Shining a Light on Athletes Sleep: Development of a Screening Nomogram to Flag Athletes at Risk of Poor Sleep Quality

This cross-sectional study developed a high-specificity screening nomogram based on LASSO-selected light exposure behaviors, specifically pre-bed and nighttime phone use, to effectively identify athletes at risk of poor sleep quality.

The Gist

Imagine your body is a high-performance race car. To win the race, it needs the right fuel, the best mechanics, and, crucially, a good night's rest in the garage. But what if the garage lights are left on, or the driver keeps checking their phone while trying to sleep? The engine never truly cools down, and the car isn't ready for the next day.

This paper is like a mechanic's new, super-fast diagnostic tool. It helps sports coaches and doctors figure out which specific "garage light" habits are keeping athletes awake, without needing to spend hours analyzing complex data.

Here is the story of the study, broken down simply:

1. The Problem: Too Much Noise in the Garage

Scientists knew that light affects sleep. They also knew that athletes often sleep poorly. But trying to measure exactly how much light an athlete sees is like trying to count every single grain of sand on a beach. It's too hard, too slow, and too complicated for a busy sports team.

There were also too many questions. If you ask an athlete 23 different questions about their light habits (like "Do you wear blue-light glasses?" or "How often do you walk outside?"), it's like trying to find a needle in a haystack. Most of those questions might not actually be the problem.

2. The Solution: The "Feature Reduction" Filter

The researchers used a smart computer algorithm (called LASSO) to act like a gold panner. Imagine sifting through a river of sand (the 23 questions) to find the actual gold nuggets. The computer sifted through all the data and realized that almost all the "sand" (the other questions) didn't matter.

It found only three gold nuggets that were actually causing the sleep trouble:

1. Using your phone right before bed.
2. Checking your phone when you wake up in the middle of the night.
3. Checking your smartwatch when you wake up in the middle of the night.

That's it! Just those three habits were the biggest red flags for poor sleep in athletes.

3. The New Tool: The "Sleep Score" Calculator

Once they found the three key habits, the team built a Nomogram. Think of this as a visual cheat sheet or a slider-calc for doctors and coaches.

Instead of doing complex math, a coach can look at the chart:

• "How often does the athlete use their phone before bed?" (Never, Rarely, Sometimes, Often, Always).
• "How often do they check their phone/watch at night?"

They draw a line up to get a "score," add the scores together, and draw a line down to see the risk percentage.

The Analogy: It's like a weather forecast. You don't need to be a meteorologist to know that if the sky is dark, the wind is howling, and the barometer is dropping, a storm is coming. This tool tells the coach: "Hey, based on these three phone habits, there is a 46% chance this athlete is having a stormy night of sleep."

4. What Did They Find?

The tool worked surprisingly well.

• High Accuracy: It correctly identified athletes with bad sleep 92% of the time (when it said they had a problem, they usually did).
• The "Why": It turns out it's not just the light from the screen that's the problem. It's likely the content. Checking a phone or watch at 3:00 AM isn't just about blue light; it's about your brain getting excited, anxious, or stimulated by what you see. It's like revving the engine of a car that's supposed to be parked.

5. Why This Matters for Everyone

This isn't just for Olympic athletes. The study suggests that for anyone trying to sleep better, the most important thing to fix isn't necessarily buying expensive blue-light glasses or sleeping in a pitch-black cave.

The takeaway is simple:
If you want to sleep like a champion, stop treating your bed like a command center.

• Put the phone away before you close your eyes.
• If you wake up in the night, do not check your phone or your watch.
• Let your brain know that the "garage" is closed for the night.

The Bottom Line

The researchers created a simple, 3-question checklist that acts as a "sleep radar." It helps coaches spot athletes who are struggling with sleep due to their phone habits, so they can give them the right advice immediately. It turns a complex scientific mystery into a simple, actionable rule: Put the phone down, and let the sleep begin.

Technical Summary

1. Problem Statement

Elite athletes frequently suffer from insufficient sleep quality, which is critical for recovery and performance. While light exposure is a primary regulator of circadian rhythms, existing methods to assess its impact in sports settings have significant limitations:

• Objective Measures: Wearable light sensors are often logistically impractical due to issues with placement, clothing obstruction, and data processing requirements.
• Subjective Measures: Existing self-report questionnaires (e.g., Harvard Light Exposure Assessment) are often too complex, relying on specific light source classifications that lack intuitive applicability for athletes.
• The Gap: There is a lack of a concise, practical screening tool that identifies specific light-related behaviors most strongly associated with poor sleep quality in athletes, allowing high-performance practitioners to efficiently prioritize athletes for further assessment.

2. Methodology

Study Design: Cross-sectional study (Level 3 Evidence).

Participants:

• Sample: 121 athletes (50 female, 71 male; mean age 28 ± 9 years) from 20 countries and various competitive levels (recreational to elite/international).
• Exclusion: Participants with sleep disorders, sleep medication use, or supplements affecting light perception were excluded.

Instruments:

• Light Exposure Behaviour Assessment (LEBA): A 23-item questionnaire covering five factors (blue light filters, outdoor time, device use in bed, pre-bedtime light, morning/daytime light). Items are scored on a 5-point Likert scale.
• Pittsburgh Sleep Quality Index (PSQI): Used as the outcome variable. A global score >5 defined "poor sleep quality."

Statistical Analysis:

• Feature Reduction: Least Absolute Shrinkage and Selection Operator (LASSO) regression was employed to identify the minimal subset of LEBA items that best predicted poor sleep quality while preventing overfitting.
• Model Comparison: Three logistic regression models were trained and compared:
  1. Full-feature model: All 23 LEBA items.
  2. Factor-specific model: Only the 5 items from Factor 3 (device use in bed).
  3. Feature-reduced model: Only the items selected by LASSO.
• Validation: Data was split 80:20 (Training: N=98; Test: N=23). Performance was evaluated using 5-repeated 10-fold cross-validation on the training set and held-out testing on the independent set. Metrics included Area Under the Curve (AUC), Sensitivity, and Specificity.
• Tool Development: A nomogram (graphical calculator) was constructed using the rms package in R based on the best-performing model to translate predictor scores into probability estimates.

3. Key Contributions

• Identification of Key Predictors: The study isolated three specific behaviors from the 23-item LEBA questionnaire that are most strongly associated with poor sleep quality in athletes:
  1. Phone use before bed.
  2. Checking the phone during the night.
  3. Checking a smartwatch during the night.
• Development of a Clinical Nomogram: The authors created a visual, practitioner-friendly tool that allows sport scientists and clinicians to input these three behaviors and instantly estimate the probability of an athlete having poor sleep quality (PSQI >5).
• Demonstration of Feature Reduction: The study proved that a parsimonious model (3 items) outperforms complex models (23 items) in predictive accuracy and generalizability, offering a low-burden alternative to comprehensive questionnaires.

4. Results

Model Performance:
The feature-reduced model (3 items) demonstrated superior performance compared to the full-feature and factor-specific models.

• Independent Test Set Performance:
  • AUC: 0.83 (Highest among all models).
  • Sensitivity: 0.70.
  • Specificity: 0.92 (High specificity indicates that athletes flagged by the tool are very likely to genuinely have poor sleep quality).
• Comparison: The full-feature model had a lower test AUC (0.79) and lower specificity (0.77), suggesting overfitting to the training data.

Key Findings:

• LASSO regression successfully reduced the 23 variables to the three identified above, all belonging to Factor 3 (using phone/smartwatch in bed).
• The nomogram example showed that an athlete who "often" uses a phone before bed, "sometimes" checks it at night, and "rarely" checks a smartwatch at night accumulates ~130 points, corresponding to a 46% predicted risk of poor sleep quality.

5. Significance and Clinical Relevance

• Practical Application: The developed nomogram serves as a rapid (<1-2 minutes), low-burden screening tool for high-performance environments. It enables practitioners to quickly identify athletes at risk without needing complex objective monitoring or lengthy questionnaires.
• High Specificity: With a specificity of 0.92, the tool is highly effective at ruling in poor sleep quality when a flag is raised, warranting further investigation into underlying causes (e.g., training load, anxiety, or specific light interventions).
• Behavioral Insight: The findings suggest that the association between light and sleep in athletes is driven not just by spectral composition (blue light) but potentially by the cognitive engagement and psychological mechanisms (e.g., "orthosomnia" or sleep anxiety) associated with checking devices during the night.
• Future Directions: The authors recommend using this tool as a starting point for targeted education and interventions. Future research should validate the nomogram against objective measures (wearable sensors, melatonin levels) and examine causal relationships through longitudinal designs.

Conclusion:
This study successfully translates complex light exposure data into a practical, evidence-based screening tool. By focusing on three specific device-related behaviors, it provides a scalable solution for monitoring sleep health in diverse athletic populations, bridging the gap between research findings and applied sports medicine practice.