Sleep and performance in young archers: Guru 2019

Fifteen male archers aged 14 to 18 (mean age 15.3), training and living at an Indian Army sports institute in Pune, answered a sleep questionnaire before and after a year of changes to evening habits and schedules. At the start they slept little on school days: they went to bed at about 22:27, got up at 5:28 and took 28 minutes on average to fall asleep, sleeping about 6 hours 27 minutes against 8 hours 39 minutes at weekends.

A year later, among the 13 who repeated the questionnaire, weekday time to fall asleep had dropped to 15 minutes (p = 0.008), the study's primary outcome. Weekday wake-up time was 29 minutes earlier, at 4:59. Sleep duration rose by about 10 minutes on weekdays and 46 at weekends, but neither change is significant (p = 0.492 and 0.059), and the same goes for daytime sleepiness and naps.

Training score, out of a maximum of 1440, went from 1268 to 1291 in the 10 archers who shot on both occasions (p = 0.018). For comparison the authors retrieved the scores of 8 archers from the previous year, similar in age and training, who without any intervention had gone from 630.25 to 647.25 out of 720 (p = 0.269).

The study is a before-and-after design with no concurrent control group, registered with the Clinical Trials Registry of India. The archers, chosen at random from those at the institute, had to be aged 14 to 18, have at least 24 weeks of practice and have competed at least at national championship level; those in a pre-competition phase were excluded. All lived in the same accommodation and ate in the same mess.

Sleep was measured with the School Sleep Habits Survey, a 63-item questionnaire covering the previous two weeks, filled in in small groups with help from experts, in May to August 2017 and again in May to August 2018. Sleep duration was worked out by subtracting bedtime from wake-up time.

The intervention lasted a year and involved athletes, coaches and administrators: sessions on sleep hygiene, training never extended into the evening or started too early, an evening reminder from the hostel warden, handing in phones and gaming devices in the evening, less noise and light in the rooms, and access to a sleep physician. The score came from two training sessions in the same week as the questionnaire, with a maximum of 1440 points under World Archery rules (the text mentions 72 arrows, which would allow at most 720 points); distance and bow type are not given.

Without a concurrent control group, the effect of the intervention cannot be separated from a year of growing up and training. The comparison with the 8 archers from the previous year helps little: it was out of 720 points rather than 1440, and proportionally that group had improved slightly more (17 points out of 720, against 23 out of 1440); that its improvement is not significant owes something to the smaller sample. The score also comes from training sessions, with no distance or conditions reported.

Sleep is self-reported rather than recorded with instruments, and calculating duration as the gap between bedtime and wake-up time ignores the time taken to fall asleep. The sleep comparisons use a test for independent groups even though the people were the same, and of the 15 archers 13 remained for the questionnaire and 10 for the score. The text is not internally consistent: the drop in time to fall asleep is given as p = 0.008 with an effect size of 1.08 in the results and as p = 0.005 with 2.174 in the discussion, and the reduction in the weekday to weekend gap in time to fall asleep goes from p = 0.055 in the table to p = 0.013 in the text. The abstract reports an increase in sleep duration that is not significant. Finally, the intervention bundles many things together, from training times to phones, and there is no way to tell which of them mattered.

The study describes a problem familiar to many youth coaches: teenagers who get up before 5:30 to train, fall asleep late and catch up at weekends, with more than two hours' difference in sleep duration between school days and weekends. This descriptive picture is the most solid part of the paper, even though it comes from a questionnaire.

On the link between sleep and score the study allows no conclusions. Training scores rose, but they rise anyway over a year of training, and the comparison group from the year before had improved proportionally more. Sleep itself, apart from time to fall asleep, changed little: ten more minutes on weekdays, with wake-up half an hour earlier. For a club or centre with young athletes in residence, the paper shows that an organisational change to schedules and evening habits is feasible and was accepted by the teenagers; how much it affects performance remains to be measured in a controlled study with sleep recorded by instruments.