• caffeine;
  • coffee;
  • healthy people;
  • real-life condition;
  • sleep;
  • sleep disorder


Aim:  To report changes in the quality of sleep after drinking an evening cup of either caffeinated or decaffeinated coffee, in healthy subjects in everyday life.

Methods:  Sixty-three healthy men and women, who considered themselves to be caffeine sensitive were included in a double-blind, cross-over trial, randomised to receive either caffeinated coffee containing 90 mg of caffeine, or, as control, a dose of decaffeinated coffee containing 4.5 mg caffeine, taken after dinner. The primary outcome measure was the degree of sleep disturbance, scored on a visual analogue scale, ranging from 0 (excellent sleep) to 100 (very disturbed sleep). Ancillary criteria were patients' reported estimate of sleep latency, and how often the subjects reported waking.

Results:  Mean age of subjects was 30.5 ± 12 years. The quality of sleep was significantly worse with caffeinated (mean 30.8 ± 22.7) than with decaffeinated coffee (mean 19.5 ± 16.9), P = 0.001. Caffeinated coffee also significantly increased the sleep latency (mean difference 17 ± 31 minutes, P < 0.001) and the frequency of waking (mean 1.3 vs 0.8 episodes in the night, P = 0.006) compared with decaffeinated coffee.

Conclusions:  Even a single cup of caffeinated coffee consumed before bedtime in real-life conditions causes a deterioration in the quality of sleep in caffeine-sensitive subjects.