REF-114 · Spriet, 2014 · Sports Medicine, Vol. · Paper

Exercise and Sport Performance with Low Doses of Caffeine

B+moderate evidenceMeasured on trained to highly trained athletesWritten before October 2026, being re-checked

Spriet reviews the case for low doses of caffeine. Most research has used moderate-to-high doses (5-13 mg/kg body mass), which have profound whole-body effects, variable results and some undesirable side effects. Low doses (< 3 mg/kg, about 200 mg) are also ergogenic in some exercise and sport situations, although less well studied.

Low doses (1) do not alter peripheral whole-body responses to exercise, (2) improve vigilance, alertness, mood and cognitive processes during and after exercise, and (3) are associated with few if any side effects — so their ergogenic effect appears to come from the central nervous system.

Effects on high-intensity sprint and burst activities are unresolved, responses are variable, and athletes need to test about 200 mg before and/or during training and competition individually.

Key Findings

Conventional research doses: 5-13 mg/kg body mass — profound whole-body effects, variable results, some side effects.
Low dose: < 3 mg/kg (about 200 mg) — ergogenic in some situations; less studied.
Low doses do not change peripheral whole-body responses to exercise; the benefit appears to be CNS-mediated (vigilance, alertness, mood, cognition).
Few if any side effects at low doses.
Unresolved: high-intensity sprint and burst activities; individual variability.

Practical Application

Caffeine at under 3 mg/kg (about 200 mg) is ergogenic in some exercise and sport situations with few if any side effects, and works through the central nervous system — improved vigilance, alertness, mood and cognition — rather than through changes in whole-body exercise responses; the high doses of the classic research (5-13 mg/kg) are not needed to get an effect.
Because responses vary, each athlete should test roughly 200 mg before and/or during training sessions before relying on it in competition; the paper leaves the timing to the athlete.
Whether low doses help sprint or burst efforts is unresolved; do not assume a benefit there.
Not from this paper: the position-stand dosing range (3-6 mg/kg, anhydrous form, timing) is in REF-115; mechanisms and metabolism in REF-116.
Not from this paper; the base's own practice: for races under about 2 h we take about 200 mg 30-60 min before the start; for longer races a second smaller dose in the second half; caffeine is reserved for key sessions and races rather than daily use.

Limitations

Narrative review; the low-dose evidence base is described by the author as sparse; no effect sizes or trial counts in the abstract.
How this entry was checked and changed
Why this level

regraded 2026-09-27 from A: narrative review without systematic method, from a leading caffeine research group (Spriet); no pooled data

Content correction 2026-09-27
"Traditional 6 mg/kg (480 mg for 80 kg)" -> the abstract's range is 5-13 mg/kg; "2-3% power improvement", "50-70% fewer side effects", "200 mg + 150 mg at 1 h", "total 200-350 mg", "> 500 mg daily", "30-45 min pre-race" -> not in the paper; deleted or relabelled as base practice.
"Not fat oxidation" -> the abstract says low doses do not alter peripheral whole-body responses, so the effect appears central; wording aligned.
Level A -> B+ (narrative review); Sport cycling -> general.