alt.hn

2/9/2026 at 11:29:51 PM

Coffee and Tea Intake, Dementia Risk, and Cognitive Function

 https://jamanetwork.com/journals/jama/article-abstract/2844764

by bookofjoe

2/10/2026 at 3:45:52 PM

Per [0] risk factors for dementia are "lower levels of education, high blood pressure, hearing impairment, smoking, obesity, depression, physical inactivity, diabetes, and low social contact."

A recreational stimulant often taking in social settings seems to hit a lot of those.

The interesting one is blood pressure. Caffeine would spike it, but I have long wondered if "exercising" the vasculature with transients is a good idea and long-term elevation is the real killer. Like school fire drills.

[0]: https://en.wikipedia.org/wiki/Dementia

by randusername

2/10/2026 at 10:55:39 PM

It is correlation then, not causation. Wouldn’t be surprised if this study was funded by corps. to promote those stimulants, like the Harvard and sugar or red wine everyday studies.

PS: Don’t trust in “no conflict of interest” claims

by mgh2

2/10/2026 at 7:07:31 AM

> higher caffeinated coffee intake was significantly associated with lower dementia risk (141 vs 330 cases per 100 000 person-years comparing the fourth [highest] quartile of consumption with the first [lowest] quartile; hazard ratio, 0.82 [95% CI, 0.76 to 0.89])

That's a very big difference.

by marticode

2/10/2026 at 10:46:36 AM

Doing lots of cognition heavy tasks also seems to correlate with lower dementia risk; Basicly the premise of "brain workout" apps. Thought I think if you're going to be doing brain exercises, that time would be better spent learning a new langauge or learning math or physics or art.

Incidently, People who study or do a lot of thinking heavy deep work also tend to be coffee addicts. coincidence?

by cultofmetatron

2/10/2026 at 4:31:53 PM

> Incidently, People who study or do a lot of thinking heavy deep work also tend to be coffee addicts. coincidence?

Coffee can be quite expensive, and thinking work that requires a lot of education tends to correlate with higher income, so I wonder if coffee drinking just acted as another wealth indicator.

by Qem

2/10/2026 at 5:13:28 PM

You can get very, very cheap coffee.

Good coffee can be expensive, I agree with that.

by galleywest200

2/10/2026 at 1:29:01 PM

Looking at the chart the difference is even larger. The cohort quartiles have different drinks/day. looking at just the NHS cohort the difference between 0 and 4.5 drinks/day was an incident rate of 354 to 95 per 100k person-years.

by ticulatedspline

2/10/2026 at 10:48:21 AM

I wonder if this translates to other stimulant use like amphetamines.

The positive effect may be negated by hypertension induced by said stimulants, which would then lead to organic dementias.

by xattt

2/10/2026 at 7:27:52 AM

So a 100% reduction in risk? Or is it 50% reduction in risk.

by dyauspitr

2/10/2026 at 7:24:15 PM

I have to remind myself to stay within the two to three cups a day recommendation.

There was a study in 2021 that found that drinking more than six cups of coffee a day was associated with a 53% increased risk of dementia and smaller total brain volume.

https://cardiologycoffee.com/blogs/news/new-study-says-coffe...

by speak_plainly

2/10/2026 at 1:49:16 PM

Skimming the study leaves some weird results.

- There's a sharp decrease in incidence between 2.5 drinks/day and 4.5 (3rd/4th quartile)

- Technically decaffeinated coffee actually had more dementia cases in the NHS cohort

- The highest decaf group is 1 drinks/day

- They didn't track the kinds of tea people were drinking (black, green, herbal)

- Drinking one cup of tea seems nearly as effective as 2.5 cups of coffee.

while it seems they were controlling for caffeine intake the tea vs coffee groups would have had wildly different intake with similar results.

also I couldn't tell if they tracked if the people drinking 40oz of coffee a day simply dropped dead before they could go crazy

by ticulatedspline

2/10/2026 at 3:22:54 PM

The only thing that feels better than a caffeine buzz is having my biases confirmed. Thanks, JAMA! Can I have mine with cream, sugar, and an hour of CME credit?

by overtone1000

2/10/2026 at 1:06:58 PM

https://en.wikipedia.org/wiki/Gustav_III_of_Sweden%27s_coffe...

Needs an obligatory mention

by srean

2/10/2026 at 1:22:56 PM

> The tea drinking twin died first at the age of 83, long after the death of Gustav III, who was assassinated in 1792.

So

being King << drinking Tea < drinking Coffee

by elcapitan

2/10/2026 at 3:05:11 PM

Yeah. Being king or an heir to the throne is certainly very hazardous.

by srean

2/10/2026 at 8:31:18 AM

> The most pronounced associated differences were observed with intake of approximately 2 to 3 cups per day of caffeinated coffee or 1 to 2 cups per day of tea.

Weird, tea is supposed to have half the caffeine of coffee.

by krzat

2/10/2026 at 9:01:53 AM

Both coffee and tea have thousands of other constituent parts which could play a role.

The decaffeination process also removes more parts than just the caffeine.

by londons_explore

2/10/2026 at 4:05:27 PM

Anecdotally, I never feel more awake when having coffee. I can drink coffee at night and sleep great. On the other hand, if I try to sleep too soon after tea I feel very mentally awake and can’t sleep for restlessness. Maybe Qi is real!

by whaleidk

2/10/2026 at 1:02:51 PM

3 cups of coffee a day is way too many

by jdawg777

2/10/2026 at 3:09:01 PM

A cup of coffee has 95mg caffeine, and the FDA has long held 400mg as a reasonable stopping point for the day.

I guess I'm not really concerned about three cups of coffee in that case.

by ahhhhnoooo

2/10/2026 at 2:09:25 PM

For toddlers, absolutely.

by CoastalCoder

2/10/2026 at 4:28:02 PM

That's not even enough for doctors to consider "drinking coffee" on a chart lol.

by redwall_hp

2/10/2026 at 2:39:43 PM

I'm still spitballing here, but I speculate it's because caffeine increases calorie consumption*, reducing diabetes risk. Diabetes is a major risk factor for dementia.

*calorie expenditure

by ETH_start

2/10/2026 at 4:42:02 PM

Another possibility: Coffee intake is a proxy measurement for bitter taste tolerance, as many people take theirs without sugar. As healthy vegetables and low sugar foodstuff tend to be bitter, coffee and healthy food may be correlated to some measure.

by Qem

2/11/2026 at 2:44:31 AM

I'm trying to find an explanation for why people who drink caffeinated coffee have a much lower risk of dementia than people who drink decaffeinated coffee.

by ETH_start

2/11/2026 at 6:42:36 PM

Could be the case caffeine actually gets you addicted to bitter taste, instead of just indicating people with bitter taste preference, Pavlov style? Coffee is bitter, coffee is addictive. Perhaps this addiction to bitterness transfer to other foodstuff (vegetables). Decaffeinated coffee wouldn't allow the same transfer, by lack of the addictive ingredient.

by Qem

2/11/2026 at 6:18:00 PM

A: The cardiovascular impact of caffeine on blood circulation through the brain and its stimulation of the enteric nervous system.

by vaxman

2/10/2026 at 2:57:43 PM

Excuse the basic question- is that a typo, or do you really mean that higher calorie intake reduces diabetes risk?

If that is correct, why?

I thought higher calorie intake leads to weight gain which increases diabetes risk.

by testing22321

2/10/2026 at 3:19:04 PM

Sorry I meant calorie burning. Consumption in the sense of metabolizing it.

by ETH_start

2/10/2026 at 4:37:28 PM

Man anything can cause dementia huh.

by hiprob

2/10/2026 at 4:14:05 PM

The public communication around research like this is terrible.

> "2 to 3 Cups of Coffee a Day May Reduce Dementia Risk. But Not if It’s Decaf." - NYT

> "Daily cups of caffeinated coffee or mugs of tea may lower dementia risk." - Science News

"Reduce," "Lower" - this is all causal language for a study that is purely observational. The authors do a good job keeping causal language out of the paper, so why can't media do the same?

This leads to an environment where everyone knows that "correlation != causation," but almost nobody understands why.

by tfirst