Caffeine intake → Sleep duration and regularity
Higher Caffeine intake moves Sleep duration and regularity only in some conditions. Read the mechanism.
In twenty young men, Weibel 2021 found that 450 mg of daytime caffeine a day left nighttime sleep length and stages unchanged, but delayed REM and made waking harder when sleep was pushed into the morning.
Randomized trial · Moderate · N=20
Supported by an experiment or a tightly identified design. Read the caveats.
Mechanism
Caffeine blocks adenosine A1 and A2A receptors. An evening dose can blunt homeostatic sleep pressure: longer sleep latency, less slow-wave activity. Habitual daytime use can adapt those homeostatic marks. Weibel, Lin, Landolt, Cajochen, and Reichert ran the same twenty men through caffeine, withdrawal, and placebo. Three 150 mg capsules a day for ten days. Last capsule about eight hours before habitual bedtime. On a habitual night (Scientific Reports, 2021), polysomnography did not move total sleep, latency, architecture, or how the night felt. Sigma power between 12 and 16 Hz fell on both caffeine and withdrawal nights. The authors read that as early overnight withdrawal, not deeper sleep. During the ambulatory week, bedtime sat about 25 minutes later on caffeine than on placebo. The companion recording (Journal of Biological Rhythms, 2021) started five hours after habitual bedtime, after a one-hour evening nap, around the circadian peak of REM. REM latency rose from about 54 minutes on placebo to about 79 on caffeine. REM accumulated later. Men reported more trouble waking and more tiredness. Total sleep and stages still did not differ. The sign depends on when you sleep. A habitual night in good sleepers who already drink coffee is not the same exposure as a delayed night.
Caveats
Twenty men, 18 to 35, habitual 300 to 600 mg a day, good sleepers. No women. One lab, one protocol. The nighttime file lost ten recordings to technical problems. One man was dropped for noncompliance. This page is adult daytime caffeine and nighttime or delayed sleep. It is not apnea of prematurity, not oolong tea plus fat oxidation, not caffeine plus zolpidem or trazodone, not slow-release caffeine after two days awake, not a tyrosine-phentermine-amphetamine stack, and not galanga for alertness. Evening caffeine close to bedtime is a different timing. These capsules stopped in the afternoon. The men already drank coffee; people who quit because it wrecks their sleep are not in the sample. This is not medical advice and not a recommendation to drink coffee, to quit, or to treat a number.
Effect
qualitative. Weibel Sci Rep 2021: nighttime TST, latency, architecture, and subjective quality not different after 9 days of 3x150 mg. J Biol Rhythms: REM latency ~54 vs ~79 min when sleep started ~04:22. Same 20 men.
Nodes
Named confounders
Cite this page
What Causes What. “Caffeine intake → Sleep duration and regularity.” https://whatcauseswhat.org/edges/e-caffeine-sleep (atlas updated 2026-09-02).
Papers
- The impact of daily caffeine intake on nighttime sleep in young adult men.
J Weibel, YS Lin, HP Landolt, J Kistler, S Rehm, KM Rentsch, H Slawik, S Borgwardt · 2021 · Scientific reports
- Regular Caffeine Intake Delays REM Sleep Promotion and Attenuates Sleep Quality in Healthy Men.
J Weibel, YS Lin, HP Landolt, C Berthomier, M Brandewinder, J Kistler, S Rehm, KM Rentsch · 2021 · Journal of biological rhythms