This will be a longitudinal study that aims to assess the effects of a sleep education and hygiene intervention on the sleep of young soccer players.

The primary outcome of interest is the difference-in-differences in objectively measured sleep duration from baseline to post-intervention between the intervention and control groups.

METHODS Study Design This is an experimental study that aims to evaluate the impact of a sleep hygiene program on the sleep patterns of high-level adolescent male soccer players. Before the beginning of the study, subjects will fill out standardized sleep questionnaires. Then, subjects' sleep will be monitored using actigraphy and sleep diaries for 6 consecutive weeks (Weeks 1-2 Baseline; Weeks 3-4 Intervention; Weeks 5-6 Post-Intervention) and another 2 weeks, 6 weeks after the conclusion of the post-intervention period. The sleep diaries will be completed every morning, and information regarding bedtime and waketime will be combined with actigraphy data. After the baseline period, sleep data will be analyzed, and a tailored a sleep hygiene and education program will be built, tailored around the specific needs of the group of participants. At the same time, subjects will be randomly assigned to either control or intervention group.

Sample Size Calculation and Participant Recruitment Sample size was determined using simulation-based power analyses focused on the primary intervention effect, defined as the difference-in-differences (DiD) in sleep duration from baseline to post-intervention between the intervention and control groups. This effect reflects whether the change in sleep duration over time differs between groups and constitutes the primary outcome of interest. Although sleep was monitored across multiple periods, the power analysis focused on the pre-specified baseline to post-intervention contrast, as this corresponds to the primary causal effect of interest.

Expected effect sizes were informed by preliminary internal actigraphy data and recent sleep hygiene intervention literature. Specifically, the intervention group was expected to increase sleep duration by approximately 25 minutes from baseline to post-intervention, while the control group was expected to increase by approximately 5 minutes over the same period, resulting in an anticipated DiD of 20 minutes. Within-subject variability in sleep duration was estimated at 35-45 minutes and between-subject variability at 60-70 minutes, based on internal pilot data.

Power simulations were conducted assuming a 2 × 2 mixed design (group × time: pre- vs post-intervention), a common within-condition standard deviation of 35 minutes, and a within-subject correlation of 0.5. The required sample to achieve 80% power at an alpha level of 0.05, based on 2000 simulations runs using the ANOVA_power shiny app was tested. This was achieved using 50 participants (25 per group), with the group × time interaction corresponding to the DiD effect, for an approximate power of 81%.

Considering possible dropouts, missing data, available equipment and logistical constraints, 52 subjects will be recruited to participate in the study. According to the participant classification framework, recruited participants will be high-level male soccer players, competing in the Portuguese U17 and U19 national championship.

Written informed consent will be obtained prior to data collection from participants or their legal guardians (in case of participants younger than 18 years old). Pre-specified exclusion criteria included: having travelled across more than two time zones in the past three months, history of major injuries in the previous 3 months, prior history of a sleep-related disorder, and current medication that may interfere with sleep. The study will be conducted according to the Declaration of Helsinki and was approved by the Ethics Committee of the Portugal Football School (PFS 23/2023).

Randomization Participants were allocated to either the sleep hygiene intervention or control group using stratified block randomization. Stratification factors were sleep difficulty score (ASSQ categories) and sleep hygiene index (good, normal, poor). Within each stratum, participants were randomized in blocks of size 2 or 4 with a 1:1 allocation ratio. Random block sizes were used to reduce allocation predictability.

Sleep data analysis and program construction At the end of the baseline period, with the data gathered from the initial two weeks, the lead researcher will create a sleep education program, consisting of two 45-minute sessions delivered for a 2-week period (1 session per week) by the same specialist, and an individualized report containing each subjects' own sleep data (averaged and daily time in bed, sleep duration, sleep efficiency, bedtime, and waketime) and individualized recommendations. This report will be sent at the end of the intervention period. The recommendations will be centered at 1) improving time in bed and consequently increase sleep duration, 2) improving sleep schedules regularity and minimizing social jetlag, and 3) improving sleep hygiene behaviors reported through the Sleep Hygiene Index.

Sleep Education Sessions: The sessions will be designed to provide subjects with essential knowledge about sleep and its impact on performance and recovery, present strategies and behaviors (sleep hygiene) to improve sleep quantity and quality and tailored to each subject, considering the results of the Sleep Hygiene Index. In the space of two weeks, two 45-minutes sessions will be scheduled, one in the first week and another in the second week, and both will be given by the same specialist. The first session will address general sleep knowledge, including sleep stages, their benefits, sleep's role on athletic performance and recovery, and data from adolescents and adolescent athletes. The second session will be focused on sleep hygiene, discussing environmental factors such as light exposure, room temperature, and topics such as jet lag, sleep pressure, and technology's role on sleep.

Measures Sleep-related questionnaires At study onset, participants will complete five online self-administered questionnaires. These included the reduced Morningness-Eveningness Questionnaire (r-MEQ) for chronotype assessment, the Pittsburgh Sleep Quality Index (PSQI) for subjective sleep quality, the Epworth Sleepiness Scale (ESS) for perceived sleepiness, the Athlete Sleep Screening Questionnaire (ASSQ) for sleep difficulties and risk of sleep disturbances, and the Sleep Hygiene Index (SHI) to assess the practice of sleep hygiene behaviors. All questionnaires were completed prior to the main data collection period. After the post-intervention period, participants will complete the PSQI, ASSQ, ESS, and SHI again.

r-MEQ (Portuguese version): scores on the r-MEQ range from 4 to 25, and subjects are classified as morning-type (scores from 18-25), evening-type (scores from 12-17), or neither type (scores from 4-11).

PSQI: 19-item questionnaire organized into seven dimensions: subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleeping medication, and daytime dysfunction. A higher score indicates greater sleep disturbances, and the developers have suggested a cut- off score of 5 for the global scale to classify subjects with poor sleep quality.

ESS: total daytime sleepiness score ranges from 0-24 (0-7 normal, 8-9 mild excessive, 10-15 moderate excessive, and ≥ 16 severe excessive).

ASSQ: 15-item questionnaire that assesses sleep and circadian factors of sleep duration, sleep quality, symptoms of insomnia, and chronotype with a timeframe of "over the recent past". The sum of the first seven items comprises the sleep difficulty score (SDS) which is categorized into four intensities of the symptoms: none (0-4), mild (5-7), moderate (8-10), and severe (11-17).

SHI: 13-intem questionnaire used to assess habits and behaviors that affect sleep quality, with higher scores indicating poorer sleep hygiene. The questionnaire assesses topics such as regular sleep schedules, bedtime routine behaviors, and bedroom environment with the aim of identifying inadequate sleep practices that can be improved.

Sleep diaries Each morning and throughout the study period, participants will receive a daily link from the research staff and regular reminders from the club staff to maximize compliance. Each diary entry will include information regarding bedtime, wake-up time, and perceived fatigue, using the Rating-of-Fatigue scale. A researcher constantly checked data entries to monitor compliance and address missing data promptly.

Statistical Analysis Primary analysis The primary analysis will assess the intervention effect using a difference-in-differences approach, comparing changes in sleep duration from baseline to post-intervention between the intervention and control groups. This effect will be tested via the group × time interaction in a linear mixed-effects model.

Secondary analyses Secondary analyses will examine (i) difference-in-differences from baseline to the intervention period, (ii) within-group changes in sleep duration across periods, and (iii) longer-term follow-up effects. These analyses will be considered supportive and exploratory.

Statistical significance for the primary analysis will be defined as p < 0.05.

Missing data Linear mixed-effects models will be used to analyze sleep outcomes, allowing inclusion of all available data under a missing-at-random assumption. No imputation of missing sleep data is planned.

Inclusion Criteria:

• Are actively playing soccer in the under-17 or under-19 of the soccer academy participating in the study.
• Are between 16 and 19 years old.
• Are available (i.e., no planned time off) during the monitoring period of the study.

Exclusion Criteria:

• Recent history of injury.
• In injury recovery process (i.e., return to play).
• Taking any medication that may impact sleep.
• Has travelled through multiple time zones in the past 3 months.

Data will be anonymized with each participant received a unique identifier code that only the responsible researcher will have access to. IPD data regarding the main study outcomes (sleep, perceptual measures, and sleep questionnaires) will be shared only with the core research team, if necessary, to discuss statistical analysis procedures.