Hacker News Reader

Comment by mapt | Hacker News Reader

This is a distinct claim. Caffeine can disrupt sleep even 48 hours later, a little bit; It is traditionally modelled with an elimination half-life of 5 hours, meaning 1/2 effect at 5 hours, 1/4 effect at 10 hours, 1/8 effect at 15 hours, an exponential decay curve.

The claim being made is that due to cascading decay of a secondary metabolite that does a lot of the work producing the clinical effect, caffeine elimination is a much more linear, slow process that only reaches half effect at around 10 hours and 1/4 effect at 17 hours, 1/8 effect at 23 hours.

View on HN

Just because the halflife leaves a measurable amount in your system, doesn't mean that that amount is enough for measurable outcomes.

In your example, a 200 mg caffeine intake in the morning, least to 100mg at noon, 50mg at 5PM, 25mg at 10PM. Yes that means you still have 25mg of caffeine. But it's unlikely to have an outcome you can measure since it's below a minimum threshold.

This depends on your genetics - there are different groups of caffeine metabolizers. I'm in the group that's super sensitive to caffeine and I can feel effects from less than 20mg.

Sure, but you're consumption is probably lower.

This may explain why cold turkey effects for me take 24h to start

I went decaf drinks only back in 2024. I was fine and thinking “what’s the big deal” until day three. I will never forget that day. So horrible.

Still decaf only. Has been a pretty positive change for me. Kicked the soda habit completely. Sleep is better. I find I’m even all day. I generally only get tired when I’m bored.

Decaf only is great for coffee lovers. Instant decaf is of course just a way to spoil perfectly good hot water. But decaf beans from a good roaster are good then grind and make what you desire - espresso, pour over, french press, moka etc.

Maybe preground OK for non-espresso too?

"Caffeine can disrupt sleep even 48 hours later, a little bit;"

The literature for AASM protocols suggests 41 hours without caffeine is enough to safely control for those potential effects

The point is to communicate that it's not a steep step change. In the original caffeine-only model, it's a smooth exponential decay function. There is still measurable caffeine in your blood N hours later, at levels approximately 1/(2^(N/5)) as high as the peak concentration.

In the new model, it's still smooth decay, but it's a compound exponential decay which is spread out over a longer time period, and close to linear for a while after that, before going on a longer exponential decline.

Reply on news.ycombinator.com