7 min read

Can we measure low energy availability?

The definition of relative energy deficiency in sport (REDs) is that low energy availability (LEA) is the cause of a wide range of symptoms that are common amongst athletes. In the current REDs model, low energy availability is depicted at the centre of a wheel with numerous spokes. Each spoke represents a grouping of symptoms or suggested consequences of LEA. Given LEA is central to the REDs model, this blog asks the question: can we measure LEA?

Infographic for the difficult calculating low energy availability

Challenges defining low energy availability

Of course to provide evidence for this model and to demonstrate that the symptoms are indeed caused by LEA one would need to

Measure energy availability
Decide what is ‘low’ energy availability (LEA) and what is ‘normal’ or ‘adequate’ energy availability and
Decide whether the LEA is problematic or not (this is a criterium that was added in the latest consensus statement on REDs).

In the literature we find a cutoff value or threshold of energy availability of 30 kcal/kg FFM/day. This, for many years served as a dividing line between low and adequate energy availability. This threshold was derived from short-term laboratory-based studies conducted in sedentary women, showing that exposure to ≤ 30 kcal/kg FFM/day for, typically, ≤ 5 days resulted in the alteration of a number of hormonal and metabolic parameters that resemble those of amenorrhoeic females and were consistent with the body entering into an energy preservation state (1).

However, with time we came to realise that reality is a lot more complex than that. Any threshold could be different from person to person, alterations in hormones may not always reflect problems, other factors may also affect these hormones and there are several issues with measuring energy availability. Many of these issues, particularly when trying to assess energy availability in free-living athletes, have been discussed previously (2).

The current REDs consensus statement has moved away from trying to determine energy availability and doesn’t incorporate energy availability anymore as a diagnostic criterion (3). This is curious and astonishing because we can then have no certainty if it is energy availability or something else causing all the symptoms that are attributed to low energy availability, something we discuss in our latest article ‘Does REDs exist?’ published in Sports Medicine (4).

Below we detail a few inter-related problems that make assessing energy availability and defining ‘low’ energy availability problematic.

Laboratory vs. field measures of low energy availability

The formula to define energy availability was developed for use in laboratory settings, where individuals are in a highly controlled environment, the diets are typically custom-made and provided to the participants, exercise is tightly controlled and exercise energy expenditure is measured, and fat-free mass is typically measured by gold-standard methods (such as DXA or underwater weighing).

In free-living conditions things are considerably more chaotic and difficult, and the assessment of energy availability is fraught with error. A large part of the literature on energy availability in athletes (observational studies and cross-sectional studies), seem to completely ignore this important limitation. Assessments of energy availability are presented as accurate or factual data.

Problems with accuracy in assessing components of low energy availability

Any measurement is bound to have a degree of error. The assessment of energy intake (EI) in particular has a very large error of measurement. A recent study has shown that the systematic error of assessment of EI self-reported by athletes, is on average underreporting EI by 19% (5). That’s on average 667 kcal/day under-estimation (roughly equivalent to a full meal), and means that it is not all that rare that individual assessments of individuals can be off by 40% or so.

The other measurements required for the calculation of energy availability are energy expenditure of exercise and fat-free mass. It is well known that measurements of energy expenditure can also be inaccurate and have large errors and fat-free mass as well, even when gold standard techniques are used. For example, if DXA scans are not strictly standardised, results will vary. It is also known that different DXA scanners will provide different results. Without getting into too much details on how the error measurement of exercise energy expenditure (EEE) and fat-free mass (FFM) can add up, it is obvious that if you calculate energy availability (EA) from 3 numbers that each can have a substantial error, it is impossible to base a diagnosis on this outcome.

It is also worth highlighting that even when considering only the lack of accuracy in the assessment of EI, the majority of studies assessing EA in athletic populations (see [6] as an example) have likely reported a significantly inflated prevalence of LEA.

The formula problem

The formula for assessment of EA appears to be rather straightforward:

EA= (EI-EEE)/FFM

Assuming the assessment of EI, EEE and FFM are accurate, there is hardly any room to make mistakes, right?

Not quite: there has been different iterations of this formula (1), which the vast majority of studies on energy availability appear to have ignored, and it is rarely specified which formula has been used to perform the calculations. A key difference between these formulae lies with the EEE variable. In two different iterations EEE may be calculated as Total EEE or Net EEE.

Total EEE is the total amount of energy used during exercise. Whereas Net EEE is the total amount energy minus the theoretical contribution of resting metabolic rate (RMR) for the duration of exercise.

This means that EA values calculated using Total EEE are lower than those using Net EEE, and the difference gets bigger with longer durations of exercise. Therefore, we may have inadvertently been comparing apples to pears when trying to understand different levels of EA reported in different studies, making the use of a universal threshold problematic based on current literature.

"we may have inadvertently been comparing apples to pears when trying to understand different levels of [energy availability] reported in different studies"

The NEAT problem

The body does not differentiate between energy expended in exercise or energy expended in other energy-consuming tasks, such as walking to the supermarket, running to catch the bus or fidgeting, among many others. This component of total daily energy expenditure is called ‘non-exercise activity thermogenesis’ (NEAT), and it can be a significant contributor to daily energy expenditure. There is no consensus on what is the ‘allowance’ of NEAT in the current EA formula, before it is considered significant, and also how it should be assessed.

The individual variability problem

Even if we could measure EA with 100% precision, to date we have no information on how different individuals may respond to different levels of EA. Based on the best laboratory-based studies, it seems possible that there exists significant inter-individual variation to the response of the same amount of energy availability, therefore, utilising a universal threshold for EA appears to be premature.

The duration problem

Even if we could measure EA with 100% precision, knew to what extent NEAT should be considered in the calculation, and what is the inter-individual variability to varying degrees of EA, we do not know for how long EA should be assessed before it is representative of what is happening in an individual’s life or it can be linked to ‘negative’ effects, other than affecting the concentration of a few hormones.

Assessing EI and EEE is rather time consuming and a burden to the athlete and nutritionist. Short-term (e.g. 3-7 day) assessments of EA may not be representative of what is happening long-term in an individual’s life, and it is unknown how long it may take until it has any significant effects in health or performance outcomes.

Conclusion: is the low energy availability concept completely useless?

This is just a simple and non-exhaustive list of problems associated with measuring EA and defining LEA, particularly in field conditions. The list of problems is longer, but these are possibly the most significant and easier to explain.

Does it mean that measuring EA is completely useless? No. In some cases it may be useful to detect significant problems with fuelling and LARGE difference between expected and observed EA values, which may be suggestive of what may be sub-optimal fuelling, and also may be interesting when considering patterns of EA in athletes (7).

However, all these problems highlight that it is nearly impossible to know if EA (LEA in particular) is the only or main source of the problem in athletic in the REDs and Triad models, because we cannot really measure it accurately. These problems also highlight how that we must not take EA values at face value, and carefully consider how they have been measured and calculated.

Most importantly we should always have these limitations in the back of our mind when we read papers on REDs…and definitely in practical situations: is it really low energy availability? How can we know for sure that it is LEA if we cant measure it?

Reference

Areta, J. L., Taylor, H. L., & Koehler, K. (2021). Low energy availability: History, definition and evidence of its endocrine, metabolic and physiological effects in prospective studies in females and males. European Journal of Applied Physiology, 121(1), 1–21.
Burke, L. M., Lundy, B., Fahrenholtz, I. L., & Melin, A. K. (2018). Pitfalls of Conducting and Interpreting Estimates of Energy Availability in Free-Living Athletes. International Journal of Sport Nutrition and Exercise Metabolism, 28(4), 350–363.
Mountjoy, M., Ackerman, K. E., Bailey, D. M., Burke, L. M., Constantini, N., Hackney, A. C., Heikura, I. A., Melin, A., Pensgaard, A. M., Stellingwerff, T., Sundgot-Borgen, J. K., Torstveit, M. K., Jacobsen, A. U., Verhagen, E., Budgett, R., Engebretsen, L., & Erdener, U. (2023). 2023 International Olympic Committee’s (IOC) consensus statement on Relative Energy Deficiency in Sport (REDs). British Journal of Sports Medicine, 57(17), 1073–1097.
Jeukendrup, A. E., Areta, J. L., Van Genechten, L., Langan-Evans, C., Pedlar, C. R., Rodas, G., Sale, C., & Walsh, N. P. (2024). Does Relative Energy Deficiency in Sport (REDs) Syndrome Exist? Sports Medicine.
Capling, L., Beck, K., Gifford, J., Slater, G., Flood, V., & O’Connor, H. (2017). Validity of Dietary Assessment in Athletes: A Systematic Review. Nutrients, 9(12), 1313.
Logue, D., Madigan, S. M., Delahunt, E., Heinen, M., Mc Donnell, S.-J., & Corish, C. A. (2018). Low Energy Availability in Athletes: A Review of Prevalence, Dietary Patterns, Physiological Health, and Sports Performance. Sports Medicine, 48(1), 73–96.
Taylor, H. L., Garabello, G., Pugh, J., Morton, J., Langan-Evans, C., Louis, J., Borgersen, R., & Areta, J. L. (2022). Patterns of energy availability of free-living athletes display day-to-day variability that is not reflected in laboratory-based protocols: Insights from elite male road cyclists. Journal of Sports Sciences, 40(16), 1849–1856.