Sleep loss, caffeine and young archers: Park 2017 study

Six boys from a Korean middle-school archery team, with a mean age of 14 and a half and almost five years of shooting, shot at 30 metres under three conditions a week apart: after a normal night's sleep, after a sleepless night with a placebo capsule, and after a sleepless night with a capsule of caffeine at 2.5 mg per kilogram of body weight. The score over six ends did not change in any meaningful way: 336.83 after normal sleep, 334.83 with the placebo and 339.16 with caffeine, differences the analysis cannot tell apart from chance (p = 0.407).

Heart rate during the test, on the other hand, was lower with caffeine (p = 0.019): 93.90 beats per minute at the first end, against 102.25 with the placebo and 111.41 after normal sleep, while at the third and sixth ends the three conditions were much closer. Blood glucose did not change. Growth hormone was lower in the caffeine condition and, in all three conditions, higher after shooting than before, with no interaction between condition and time of measurement. The authors conclude that caffeine and the sleepless night had a partly positive effect on heart rate and performance, and none on glucose or growth hormone.

The study was carried out in 2014 with the archery team of a middle school, using repeated measures on the same boys. The three conditions always came in the same order, one per week: normal sleep first, then the sleepless night with the placebo, and finally the sleepless night with caffeine. On the sleepless nights the boys stayed awake from 10 pm to 7 am, fasting, and were checked every hour; the test day was followed by a rest day and five days of normal training. The capsule, taken at 8.20 am after breakfast, contained either pure caffeine or, as the placebo, glucose, and the boys did not know which they were given.

Blood was drawn from a vein three times, at 7.20, 9.10 and 11.00, that is on waking, before shooting and after shooting. The 30-metre test began at 10 am, with the range at 26 degrees and 60% humidity, and lasted six ends; the paper does not give the number of arrows per end, but totals above 330 points suggest 36 arrows. Heart rate was recorded with a Polar monitor at the first, third and sixth ends. Comparisons used repeated-measures analysis of variance.

Six boys are too few to detect differences of a few points over 36 arrows, and the fixed order of conditions matters: caffeine always came last, after two more weeks of training and with one sleepless night already behind them, so familiarity and experience are mixed up with the effect of the substance. No capsule was given after normal sleep, so even the comparison between sleep and the sleepless night is not entirely like for like.

The growth hormone data call for caution: in the caffeine condition the level was already lower on waking (0.390 ng/ml against 0.786 and 0.808), before the capsule had been taken, so the difference between conditions cannot be due to caffeine. For heart rate the authors themselves note that, without an interaction, the effect cannot be put down to caffeine alone. The conclusion of a positive effect on performance finds no support in the score, which did not change; the text also sometimes calls the placebo group the glucose group, and the text and the table note disagree about when growth hormone was highest (this record follows the table). The discipline is not stated: the 30-metre distance and the Korean school setting suggest recurve.

For a young archer the most concrete finding is that a whole night without sleep did not wreck the score at 30 metres: 334.83 points against 336.83, in six boys used to shooting. With numbers this small and a single test day, that does not mean sleep is unimportant, only that over a short distance and a short test any loss was smaller than the study could detect.

On caffeine the study does not allow anyone to conclude that it helps a boy who has slept badly: the score did not move, the lower heart rate may reflect the order of the tests, and the dose, chosen to stay below anti-doping thresholds, was given to 14-year-olds who did not normally take caffeine. For anyone coaching this age group it is one more piece of information, not a recommendation.