REF-1317 · Gallant et al., 2025 · Sports Medicine · Paper · PubMed

How common is low energy availability in athletes?

Common: 44.7% of 6118 athletes in 46 studies were classed as having it, men (49.4%) as often as women (44.2%). Screen before any weight-loss plan.

Main caveat: From the abstract only. The full text (Springer, paywalled; the repository copy listed by OpenAlex sits behind a bot check) was not read.
moderate evidenceMeasured on trained to elite athletes

Gallant and colleagues (California Northstate University College of Medicine and others) set out to estimate how many athletes have low energy availability (LEA) or relative energy deficiency in sport (REDs), and to give the first secondary analysis of how LEA relates to performance and injury.

Following a registered protocol, they searched PubMed, Embase and the Cochrane databases for studies of the prevalence of LEA or REDs, or of their effects on performance or injury. Fifty-nine studies were included.

Across 46 prevalence studies, 2737 of 6118 athletes (44.7%) were classed as having LEA — 44.2% of the women and 49.4% of the men — and across eight studies 460 of 730 (63.0%) were at risk of REDs.

Athletes with LEA had lower running and endurance performance, training response, coordination, concentration, judgment, explosive power and agility than athletes with normal energy availability, and were more often absent from training through illness.

Whether LEA raised injury risk in general was mixed, but most studies concluded that athletes with LEA had impaired bone health and a higher risk of bone stress injuries. The authors suggest coaches use a questionnaire such as the LEAF-Q to identify athletes at risk early.

Our reading: the direction agrees with the IOC consensus already in the base (REF-341) and with the base's screen-first rule for weight loss (SYNTH-54); what it adds is a prevalence figure that includes men. Without the full text we cannot say how LEA was classified in each study, so the figure is a rough average, not a measured rate.

Protocol

Design: systematic review registered on PROSPERO (CRD42023469253), PRISMA; searches of PubMed, Embase and Cochrane databases (dates not given in the abstract).
Inclusion: articles on the prevalence of LEA or REDs, the impact of LEA or REDs on athletic performance, or their impact on injury.
Included: 59 studies; prevalence of LEA from 46 studies (n = 6118 athletes), risk of REDs from eight studies (730 athletes).
Not in the abstract: how LEA and REDs risk were defined or measured, study designs, sports and competitive levels, risk-of-bias tool, heterogeneity, and whether proportions were pooled by a random-effects model or summed.
Funding and interests: not stated in the abstract; Crossref lists no funder (full text not read).

Key Results

Outcome (abstract)StudiesResult
LEA, all athletes462737 of 6118 (44.7%)
LEA, womenwithin the 4644.2%
LEA, menwithin the 4649.4%
At risk of REDs8460 of 730 (63.0%)
Performance with LEA vs normal energy availabilitynot statedlower run performance, training response, endurance performance, coordination, concentration, judgment, explosive power and agility
Illnessnot statedmore absence from training due to illness
Injury in generalnot statedmixed results
Bonenot statedmost studies: impaired bone health and higher risk of bone stress injuries
Our arithmetic: 2737 / 6118 = 44.7%, so the reported figure is the simple ratio of the counts; whether a weighted or random-effects estimate was also computed is not in the abstract.
No effect sizes, confidence intervals or numbers of studies are given in the abstract for any performance, illness or injury outcome.

Mechanisms

Measured (associations): athletes classed as having LEA performed worse on a range of tests and had more illness-related absence and, in most studies, poorer bone health and more bone stress injuries than athletes with normal energy availability.
Authors' interpretation: LEA arises when training and bodily energy needs exceed intake; when severe it alters metabolic rate, menstrual, immune, bone, protein-synthesis and cardiovascular function (REDs), which lowers performance and raises injury risk; early identification and correction can prevent REDs symptoms and bone stress injuries.
Inference: in observational comparisons, athletes who under-eat may also differ in training load, sport, body composition and stress; the abstract cannot show that LEA caused the performance gap. The bone finding agrees with the causal reasoning of the IOC consensus (REF-341).

Practical Application

How common is low energy availability in athletes? Common: 44.7% of 6118 athletes in 46 studies were classed as having it, men (49.4%) as often as women (44.2%). Screen before any weight-loss plan.
Ask before advising a deficit (B+, from the abstract): with roughly every second athlete in these studies classed as under-fuelled, the base's rule to screen for REDs indicators before any weight-loss conversation stands (SYNTH-54, built on REF-341).
Men too: the male prevalence was no lower than the female; libido, morning erections and low testosterone belong in the screen as much as periods do (REF-341; REF-343: 40% of the men among elite distance athletes had low testosterone).
A bone stress injury is a fuelling question first: most included studies linked LEA to poorer bone health and more bone stress injuries — the base already treats a bone stress injury as a stop-and-refer sign (SYNTH-54; REF-3497).
A falling performance or a stalled block: rule out under-fuelling before calling it overtraining (the direction of this review; REF-344: 18 of 21 overload studies with dietary data found the "overtrained" athletes under-fuelled).
Questionnaires: the authors suggest the LEAF-Q; it is the Low Energy Availability in Females Questionnaire, so it does not cover men (inference from its name; the abstract does not discuss this). Not from this paper; the base's own practice: ask the IOC indicators for both sexes and refer, without assigning a severity (REF-341).

Limitations

From the abstract only. The full text (Springer, paywalled; the repository copy listed by OpenAlex sits behind a bot check) was not read. Definitions and tools for LEA and REDs, sports, levels, study designs, risk of bias, heterogeneity and funding are not in the abstract.
A crude prevalence. 44.7% is the ratio of summed counts across 46 studies with unstated methods and sports; it is not a rate for a given sport, sex or level.
"Meta-analysis" without an effect. The title says meta-analysis, but the abstract reports no pooled effect size for performance, illness or injury; those findings are a direction count, and "mixed" for injury.
Associations. The designs behind the performance findings are not given; differences between athletes with and without LEA do not show that LEA caused them.
Screening tool. The recommended LEAF-Q is built for women, while a male prevalence is reported; how men were classified is not stated in the abstract.
Reporting. The abstract calls this the first secondary analysis of LEA and REDs on performance and injury; the IOC 2023 consensus (REF-341) already maps performance outcomes, by expert synthesis rather than systematic review.
Funding. Not stated in the abstract; no funder in the Crossref record.
How this entry was checked and changed
Why this level

(from the abstract) — a systematic review of 59 studies, which the rubric would place at A- if the pooling were shown; held at B+ because the abstract reports prevalence as a crude count (2737 of 6118) with no confidence interval, heterogeneity or risk-of-bias assessment, does not say how low energy availability or REDs risk was defined or measured, and summarises performance and injury as a direction count of studies whose designs are not given