#66: A big coffee study won't change what I do....

18 snips

Mar 19, 2026

They scrutinize a big study linking two to three cups of coffee to lower dementia risk and explain why observational math can mislead. They debate puzzling patterns around tea, decaf, and caffeine. They compare randomized trials on heart rhythm, including findings on harmless irregular beats and surprising recurrence results. The takeaway: enjoy coffee if it suits you, but prioritize good sleep over chasing headlines.

Ask episode

AI Snips

Chapters

Transcript

Episode notes

ANECDOTE

Barista Machine Won't Be An HSA Expense

Bobby joked about wanting to deduct his barista machine as a medical expense if coffee prevented dementia.
He noted IRS rules disallowing HSAs for general wellness, making the thought humorous and impractical.

INSIGHT

Observational Coffee Study Shows Association Not Proof

Large observational cohorts can show associations but not causation for coffee and dementia.
Harvard JAMA analysis used 131,000 people over decades but relied on self-reported diet and studies not designed for coffee.

INSIGHT

Tea Matches Coffee But Decaf Fails To Explain Mechanism

Tea showed similar dementia benefits while decaf did not, creating conflicting clues about whether caffeine or other compounds matter.
Tea's lower caffeine and decaf's lack of benefit make it hard to pin the effect on any single component.

Get the Snipd Podcast app to discover more snips from this episode

Get the app

Send us Fan Mail

Visit my website

Bold claims make great headlines; clear evidence makes better habits. We take a hard look at the widely shared study suggesting two to three cups of coffee cut dementia risk by 20 percent and unpack what those numbers really mean for your brain and your daily routine.

First, we break down the Harvard JAMA research: massive cohorts of nurses and physicians, decades of follow-up, and self-reported diet data that carry real strengths and built-in limits. We explore why observational studies can’t prove causation, how confounders like sleep, exercise, and income can bend results, and why tea showing similar benefits while decaf shows none points to caffeine yet refuses a tidy explanation. Then we translate relative risk into absolute terms to show how a big percentage drop can still be a small difference in real life, and we discuss the publication bias that comes from testing many hypotheses and promoting only the eye-catching hits.

Next, we turn to trials where the science gets sharper. The CRAVE study randomized coffee days in healthy adults with continuous heart monitoring and found no rise in atrial abnormalities that lead to atrial fibrillation, though there was a bump in benign PVCs. For those with a history of AF, the DCAF trial offers a surprise: participants who kept drinking coffee had almost half the recurrence rate compared with those who quit, suggesting caffeine didn’t worsen outcomes and might even help. The message for most people is reassuring—coffee isn’t the arrhythmia trigger it’s often made out to be.

Our bottom line is practical and personal. If coffee fits your life and doesn’t wreck your sleep, enjoy one or two cups without expecting miracles. Protect your rest first, because sleep debt is a far clearer risk to cognition than a second espresso is a remedy. Stay curious, ask how a study was designed, and look for consistent results across methods before changing routines. If you learned something helpful, tap follow, share this episode with a friend who loves their morning brew, and leave a quick review to help others find the show.