Inter-relationship between sleep quality, insomnia and sleep disorders in professional soccer players

This was not an intervention study — it was a **cross-sectional observational survey** that measured three things in professional soccer players:

**Sleep quality** (how well they slept, not just how long)

**Insomnia symptoms** (difficulty falling asleep, staying asleep, or waking too early)

**Daytime sleepiness** (how likely they were to fall asleep during the day)

The researchers did not test any treatment or intervention. They simply gave three questionnaires to players at one point in time and looked at how these sleep measures related to each other and to player characteristics like age, body fat, and ethnicity.

The outcome measures were:

Prevalence (percentage) of poor sleep quality

Prevalence of insomnia

Prevalence of excessive daytime sleepiness

Correlations between these three sleep measures

**111 professional male soccer players** from the Qatar Stars League (the top professional football league in Qatar)

Mean age: 23.7 years (standard deviation 4.8 years; range approximately 18–33)

Mean body mass index (BMI): 22.9 kg/m² (lean, athletic build)

Mean body fat percentage: 12.4%

Ethnic composition: 50.5% Middle Eastern/North African, 34.2% African, 15.3% other

All players were from the same league but likely from multiple clubs (the paper does not specify how many clubs)

No exclusion criteria reported — this was a convenience sample of available players

**Why this matters for your self-experiment:** These are elite athletes who train daily, have structured schedules, and likely have access to sports science support. If *they* have poor sleep, it's not just a "lazy person" problem. Sleep issues can affect anyone, including people who are highly motivated and physically active.

Three validated questionnaires were used:

1. **Pittsburgh Sleep Quality Index (PSQI)** — measures sleep quality over the past month. Score range 0–21, with higher scores = worse sleep. The cutoff for "poor sleep quality" is PSQI ≥ 5. The PSQI covers seven components: subjective sleep quality, sleep latency (how long to fall asleep), sleep duration, sleep efficiency (time asleep vs. time in bed), sleep disturbances, use of sleep medication, and daytime dysfunction.

2. **Insomnia Severity Index (ISI)** — measures insomnia symptoms over the past month. Score range 0–28, with higher scores = worse insomnia. Cutoffs: 0–7 = no clinically significant insomnia, 8–14 = subthreshold insomnia, 15–21 = clinical insomnia (moderate severity), 22–28 = clinical insomnia (severe). The authors used ISI ≥ 11 as their cutoff for "insomnia" — this is a slightly lower threshold than the standard clinical cutoff of 15, meaning they were capturing milder cases.

3. **Epworth Sleepiness Scale (ESS)** — measures daytime sleepiness (how likely you are to doze off in eight different situations, like sitting and reading, watching TV, or as a passenger in a car). Score range 0–24, with higher scores = sleepier. The cutoff for excessive daytime sleepiness is ESS > 8.

**Important measurement note:** All three are self-report questionnaires. No objective sleep measures (like actigraphy or polysomnography) were used. This means the data reflect what players *remember* and *report* about their sleep, not necessarily what actually happened. People are notoriously bad at estimating how long it takes them to fall asleep or how many times they wake up at night.

**Study design:** Cross-sectional observational study. This means all measurements were taken at a single point in time. There was no follow-up, no intervention, and no control group.

**Randomisation:** None. This was not an experiment — there was nothing to randomise. Players were recruited from the league, and all who agreed to participate were included.

**Blinding:** None. Players filled out questionnaires about their own sleep. There was no treatment to blind.

**Duration:** Single assessment. Each player completed the three questionnaires once. The PSQI asks about the past month, the ISI asks about the past month, and the ESS asks about "recent times" — so the data cover roughly the past month of each player's life, but there was no repeated measurement.

**Statistical approach:**

Prevalence was reported as percentages with 95% confidence intervals

Associations between sleep measures were tested using Pearson correlations (r values)

Differences between groups (e.g., by ethnicity, by BMI category) were tested using t-tests or ANOVA

Significance was set at p < 0.05

**What this design can and cannot prove:**

*Can prove:* The prevalence of poor sleep quality, insomnia symptoms, and daytime sleepiness in this specific population at this specific time. It can also show that these three sleep problems are correlated — people with poor sleep quality tend to have more insomnia symptoms and more daytime sleepiness.

*Cannot prove:*

**Causation** — You cannot say that poor sleep quality *causes* insomnia or vice versa. They are correlated, but the direction is unknown.

**Temporal changes** — With only one measurement, you cannot see if sleep quality changes across a season, after games, or with training load.

**Objective sleep** — Without actigraphy or polysomnography, you cannot confirm whether self-reported sleep matches actual sleep.

**Generalisability to other populations** — These are male professional soccer players in Qatar. Results may not apply to female athletes, amateur athletes, non-athletes, or people in different climates/cultures.

**Major methodological weaknesses:**

1. **No objective sleep measurement** — Self-report is prone to recall bias, social desirability bias, and simple error.

2. **Single time point** — Sleep varies night to night and week to week. One snapshot may not represent typical sleep.

3. **No control for training load or game schedule** — Players may have completed questionnaires after a night game, a travel day, or a rest day, which would dramatically affect responses.

4. **Convenience sample** — Players who agreed to participate might differ from those who declined (e.g., players with worse sleep might be more or less likely to volunteer).

5. **No exclusion criteria** — Players with diagnosed sleep disorders, medical conditions, or medication use were not excluded, which could inflate prevalence estimates.

6. **ISI cutoff of ≥ 11** — This is lower than the standard clinical cutoff of 15, so the 27% insomnia prevalence may overestimate clinically significant insomnia.

**Primary outcomes (prevalence):**

**Poor sleep quality (PSQI ≥ 5):** 68.5% of players (76 out of 111). 95% confidence interval: 59.8% to 77.1%. This means roughly 7 out of 10 players had sleep quality that is considered poor by clinical standards.

**Insomnia symptoms (ISI ≥ 11):** 27.0% of players (30 out of 111). 95% CI: 18.8% to 35.3%. About 1 in 4 players had at least subthreshold insomnia.

**Excessive daytime sleepiness (ESS > 8):** 22.5% of players (25 out of 111). 95% CI: 14.7% to 30.3%. About 1 in 5 players reported being sleepy enough to doze off during daily activities.

**Secondary outcomes (correlations between sleep measures):**

Sleep quality (PSQI) was **positively correlated** with insomnia (ISI): r = 0.42, p < 0.001. This is a moderate correlation — meaning players with worse sleep quality tended to have more insomnia symptoms, but the relationship is not extremely strong (r² = 0.18, so only 18% of the variance in one measure is explained by the other).

Sleep quality (PSQI) was **positively correlated** with daytime sleepiness (ESS): r = 0.23, p = 0.018. This is a weak correlation — worse sleep quality is associated with more daytime sleepiness, but the relationship is small (r² = 0.05).

Insomnia (ISI) was **positively correlated** with daytime sleepiness (ESS): r = 0.24, p = 0.012. Again, a weak correlation.

**Non-significant findings:**

Age was not associated with any sleep measure (p > 0.05 for all)

Body fat percentage was not associated with any sleep measure (p > 0.05)

BMI was not associated with any sleep measure (p > 0.05)

Ethnicity (Middle Eastern/North African vs. African vs. other) was not associated with any sleep measure (p > 0.05)

**Important nuance:** The authors report that 68.5% had poor sleep quality, but they also note that the mean PSQI score for the whole group was 6.6 ± 3.3. This is just above the cutoff of 5, meaning the average player is right on the border of poor sleep. The distribution matters — some players had very poor sleep (PSQI > 10), while others had good sleep (PSQI < 5).

Let's translate these numbers into something tangible:

**68.5% poor sleep quality** — If you have 10 teammates, about 7 of them are reporting sleep quality that is worse than what is considered healthy. This is not a "slight" problem; it's a majority problem.

**The correlation between sleep quality and insomnia (r = 0.42)** — This is roughly equivalent to the relationship between height and weight in adults. It's a real association, but knowing someone's sleep quality only tells you about 18% of the story about their insomnia. Many players with poor sleep quality did NOT have insomnia, and some with good sleep quality DID have insomnia symptoms.

**The correlation between sleep quality and daytime sleepiness (r = 0.23)** — This is weak. It means that even players who reported good sleep quality sometimes reported being sleepy during the day, and vice versa. Sleep quality and daytime sleepiness are related but not the same thing.

**What this means for you:** If you measure only one aspect of sleep (e.g., just sleep quality or just sleepiness), you might miss important information. A person can have poor sleep quality but not feel sleepy during the day (perhaps due to caffeine, adrenaline, or motivation), or they can feel sleepy despite reporting good sleep (perhaps due to undiagnosed sleep apnea, poor sleep efficiency, or high training load).

**What the authors acknowledge:**

The cross-sectional design prevents causal conclusions

Self-report questionnaires may not reflect objective sleep

The sample is limited to male professional soccer players in one league

The ISI cutoff of ≥ 11 is lower than the standard clinical cutoff of 15

**What a critical reader would add:**

1. **No actigraphy or polysomnography** — This is the biggest limitation. Self-reported sleep duration is notoriously inaccurate. People with insomnia tend to underestimate how much they sleep, while people without sleep complaints tend to overestimate. Without objective data, we don't know if these players actually slept poorly or just *thought* they slept poorly.

2. **Single time point, no seasonal variation** — Soccer players' sleep varies dramatically across a season: pre-season (high training load), competitive season (games, travel, late-night matches), off-season (reduced structure). A single measurement in one month cannot capture this.

3. **No control for confounding variables** — The study did not measure or control for: caffeine intake, alcohol use, screen time before bed, training schedule (morning vs. evening), travel (time zones, flights), game schedule (night games disrupt sleep), napping habits, or medication use (including painkillers, which many athletes take).

4. **No data on sleep duration** — The PSQI includes a component for sleep duration, but the authors only report the global PSQI score. We don't know how many hours these players actually slept. "Poor sleep quality" could mean short sleep, fragmented sleep, or perceived poor sleep despite adequate duration.

5. **Cultural and environmental factors** — Qatar has a hot climate, and training/games often occur in the evening or at night to avoid heat. Many players are expatriates living away from family. These factors could affect sleep but were not measured.

6. **No female athletes** — Sleep patterns and disorders differ between sexes. These results cannot be generalised to female soccer players.

7. **Response rate not reported** — We don't know how many players were approached vs. how many agreed to participate. If participation was low, selection bias could be severe.

For someone running their own n=1 experiment on sleep and performance:

### What to test

**Your own sleep quality** using the PSQI (free online, takes 5 minutes) — not just sleep duration

**Your insomnia symptoms** using the ISI (free online, takes 2 minutes)

**Your daytime sleepiness** using the ESS (free online, takes 2 minutes)

**Test an intervention** like: consistent bedtime/wake time, no screens 1 hour before bed, 7–9 hours time in bed, or a specific sleep hygiene protocol

### Minimum meaningful duration

**At least 2 weeks** of baseline measurement before any intervention

**At least 2 weeks** of intervention — sleep changes take time to stabilise

**Total minimum: 4 weeks** (2 weeks baseline + 2 weeks intervention)

**Better: 8 weeks** (2 weeks baseline + 4 weeks intervention + 2 weeks washout if testing multiple interventions)

### What to measure (specific metrics)

**Daily sleep diary** (not just a questionnaire at the end): record bedtime, wake time, estimated time to fall asleep, number of night wakings, total sleep time, and subjective sleep quality (1–10 scale)

**PSQI once per week** — this gives you a validated measure of overall sleep quality

**ISI once per week** — tracks insomnia symptoms

**ESS once per week** — tracks daytime sleepiness

**Optional but valuable:** Wearable sleep tracker (e.g., Oura Ring, Whoop, or actigraphy watch) for objective sleep duration, sleep efficiency, and sleep stages. Be aware that consumer wearables are less accurate than clinical actigraphy, but they're better than nothing.

**Performance metric:** Choose ONE performance measure relevant to you — e.g., reaction time (simple app-based test), grip strength, 5-minute time trial on a bike/row, or subjective energy rating (1–10) each morning

### Key confounds to control for

**Caffeine** — Keep dose and timing constant. If you change caffeine, you can't attribute sleep changes to your intervention.

**Alcohol** — Zero alcohol during the experiment, or at least keep it constant. Alcohol fragments sleep even in small amounts.

**Training load** — Log your training volume and intensity daily. A hard training day will affect sleep differently than a rest day.

**Meal timing** — Eating within 2 hours of bed can disrupt sleep. Keep meal timing consistent.

**Screen use** — Log screen time in the hour before bed. If you're testing a "no screens" intervention, this is your independent variable.

**Stress** — Rate your daily stress (1–10). Life stress is a major confound for sleep.

**Travel and schedule changes** — Note any travel across time zones, late-night events, or changes in work/school schedule.

### What a positive result would look like

**PSQI drops by ≥ 3 points** (e.g., from 8 to 5) — this is a clinically meaningful improvement

**ISI drops by ≥ 4 points