A caffeine cutoff after 2 pm: the cheapest hour of sleep you'll ever get back
Evidence is good enough to be worth trying for this claim. · Several RCTs, mostly consistent.
Caffeine six hours before bed still took about an hour off measured sleep — and people didn't feel it happening. Moving your last coffee to early afternoon is the single easiest 14-day experiment for anyone who sleeps badly.
The world's evidence
- Effect, in daily-life terms
- 400 mg of caffeine taken even 6 hours before bed cut measured sleep by about an hour in a controlled trial, and people under-noticed it; stopping by early afternoon removes most of that loss.
- Who was studied
- Controlled crossover trial (n=12) with objective sleep monitoring; consistent with caffeine's 5–6 hour half-life. Sensitivity varies widely between people.
Worth knowing
No risks. Expect a few days of afternoon dips while your body adjusts; decaf and tea before the cutoff still count toward the total.
Your evidence
Research says what happens on average. Log it for 14 days and Feelments shows what changed in your sleep, energy or mood — a pattern in your own logs, not a claim.
Sources
- Randomized trial (2013) — Drake et al., J Clin Sleep Med — 400 mg caffeine at 0, 3 or 6 h before bed all disrupted objectively measured sleep; 6 h before bed still reduced total sleep by ~1 h.
Last reviewed 2026-09-03. Verdicts are editorial assessments of published research, at the population level.
Feelments is for personal tracking and general wellness education. It does not provide medical advice, diagnosis, or treatment.