Research paper
McHill AW, Phillips AJ, Czeisler CA, Keating L, Yee K, Barger LK, Garaulet M, Scheer FA, Klerman EB. Later circadian timing of food intake is associated with increased body fat. American Journal of Clinical Nutrition. 2017 Nov;106(5):1213-1219. doi:10.3945/ajcn.117.161588
Why we cite it
The authors report theirs is the first study, in their words ”to highlight eating patterns in real-world settings by using advanced food diary techniques and reveal that the timing of caloric consumption relative to melatonin onset is associated with increased body fat percentage”, prefaced ”our findings are the first, to our knowledge”. (Corrected 2026-09-10 from ”to measure food timing relative to individual circadian phase (melatonin onset) rather than clock time”, which is broader than the claim the authors make.) Cross-sectional, 110 participants aged 18–22, seven days of intake recorded by time-stamped phone photographs.
Mode correction, 2026-09-02. This paragraph previously said the study was ”demonstrating” what matters for body composition and that it ”mechanistically explains why late-night eating promotes fat gain.” A cross-sectional study of 110 college-aged people cannot demonstrate a mechanism or a direction. What it found is an association: the higher-body-fat group ate most of their calories 1.1 h closer to melatonin onset than the leaner group (log-rank P = 0.009), while clock hour of eating showed no such difference (P = 0.72).
That contrast is the real finding and it is genuinely interesting — circadian timing tracked body fat where clock time did not. It is not evidence that late eating causes fat gain, and ”promotes” asserts exactly that.
One further caution `ksh05` already carries and this record should not contradict: the authors describe the timing effect as independent of caloric amount, macronutrients, activity and sleep duration — but that is their phrasing for a set of nulls (all P > 0.72), not the result of adjusting for them. The regression controlled for sex.