven’t, but it’s on my radar!

Expand full comment
S. Rudd's avatar

I haven't read yet but it's being shipped. I've heard good of it.

Expand full comment
Joshua E. Foster-Tucker's avatar

The “hearing aids-as-dementia-prevention” was on my comp exam last May! 🙃 We needed to figure out a way to study it whilst avoiding reverse causation and using a health system’s EHRs—reverse causation is the perennial problem of life course epi lol

Expand full comment
Joshua E. Foster-Tucker's avatar

Oh, you’d be surprised! Causal inference is a specialty in epi. If you’ve read a target trial emulation, you’ve seen causal inference using observational data! Technically, randomized-controlled trials are observational studies, too! 🥲

Expand full comment
Thomas Reilly's avatar

Not a topic I know much about clearly!

Expand full comment
Thomas Reilly's avatar

It’s very tricky to get at causation using observational data isn’t it.

I think the ‘natural experiment’ type studies come closest

Expand full comment
Shane Horan's avatar

Really good article. It makes it clear how hard it is to study dementia as a single concept, when it's mixing in Alzheimer's (already poorly understood, https://open.substack.com/pub/confidenceinterval/p/the-alien-er-and-the-amyloid-hypothesis), vascular disease, and probably some other distinct age-related disease process.

Expand full comment
Thomas Reilly's avatar

Thanks for sharing your post, I enjoyed it!

Expand full comment