Nutrition
Caffeine raised jump height and power in boxers 60 minutes after dosing
By Monica O'Reilly
Published · Updated

What the trial tested
Researchers at Beijing Sport University ran a randomised, double-blind, placebo-controlled crossover trial in 15 male boxers aged about 21, all at Chinese national second-class level or higher. Each boxer took capsules of placebo, 3 mg per kg of body mass of caffeine, or 6 mg per kg, in a random order, with one week between visits.
Participants avoided caffeine for 48 hours before each visit. Testing happened at baseline and in blocks starting 30 and 60 minutes after the capsules. Measures were countermovement jump, ballistic push-up, and the peak torque of the shoulder and knee on an isokinetic dynamometer.
Jump results at 60 minutes
Both doses beat placebo on every jump measure at the 60-minute block. Jump height was 0.039 m higher at 3 mg per kg (95% CI 0.019 to 0.059) and 0.051 m higher at 6 mg per kg (0.032 to 0.071). Peak power was 279 W and 348 W higher. The two caffeine doses did not differ from each other.
Placebo jump height fell from 0.471 m at baseline to 0.450 m at 60 minutes, while the caffeine conditions rose. The caffeine visits also started from lower baseline jumps (0.432 m and 0.440 m). The 60-minute gaps therefore combine a rise under caffeine with a drop under placebo.
Push-ups and joint torque
In the ballistic push-up, only the 6 mg per kg dose beat placebo. Peak power was 220.7 W higher (95% CI 79.7 to 361.6). The authors note that the impulse and mean power gains were only about 1.0 to 1.3 times the test's typical error, so they urge caution on those two.
Shoulder torque was higher under caffeine at selected time points, with no consistent edge for either dose. Knee extension torque showed the clearest dose gap. At 60 minutes, 6 mg per kg exceeded placebo by 41.4 N·m (95% CI 29.0 to 53.8) and the 3 mg per kg dose by 26.0 N·m.
Limits that matter
The trial had 15 boxers, so comparisons between the two doses are imprecise. The authors say a non-significant gap between doses is no evidence they are equal. Order and carryover effects were not formally tested, and blood or saliva caffeine levels, habitual caffeine intake and sleep were not measured.
Side effects were asked about verbally and not recorded with a standard tool. The trial was not prospectively registered, and no single primary outcome was set in advance. The 48-hour caffeine break may have caused withdrawal-related effects.
What it adds
All tests were laboratory measures. No one threw punches or boxed, so the trial shows changes in jump, push-up and joint torque outputs and cannot show better boxing. The authors report no funding and no conflicts of interest. They found 6 mg per kg was not consistently better than 3 mg per kg across outcomes.
This is news reporting and not medical advice.