2023 International Olympic Committee’s (IOC) consensus statement on Relative Energy Deficiency in Sport (REDs)

This is not an experimental study but a consensus statement—a structured review and expert synthesis of the existing scientific literature. The authors did not test a single intervention. Instead, they:

• Reviewed and updated the definition of REDs, moving from a focus on female athletes (the "Female Athlete Triad") to a sex-inclusive syndrome affecting both men and women.
• Developed a new Physiological Model that distinguishes between "problematic" LEA (prolonged, severe, or repeated exposure causing harm) and "adaptable" LEA (short-term, mild exposure that the body can compensate for).
• Created the REDs Clinical Assessment Tool Version 2 (REDs CAT2) for clinicians to diagnose and stratify risk based on accumulated signs and symptoms.
• Outlined guidelines for safe body composition assessment and prevention/treatment principles.
• Proposed methodological best practices for future REDs research.

The "outcome measures" are the consensus statements themselves—agreed-upon definitions, models, and clinical tools—not quantitative results from a single experiment.

No participants were directly studied in this consensus statement. However, the authors synthesised data from:

• Over 170 original research publications published between 2018 and 2023.
• Studies on female and male athletes across a wide range of sports (endurance, aesthetic, weight-class, team sports).
• Populations ranging from elite Olympians to recreational athletes, aged approximately 15–50 years.
• Clinical populations with diagnosed eating disorders, as well as non-clinical athletes with unintentional LEA.

The statement explicitly notes that earlier REDs research focused heavily on female athletes, but the 2023 update includes emerging data on males, including studies showing that LEA suppresses testosterone, impairs bone health, and reduces libido in men.

No direct measurements were taken. Instead, the authors used a structured consensus process:

• Literature review: Systematic search of PubMed, Web of Science, and SportDiscus for studies on LEA, REDs, energy availability, and related health/performance outcomes.
• Expert panel deliberation: 15 international experts from sports medicine, nutrition, physiology, and psychology met in multiple rounds to discuss and vote on statements.
• Delphi method: Statements were iteratively refined until ≥80% agreement was reached among panel members.
• Clinical tool development: The REDs CAT2 was built from previously validated screening tools (e.g., the Low Energy Availability in Females Questionnaire, LEAF-Q; the Male Athlete Triad Screening Tool) and expert consensus on new items.

Study design: This is a consensus statement, not a primary research study. It follows the standard IOC methodology for expert consensus documents: a systematic literature review combined with a modified Delphi process to achieve agreement among a panel of experts.

How the consensus was built:

1. Literature search: The authors identified and reviewed all original research, systematic reviews, and meta-analyses on REDs published since the 2018 consensus.
2. Drafting: A writing group produced initial statements covering definitions, models, clinical tools, prevention, and research methods.
3. Delphi rounds: The full expert panel reviewed and voted on each statement. Statements that did not reach ≥80% agreement were revised and re-voted. This process continued until consensus was achieved.
4. Final approval: The final document was approved by all authors and the IOC.

What this design can prove:

• Expert opinion and clinical consensus: The statement represents the current best thinking of leading researchers and clinicians. It provides a framework for diagnosis and treatment that is grounded in the available evidence.
• Identification of knowledge gaps: The authors explicitly list areas where evidence is weak or absent, guiding future research.

What this design cannot prove:

• Causality: The consensus statement does not provide new experimental data. It cannot prove that LEA causes specific outcomes—it synthesises existing evidence, much of which is observational or cross-sectional.
• Effect sizes: No new effect sizes, confidence intervals, or p-values are generated. The statement relies on the statistical findings of the original studies it reviews.
• Comparative effectiveness: The statement does not compare different interventions (e.g., dietary counselling vs. psychological therapy) head-to-head.

Major methodological weaknesses:

• Publication bias: The consensus is based on published literature, which tends to favour positive or significant findings. Negative or null studies on LEA may be underrepresented.
• Expert bias: The panel members are all prominent REDs researchers who have published extensively on the topic. There is a risk of groupthink or over-interpretation of weak evidence.
• Lack of quantitative synthesis: The statement does not include a meta-analysis. It summarises findings narratively, which can be less precise than pooled effect estimates.
• No preregistration: Unlike a systematic review, the search strategy and analysis plan were not preregistered, increasing the risk of selective reporting.

The consensus statement produced several key conclusions, synthesised from the literature and expert agreement:

• Definition update: REDs is now defined as "a syndrome of impaired health and/or performance caused by prolonged and/or severe low energy availability (LEA), with or without disordered eating, in both female and male athletes."
• Prevalence: LEA is estimated to affect 20–60% of female athletes and 15–40% of male athletes, depending on sport and assessment method. Prevalence is highest in endurance sports (e.g., long-distance running, cycling), aesthetic sports (e.g., gymnastics, figure skating), and weight-class sports (e.g., wrestling, rowing).
• Health consequences (from synthesised studies):
  • Bone health: LEA reduces bone mineral density (BMD) by 0.5–1.0 standard deviations below age-matched norms in female athletes, increasing stress fracture risk by 2–4 times. In males, LEA suppresses testosterone (by 30–50% in some studies), leading to similar BMD reductions.
  • Menstrual function: LEA causes functional hypothalamic amenorrhea in 10–40% of female athletes, with recovery of menses typically requiring 6–12 months of adequate energy intake.
  • Metabolic rate: Resting metabolic rate (RMR) decreases by 10–20% in response to sustained LEA, reflecting metabolic adaptation. An RMR ratio (measured/predicted) <0.90 is a clinical marker of LEA.
  • Immune function: LEA increases risk of upper respiratory tract infections by 2–3 times, likely due to impaired immune cell function.
  • Mental health: LEA is associated with a 2–3 fold increase in depressive symptoms, anxiety, and disordered eating behaviours.
  • Performance: LEA impairs endurance performance (time to exhaustion reduced by 10–20%), muscle strength (5–15% reduction), and cognitive function (reaction time slowed by 5–10%).
• New Physiological Model: The model distinguishes between:
  • Problematic LEA: Prolonged (>30 days), severe (<30 kcal/kg fat-free mass/day), or repeated exposure that overwhelms compensatory mechanisms, leading to health and performance decline.
  • Adaptable LEA: Short-term (<14 days), mild (30–45 kcal/kg fat-free mass/day) exposure that the body can compensate for without long-term harm.
• REDs CAT2: The new clinical tool stratifies athletes into three risk categories:
  • Low risk: No or few signs/symptoms; can continue full training and competition.
  • Moderate risk: 2–4 signs/symptoms (e.g., menstrual dysfunction, low BMD, low RMR); requires modified training and referral to a sports dietitian.
  • High risk: ≥5 signs/symptoms or any severe outcome (e.g., stress fracture, eating disorder); requires medical leave from training and multidisciplinary treatment.
• Body composition assessment: The statement recommends against routine body composition testing for all athletes, citing risk of triggering disordered eating. If performed, it should be done by trained professionals with clear clinical indication and athlete consent.

Since this is a consensus statement, effect magnitudes are drawn from the underlying studies:

• Bone density loss: A female athlete with LEA for 6–12 months may lose 0.5–1.0 standard deviations in lumbar spine BMD. This is roughly equivalent to 5–10 years of normal age-related bone loss, or about a 10–20% increase in fracture risk per standard deviation drop.
• Testosterone suppression in males: LEA can reduce total testosterone by 30–50% (e.g., from 500 ng/dL to 250–350 ng/dL), which is comparable to the drop seen in men with hypogonadism or after 1–2 weeks of severe sleep deprivation.
• Metabolic adaptation: A 10–20% reduction in RMR means an athlete who normally burns 1,800 kcal/day at rest might drop to 1,440–1,620 kcal/day. This makes weight loss progressively harder and increases risk of weight regain.
• Performance decline: A 10–20% reduction in time to exhaustion (e.g., from 60 minutes to 48–54 minutes) is a meaningful drop for any endurance athlete. A 5–15% strength loss (e.g., from a 100 kg squat to 85–95 kg) would be noticeable in training.
• Infection risk: A 2–3 fold increase in upper respiratory infections means an athlete who normally gets 1–2 colds per year might get 3–6 per year, potentially disrupting training cycles.

The authors explicitly acknowledge several limitations, and a critical reader would note additional ones:

Acknowledged by authors:

• Heterogeneity of studies: The underlying research uses different definitions of LEA, different measurement methods (e.g., food diaries vs. doubly labelled water), and different populations, making synthesis difficult.
• Lack of longitudinal data: Most studies are cross-sectional or short-term (days to weeks). Long-term data on the health and performance consequences of chronic LEA are sparse.
• Sex differences: While the statement includes males, the evidence base for REDs in men is still much smaller than for women. Many findings in males are extrapolated from female data.
• Measurement challenges: Energy availability (EA = energy intake – exercise energy expenditure / fat-free mass) is notoriously difficult to measure accurately in free-living athletes. Errors in any component (diet, exercise, body composition) can misclassify athletes.
• Causality: Most evidence is observational. Few randomised controlled trials have manipulated EA and measured health/performance outcomes over months or years.

Additional limitations a critical reader would note:

• No quantitative meta-analysis: The statement does not pool effect sizes across studies, so the magnitude of each outcome is based on narrative summary rather than a weighted average. This reduces precision.
• Expert panel composition: All 15 experts are from Western countries (USA, UK, Canada, Australia, Europe). Cultural and ethnic diversity in REDs research is limited, and the consensus may not generalise to non-Western populations.
• Conflict of interest: Several authors have received funding from sports organisations (e.g., IOC, national sports institutes) that have a vested interest in athlete health but also in performance. No industry funding is declared, but institutional biases may exist.
• Clinical tool validation: The REDs CAT2 is introduced but has not yet been validated in prospective studies. Its sensitivity and specificity for detecting REDs are unknown.
• Overemphasis on elite athletes: The statement focuses heavily on elite and Olympic-level athletes. The applicability to recreational athletes, weekend warriors, or non-athletes with LEA (e.g., people with eating disorders or restrictive diets) is unclear.

For someone running their own n=1 experiment to explore whether low energy availability is affecting their health or performance:

What to test

• Intervention: Increase daily energy intake by 200–400 kcal/day, specifically from carbohydrates and fats, while keeping exercise volume constant. Alternatively, reduce exercise energy expenditure by 10–20% (e.g., shorten sessions by 15–30 minutes) while keeping intake constant.
• Dose: Aim for an energy availability of ≥45 kcal/kg fat-free mass/day (the threshold for "adequate" EA). For a 70 kg person with 15% body fat (fat-free mass ≈ 60 kg), this means consuming at least 2,700 kcal/day from food, plus whatever is burned through exercise.
• Comparator: Your baseline state (before the intervention) or a control period where you maintain your usual diet and training.

Minimum meaningful duration

• For metabolic and hormonal changes: At least 14–21 days. RMR and testosterone begin to change within 3–7 days of LEA, but full adaptation and recovery take 2–4 weeks.
• For menstrual recovery (females): 6–12 months of sustained adequate energy intake may be needed to restore menses. Do not expect quick results.
• For performance: 2–4 weeks to see improvements in endurance or strength, though some athletes notice changes within 7–10 days.

What to measure

• Primary metrics:
  • Energy availability: Track daily food intake (using a food diary app like Cronometer or MyFitnessPal) and exercise energy expenditure (using a heart rate monitor or power meter). Calculate EA = (kcal eaten – kcal burned) / kg fat-free mass. Aim for ≥45.
  • Resting metabolic rate (RMR): Measure upon waking, before eating or exercise, using a handheld device (e.g., PNOE, KORR) or estimate using the Harris-Benedict equation. A measured RMR <90% of predicted suggests metabolic adaptation.
  • Body weight: Weigh daily at the same time (morning, after voiding). A stable or slightly increasing weight (0.5–1.0 kg over 2 weeks) may indicate adequate energy intake.
• Secondary metrics:
  • Sleep quality: Use the Pittsburgh Sleep Quality Index (PSQI) or a wearable (e.g., Oura Ring, Whoop). LEA often impairs sleep efficiency and increases night wakings.
  • Mood: Use the Profile of Mood States (POMS) or a simple 1–10 rating of "energy" and "irritability" daily. LEA increases fatigue and mood swings.
  • Performance: Track a standardised workout (e.g., 5 km run time, max reps at 80% 1RM) weekly. A 5–10% improvement after increasing intake is a positive sign.
  • Menstrual cycle (females): Track cycle length and presence of menses. A return of regular cycles (every 21–35 days) after amenorrhea is a strong indicator of recovery.
  • Libido (males): Rate sexual desire on a 1–10 scale weekly. LEA suppresses libido in men.

Key confounds to control for

• Sleep: Poor sleep independently lowers testosterone, impairs recovery, and increases hunger. Aim for 7–9 hours per night and keep sleep consistent.
• Stress: Psychological stress raises cortisol, which can mimic some effects of LEA (e.g., bone loss, immune suppression). Track daily stress on a 1–10 scale.
• Training load: If you increase food but also increase training, EA may not change. Keep exercise volume and intensity constant during the experiment.
• Hydration: Dehydration can lower performance and alter body weight. Drink to thirst and track urine colour.
• Supplement use: Caffeine, creatine, and other supplements can affect performance and metabolism. Keep them constant or eliminate them during the experiment.
• Menstrual cycle phase (females): Hormonal fluctuations