Duan · Nature and science of sleep 2021 · Randomized crossover trial (post hoc analysis) · n=20

Effects of Dinner Timing on Sleep Stage Distribution and EEG Power Spectrum in Healthy Volunteers.

Cited 31 times in the scientific literature.

Level 2 - randomized trial

Post hoc analysis of an individual randomized crossover trial

PubMed 34017207 · doi:10.2147/NSS.S301113 · record verified 2026-08-26

What was done

In a post hoc analysis of a randomized crossover study, 20 healthy adult volunteers (10 women) were tested under two conditions in a laboratory setting with a fixed sleep opportunity: routine dinner (RD, 5 hours before bedtime) versus late dinner (LD, 1 hour before bedtime) using an isocaloric meal. Sleep architecture was measured with overnight polysomnography, visual sleep staging, and quantitative EEG spectral power analysis (alpha, beta, delta, and theta bands) evaluated with mixed-effect spline regression.

What was found

Conventional sleep stages showed no significant differences between the routine and late dinner conditions. However, late dinner produced a 2.5% initial increase in delta power and a reciprocal 2.7% decrease in combined alpha and beta power (p < 0.0001) in the early part of the night. These spectral differences diminished over time and reversed during the latter part of the night.

Why it matters

These findings suggest that eating close to bedtime does not inherently impair gross sleep architecture in healthy adults, instead causing a transient shift toward deeper sleep early in the night followed by lighter sleep later.

Limits

The study is limited by a small sample size (n = 20) and its post hoc nature. It tested healthy individuals in an artificial laboratory environment with an acute one-night exposure, so findings cannot be generalized to chronic late-eating habits, free-living conditions, or individuals with gastrointestinal, metabolic, or sleep disorders.

Cited by