Get access to everything we publish when you sign up for Outside+.
Let’s start with some definitions, like we’re doing a speech in the second grade. Trust me, you’ll see why in a second, and then we can all share some Dunkaroos after.
“Overreaching” is an offset between stress and recovery, resulting in reduction in performance following a period of increased training. Athletes can recover in days to weeks from overreaching, whereas overtraining syndrome is a series of long-term maladaptations amounting to a medical condition that may take months for full recovery.
“Low energy availability” is an offset between energy expenditure and energy intake, which can result in negative impacts to metabolism, sex function, sex drive, bone health, the endocrine system, and the nervous system, among numerous other physiological issues constituting Relative Energy Deficiency in Sport (RED-S).
So both overreaching and underfueling result from excess stress relative to recovery practices (including rest and fueling) and, when taken to the extreme, can cause similar downstream perturbations to physiological metrics.
That creates a problem – how can you tell which is which? Or if it’s both? That’s one of the most important questions for athletes and coaches, since it determines what constitutes smart training and recovery, and what constitutes thinly veiled self-destruction.
You’ve made it through the definitions, so have some Dunkaroos. You earned it! Nap time optional, but encouraged.
Imagine those definitions as a venn diagram.
Not all cases of overreaching involve underfueling, but many do. That overlap was examined in a fantastic 2021 review article in Sports Medicine that looked at 57 studies on overtraining and 88 studies on low energy availability, finding that overtraining and low energy availability share almost all symptoms due to how they both spur “under-recovery.” That was my favorite study of 2021, a feat of ambitious research, built on top of decades of knowledge (I wrote about it here and podcasted on it here). However, there was a slight catch.
The authors used the term “training-overload/OTS [overtraining syndrome],” reviewing studies that found a sustained and significant decline in performance outcomes. In studies that met their criteria, most involved low energy/carbohydrate availability:
“Taken together, 86% of the training-overload/OTS studies reporting dietary outcomes (18 out of 21) showed reduced EA (n = 14 studies) and/or CHO availability (n = 4) between treatment groups or between pre and post concurrent with symptoms consistent with both OTS and RED-S.”
That 86% number is a massively important finding – the answer to avoiding overtraining is often related to fueling. But 86% is not 100%. Thus, even with a relatively stringent definition of what constituted training-overload/OTS, there were cases when dietary practices were unrelated to overtraining.
That makes sense intuitively. No matter how much an athlete eats, if they increase their mileage up to 200 miles per week (or the personal equivalent relative to background), their physiology will be blown to smithereens. But how can we get to more nuance? It’s not as simple as “eat enough, always, or you’ll get physiologically f**cked,” even though that’s how I often frame it. In reality, we all know athletes that sometimes underfuel, or sometimes overtrain, and don’t end up physiologically f**cked. What about those cases?
That’s where an amazing new study comes in.
The 2022 systematic review in Sports Medicine by Megan Kuikman, Alexandra Coates, and Jamie Burr built off the 2021 study to dig deeper into the tricky interplay. They noted that it is possible that “underperformance from training stress (overreaching) and low energy availability are independent factors that can coexist.” How often are they seen independently in the existing research?
Talk about another ambitiously badass study! The 2021 review in Sports Medicine was this wildly cool leap forward, taking an audacious hypothesis seen in real-world empirical observations, and finding definitive proof for it in a huge dataset. That review turned something that physiologists and coaches used to whisper about into something approaching conventional wisdom. As I put it in my summary article, “Eating enough won’t prevent every case of training-overload/OTS, but it may prevent many of them.” So cool, and so important!
Enter the 2022 review, also in Sports Medicine. It jumps in with a different methodology and inclusion criteria, addressing some open questions, and making things a bit more complicated. The scientific method is tough on tidy conclusions, but wonderful for developing more comprehensive knowledge. It’s not throwing turds in the punch bowl, it’s finding the turds that might already be floating around before we take a big sip.
The 2022 review looked at 56 studies with training blocks at least 2 weeks in length (but less than 1 year) of increased or maintained training intensity/volume. The studies also had to measure at least 2 markers of low energy availability. There were two comparison groups from the studies:
- For athletes that measured significant reductions in performance, comparisons were made for their LEA markers
- For athletes that measured 2 or more LEA markers, comparisons were made for their exercise performance
Markers of low energy availability included amenorrhea (the absence of a regular menstrual cycle), low energy availability from food and training records, decreased bone mineral density with a prior DXA scan or z-scores less than -1.0, suppressed metabolic rate, changes in certain biomarkers, and a number of other factors. Performance was evaluated via aerobic endurance tests.
There was substantial heterogeneity in study populations, durations, measures, and interventions, which will be a key consideration later. I’m foreshadowing like M. Night Shyamalan. Also, the study methodology could have missed some cases of low energy availability since not all metrics were evaluated in each study, or overreaching since it can be complex to diagnose. And given the lack of standardization across studies (plus the fact that some studies were asking different questions entirely), it’s hard to draw definitive conclusions.
14 of the studies found underperformance at the end of the training block. Of those, half included 2 or more makers for low energy availability. Of the athletes who underperformed, there were reductions in relative resting metabolic rate, fat mass, and hormone leptin.
18 studies found athletes with at least 2 markers of low energy availability. Of those, 7 showed reductions in performance. Across all of the pooled study populations, exercise performance was unaffected in athletes with at least 2 markers of low energy availability. The authors put it all together in a somewhat unsatisfying (but very important) conclusion. I’m going to paste it all, since it will tee up the next discussion:
“We found that a majority of training periods (~60%) with maintained or increased training loads did not result in impaired athletic performance or evidence of LEA. In contrast to our hypothesis, there was only evidence of LEA in half of the studies in which athletes presented with reduced performance at the end of the training block. Furthermore, the presence of ≥2 markers of LEA following training was not accompanied by changes in performance, even when accounting for differences in the length of the training block and the exercise performance test used.”
What It Means
In other words, underfueling and overreaching can coexist, but more often than not in this particular literature review, they don’t. On the surface, that sucks for tidy conclusions. I want athletes to eat enough always, because we know the long-term, terrible physiological effects that can result from underfueling. To support that statement, it’s easy to describe to athletes just how much training will suffer if they underfuel or overtrain. Eat enough always and don’t crush your body in training… OR ELSE.
But contrast that with what the athletes may see in practice. Maybe some of their college teammates treated their bodies poorly for a year or two, with underfueling and/or overreaching. Sure, some of those athletes probably ate metaphorical shit, with underperformance and stress fractures. But maybe some also became champions. It’s easy to ignore our science-backed pleas to never have low energy availability if our models don’t accommodate empirical evidence that they see with their own eyes.
And that gets to a key point. MOST IMPORTANT STATEMENT ALERT. Sustained low energy availability will backfire in a major way. A fantastic 2019 review article in the Current Opinion in Endocrine and Metabolic Research journal described how failing to eat enough food can cause a cascade of negative physiological effects. Even small within-day deficits can hurt performance and health.
Failure to eat enough ruins countless careers. In some cases, it can undercut the long-term life trajectory of athletes far off the roads, track, and trails. So why wasn’t low energy availability associated with negative performance impacts in every study in the 2022 review?
Context and Key Points
The authors highlighted three limitations that help explain the offset. All are related to the heterogeneity of the data, and they are key in how we honestly talk about these issues in running.
First, the average study duration was around 6 weeks. The authors point out that there wasn’t an association between study duration and outcomes, but it’s safe to assume that the negative consequences to the nervous and endocrine systems can take longer than a year to fully manifest themselves. Low energy availability is lighting a fuse, and often that physiological bomb just takes a bit longer to go off.
Second, a very small subset of the study population was women. As the authors say, “of the studies that did include female athletes with ≥ 2 markers of LEA, all demonstrated reductions in exercise performance.” So even in short study protocols, low energy availability started hurting the performance of female athletes. It’s likely that the same consequences would be seen in male athletes with longer time horizons. Female athletes might have almost no wiggle room to underfuel training due impacts to the endocrine system.
Failure to eat enough will undercut performance, running health, and overall life health.
Finally, the studies often do not examine how athletes adapt to training interventions over the types of time horizons that are meaningful to long-term growth. Let’s conclude with a study that the authors highlighted, from 2014 in the Medicine and Science in Sports and Exercise journal on female junior swimmers. There were two study groups: one group was experiencing low energy availability at the start of the study, and one group wasn’t. Both had performance decrements after overload training. That’s classic overreaching, and it’s expected in heavy training periods. But here’s the key part: only the group that was fueling enough progressed to higher levels after the taper. Low energy availability undercut adaptation.
I think that this study underscores an essential point in how we frame these issues. Low energy availability is not always a physiological cliff. It’s more like slowly boiling water. If your legs wind up in the water inadvertently, that may be fine, even though it’s risky and could cause health problems depending on the water’s temperature. But stay in too long, and your leg is guaranteed to melt off. Adaptation and overall health will go to shit, it’s just a matter of an uncertain time horizon.
Don’t confuse the initial period and the long-term consequences. As the college athletes may have seen, that initial period might not result in performance decreases at all. Long-term, though, the science is unequivocal. Failure to eat enough will undercut performance, running health, and overall life health.
The 2022 review is an essential leap in knowledge because it provides data to address the cognitive dissonance that you may experience if you spend time in this field. Almost every study and review article says that low energy availability is atrocious for health and performance. But that is contrasted by what athletes may see in the real world, with some prominent athletes talking about periodizing their fueling approaches, or successful teammates who might not treat their bodies as well as they should at the dinner table. If these anecdotes existed without scientific support, I think that would indicate that the scientific literature was missing something. This review helps fill that gap. Because in some cases, in extreme moderation, particularly for male athletes, low energy availability might not lead to reduced performance.
As the authors say, “Short periods of LEA may be periodized within an athlete’s training cycle without compromising performance but should be followed by periods of adequate energy intake in order to minimize the risk of injury and illness.”
Or to put it in my vocabulary, based on what I have seen in coaching and in longer-term research: Low energy availability may be fine before it starts to ruin a life.
The line between those two states is blurry, with the benefits of short-term low energy availability minimal in the best of times, and catastrophic in the worst of times. My coaching take: stay away from the line and don’t risk detonating your athletic future.
I think that the hardest part of underfueling and overreaching is not about the physiology anymore, since we know where all of these roads end with enough time. My hypothesis is that the hardest problem is related more to behavioral psychology.
An athlete sees someone have success that happens to coincide with fueling and/or training unsustainably, or maybe they have success themselves. Due to availability bias, they attribute the success to those unsustainable practices, rather than the tougher-to-see mix of genetics and zoomed-out training history. That sets an association between those behaviors and peak performance that is superstition masquerading as science. And those superstitions lead to premature peaks, short careers, and health issues that can last decades.
Eat enough, always.
Not because you’ll be immediately punished if you don’t.
But because you’ll be rewarded long-term if you do.
David Roche partners with runners of all abilities through his coaching service, Some Work, All Play. With Megan Roche, M.D., he hosts the Some Work, All Play podcast on running (and other things), and you can find more of their work (AND PLAY) on their Patreon page starting at $5 a month.