In this three-part series, I explore 10 different research questions that I feel sports science could make a big difference by attempting to answer—and in many cases, is close to doing so. In Part 1 and Part 2, the questions I explored were:
- Is a low-carb, high-fat diet effective for athletes?
- Is caffeine really ergogenic for everyone?
- Are isometric loading exercises as effective as eccentric loading exercises for hamstring injury prevention?
- What effect does the gut microbiome have on athletic performance?
- Can we develop real-time markers of exercise adaptation?
- Can we use genetic testing to predict talent?
- Do sports supplements have an additive effect, or is there a ceiling?
Obviously, I have my own biases, and some of these areas are from the fields in which I hold a strong interest, but I have tried to cast the net as wide as possible. For each question, I’ve provided:
- A brief review of what we know so far.
- Why it’s important to know more.
My expectation is that, over the next 10 years, we will get closer to more concrete answers in many of these.
Are the ‘Proven’ Effects of Ergogenic Aids and Training Interventions the Same for All Populations?
How science works is that you recruit a group of people—commonly termed “the sample population”—and then conduct your research intervention on them. Because researchers often require subjects who are somewhat similar in order to minimize sources of variation within the results, this often leads to certain groups of people being underrepresented within sports science research. As an example, men and women mightmetabolize caffeine differently, and females potentially metabolize caffeine differently at different stages of their menstrual cycles. As a result, most researchers tend not to recruit females into research exploring the use of caffeine, because they can’t easily figure out at which stage of the menstrual cycle their subjects may be at, or whether their use of oral contraceptives is affecting the results.
Is this a problem? Simply put, yes. Approximately half of the population of the world is female, and, accordingly, roughly half of all elite athletes are female. Yet, in many cases, we don’t fully understand how various interventions affect this sizeable population because they are so underrepresented within sports science research. Indeed, a recent study reported that less than 40% of subjects within sports science research were female.
Similarly, elite athletes are, by definition, quite rare. Very few individuals have the ability and luck to be able to perform at the highest level, and when they do, they’re unlikely to want to take part in research studies that may harm their performance. As a result, researchers find it very hard to conduct research, especially intervention studies, on elite athletes; consequently, elite athletes are comparatively underrepresented within sports science research.
Elite athletes are unlikely to participate in research studies that may harm their performance, says @craig100m. Share on XIs this problematic? Well, we don’t really know as it hasn’t been extensively studied, but there is at least some research that suggests elite, well-trained athletes gain more of a benefit from caffeine, and less of a benefit from other ergogenic aids such as beetroot juice. This means that, if you’re working with an elite athlete, in many cases the research on which you’re basing your decisions is perhaps not valid in the athletes you coach.
This in and of itself isn’t necessarily a major problem; sports science doesn’t exist just to enhance athletic performance, but can also be a method to improve the health of a wide range of individuals by guiding training program design and nutritional interventions. But here, again, we can run into problems; because most researchers are based at a university, there is a tendency to recruit university students to exercise intervention studies—and these subjects don’t necessarily accurately reflect the wider population.
All told, we clearly need more research in underrepresented subjects within sports science. This is especially true for female subjects, notwithstanding the issues researchers face in controlling for the menstrual cycle, and elite athletes, again keeping in mind that the recruitment of such athletes can be difficult. By being able to understand these, and other, populations, we will be better able to make evidence-led recommendations, and support practitioners who work closely with these people, enabling them to (hopefully) make better decisions.
Why Does This Matter?
Many different populations of people, especially females and elite athletes, are underrepresented within sports science research. Crucially, some evidence indicates that both groups respond differently to certain interventions, suggesting a need for more targeted research on these populations in order to better extrapolate the data from current studies and enhance their performance and health.
Can We Predict Exercise Response?
When coaches give a training program to their athletes, they are essentially making a prediction, stating that “I believe this is the best training program for you at this time.” If the training program leads to improvement in the athlete, then the coach is seen to be successful—even though we don’t know if the athlete could have gained greater improvements from a separate training program. Conversely, if the athlete doesn’t improve, then the coach can alter the training stimulus for a second training block, with the updated prediction that “I now believe that this training program is the best for you at this time.” As a result, happening upon the optimal training program for a given athlete is often a process of trial and error, with many things tested, and the ones that are perceived to work sticking, along with a large helping of luck.
One of my interests is in trying to understand whether we can predict this exercise response; that is, can we determine how much someone will improve with an exercise training program before they undertake that training program? If we can gain the ability to do this, then we can remove the trial and error process, and match athletes to the training most suited to them.
If only it were that easy.
First, a little bit of background. Since the mid-1980s, it has been demonstrated through research that not everyone gets the same improvements from exercise. Most famously, this was shown in the HERITAGE Family Study, where researchers recruited 720 subjects to a 20-week aerobic training program, putting them through a wide variety of pre- and post-training tests. Here, the results showed that, while on average, VO2 max (a measure of aerobic fitness) improved by around 380 mL O2, some subjects improved by over 1000 mL O2, while others showed no improvement (and, in actual fact, appeared to get less fit due to training—which makes no sense and is most likely measurement error).
This was also true for various health markers tested for, such as fasting insulin; most people improved, but some more than others, while others got worse. Similar results have been shown following resistance training: after a 12-week training block in one study, the mean improvement in 1RM was 54%, but varied from 0% to 250%.
The cause of the variation between individuals in response to a training program is as yet unknown, says @craig100m. Share on XIt’s clear, therefore, that there is the potential for considerable variation between individuals in response to a training program. As a result, a big question that needs answering here is “What are the causes of this variation?” We can essentially boil these causes down to two factors: “true” and “false.” “False” factors refer to things such as measurement error and random biological variation, which make it look like there is variation, when actually there isn’t (if you’re interested, this is probably the best paper on the subject).
How much of the variation in response to a stimulus is “false” is open to debate. Nevertheless, we also know that there are a number of aspects that lead to “true” (i.e., real) inter-individual variation between subjects in response to a training stimulus. In a 2017 paper, I categorized these as genetic, environmental (i.e., non-genetic), and epigenetic, and I wrote about these for SimpliFaster here. Genetics play an important role in how much an individual responds to exercise, with studies tending to find that around 50% of the variation between individuals in exercise-related traits is due to heritable factors. The other roughly 50% is therefore down to aspects such as nutrition (getting enough energy, protein, and micronutrients), sleep, and psycho-emotional factors, such as the stress levels of an individual.
In theory, if we can ensure that everyone achieves the optimal environmental factors, such as getting enough sleep and adequate nutrition, then that should help to maximize adaptation to a given training program. This leaves us with genetic factors, which, explaining around 50% of the variance in response to exercise, are clearly important; if we could understand which genetic factors explain the variation in response, and we know the genetic makeup of a given athlete, could we use this information to predict the training response?
There isn’t a great deal of research in this field, which is why I’ve identified it as a potentially important route for future research. Some studies have looked at individual genes, such as ACTN3, the famous “speed gene.” Here, it appears that individuals with a certain type of this gene respond better to high-load resistance training (i.e., lifting heavy weights), and it may also play a role in post-exercise recovery and risk of injury. A few other studies have looked at the impact of other individual genes, but the relative effect size of any individual gene on training adaptation is likely to be small. Instead, we need to identify an increased number of genes, and combine them into a single score
This is what a study in which I was an author did; here, we used the results of 15 different genes to determine whether people would respond better to high- or moderate-intensity resistance training. We then gave around half the subjects the “correct” training, and half the “incorrect” training, for an eight-week period. After the training period was completed, those who had undertaken the “correct” training, as determined by their genetics, demonstrated around three times the improvement in a countermovement jump test than those who had undertaken the “incorrect” training.
These results were both praised and criticized in equal measure, and it’s important to keep in mind that they require replication. However, they do suggest there is promise to such an approach, especially when we consider that other researchers have shown the use of similar methods to predict the response to aerobic training. Like using genetic testing to predict talent (explored in Part 2 of this series), there are ethical concerns regarding the use of genetic tests for training prescription.
It’s not just genetics that holds promise in this area. A second potentially promising biomarker is that of microRNA (miRNA). In order to adapt to exercise, our body has to produce new proteins. These proteins can themselves drive adaptation, or form part of a new structure; for example, in skeletal muscle hypertrophy, the body produces proteins that form part of the muscle, allowing it to grow.
At the cellular level, these proteins are produced from DNA. Our body “reads” this DNA, creating messenger RNA (mRNA), which travels to the ribosome, which “reads” the mRNA and creates the new protein. miRNA appears to affect this process by breaking down or destabilizing the mRNA before it can be read (among other processes), altering the expression of a given protein from a given gene. miRNAs appear to impact the adaptation to training.
As an example, researchers from a 2011 study recruited 56 men to a 12-week resistance training program, comprised of five resistance training sessions per week. They then identified the top and bottom 20% of responders in terms of increases in muscle mass, and compared differences in miRNA signature between the two. They found that three miRNAs were downregulated in low responders, and one miRNA was downregulated in the responders.
The ability to predict the response to exercise is an area requiring further research, says @craig100m. Share on XIf we can better understand whether baseline (i.e., pre-training) levels of miRNA affect the response to a training stimulus, then miRNA profiling might be useful, although at present it requires either a blood test or muscle biopsy, which may not be palatable to all. Nevertheless, the ability to predict the response to exercise is an area requiring further research, as doing so should enhance the training process quite substantially. A further understanding of the contributors to variation in training response is required here, as is the development and validation of predictive panels utilizing this information.
Why Does This Matter?
Designing training programs is essentially a matter of prediction: Coaches match athletes to the training techniques they believe will yield the greatest improvements, and then refine via trial and error. If we can predict the response to a certain type of training before that training occurs, then we may be able to more readily match athletes with the training type that will promote the greatest adaptations, and hence improve performance to a far greater extent.
Are Low Doses of Caffeine Optimally Ergogenic?
I’ve already discussed in this series how caffeine is the most established, and most used, performance-enhancing drug in sport. And, best of all, it’s legal. However, even with all the research conducted on caffeine, and even with the wide use of this ergogenic aid, there are still a number of unanswered questions regarding the practical side of its use by athletes.
One of these is the impact of individual variation on caffeine, which I covered in Part 1 of this series, as well as in this paper. A second one is whether habitual caffeine intake reduces the subsequent performance-enhancing effects of caffeine (which I discussed in this paper); the surprising thing is, we don’t really know. Finally, we don’t necessarily know which dose of caffeine is optimum for performance—and we likely never will know, given how much variation there will be between individuals as to what an optimum caffeine dose is.
Nevertheless, most caffeine guidelines suggest that the optimum caffeine dose is roughly 3-6 mg/kg of body weight, with no additional benefit of doses greater than 9 mg/kg of body weight. In 2014, Lawrence Spriet published a hugely influential review paper on the impact of lower doses of caffeine—typically 3 mg/kg or less—on performance. The main finding was that these low doses of caffeine, while not as extensively studied as higher doses, likely did exert ergogenic effects, most notably in aerobic endurance events.
What I’m interested in understanding is whether these low doses of caffeine offer similar performance-enhancing effects as more typical higher doses of caffeine. This is a potentially important question; high doses of caffeine can exert a number of side effects that may negatively impact sports performance, such as increased anxiety, gastrointestinal discomfort, and poor quality sleep following training or competition, which may negatively affect recovery. If these lower caffeine doses are as ergogenic as higher doses, then athletes susceptible to these negative side effects can just take less caffeine for the same performance improvement.
Conversely, if any athlete is using caffeine to enhance their performance, then they want the caffeine to increase performance as much as possible. For example, a runner before a competition could choose between 2 mg/kg and 5 mg/kg for their caffeine dose. If both lower and higher doses are optimally ergogenic, then they can pick the dose based on preference. However, if 2 mg/kg, while ergogenic, is not as performance-enhancing as 5 mg/kg, and they can tolerate this higher dose, then they should choose that higher dose.
The optimal dose of #caffeine will likely vary from athlete to athlete and between events and sports, says @craig100m. Share on XSo, are there any differences between lower and higher doses of caffeine, in terms of their performance-enhancing effects? It’s hard to draw really firm conclusions from the research. If we return to Spriet’s seminal paper on the topic, I count 14 studies that used a low caffeine dose and utilized a performance test (as a brief aside, I always prefer it when caffeine studies explore the impact of caffeine on some aspect of performance, such as time to cover a distance, time to exhaustion, or similar, as opposed to measuring fat oxidation rates or ratings of perceived exertion). Of these 14, only four directly compared a low caffeine dose with a higher caffeine dose, and of these four, we get contrasting results—two find that increasing the caffeine dose enhanced performance to a greater extent, while two found that both the lower and higher caffeine doses enhanced performance to the same extent.
In short, there is a relative lack of trials comparing low and higher caffeine doses for their respective performance-enhancing effects, and these trials often have conflicting results. By increasing the number of studies exploring this question, we should getter a better idea of which type of caffeine dose is most optimal for athletes, allowing us to better inform their pre-training and pre-competition caffeine strategies.
Why Does This Matter?
Because caffeine is such as well-established and well-replicated (and legal!) performance enhancer, many athletes consume it. However, we’re not quite clear on the optimal dose yet, and, more realistically, it’s likely that the optimal dose will vary from athlete to athlete, and between events and sports. Recent research has shown that lower doses of caffeine—defined as 3 mg/kg or less—can be ergogenic, but it’s not yet fully clear whether such doses are as ergogenic as the more commonly recommended 3-6 mg/kg. Fully understanding this will enable athletes to be better informed as to the optimal caffeine dose for them, enhancing performance.
There’s Still Some Way to Go
While it is tempting to think that we pretty much know all there is to know within sports science, hopefully some of the points I’ve raised in this series demonstrate that we have some way to go before this is true. This is a good thing: The use of sports science has been instrumental in enhancing performance over the last 20 to 30 years, and has even spilled over into improving the health of non-athletes. Given the progress we’ve made so far, further enhancing our knowledge in these areas will allow us to improve health and performance to an even greater extent.
The use of sports science has been instrumental in enhancing performance over the last 20-30 years, says @craig100m. Share on XFinally, this list isn’t exhaustive, and represents areas in which I have the most interest. I’d love to hear from you as to the questions you’d appreciate gaining some answers for from the field of sports science.
Since you’re here…
…we have a small favor to ask. More people are reading SimpliFaster than ever, and each week we bring you compelling content from coaches, sport scientists, and physiotherapists who are devoted to building better athletes. Please take a moment to share the articles on social media, engage the authors with questions and comments below, and link to articles when appropriate if you have a blog or participate on forums of related topics. — SF
