What we learned from getting our Sweat Test data published in an academic journal
We’ve been Sweat Testing athletes for a number of years now and as a result we’re gathering up a pretty unique and interesting database of information on the composition of athlete’s sweat. Ok, so admittedly it’s a bit of a niche interest but don’t judge me there! Some folks collect model trains or stamps, others collect sweat samples…
Whilst most of the work we do is on a one to one basis - trying to help individuals looking to improve their own performance in a lot of little n=1 experiments - there's sometimes value in zooming out and looking at the data we’ve built up as a whole in order to look for trends and general observations that can help us improve what we do.
About 18 months ago we were talking about all the information we’ve been collecting with a couple of colleagues from the sports science world (academic and Sports Scientist Ben Drury and renowned Performance Nutritionist Dr Mayur Ranchordas) and they suggested that they’d like to take a sample of our data and do some statistical analysis on it to see if it could be published in a peer-reviewed scientific article to contribute to the wider knowledge base on the topic of hydration.
Ben and Mayur were specifically keen on looking at the data we had from working with pro athletes, as data from this group is often not brilliantly well represented in scientific literature due to the obvious logistical and practical issues involved with getting sufficient numbers of pro athletes to take part in research studies.
More often than not, data for sports science research is gathered from college/university level athletes, or other easily accessible volunteers. Whilst results and conclusions drawn from this kind of work are often very informative, it’s not always possible to extrapolate findings from it to more ‘elite’ populations as, by definition, elite and professional athletes are often not like the rest of us.
Elites are often outliers and ‘freaks’ (in the nicest possible sense of the word!). As a result, any chance to study data from them is usually seen as pretty valuable. So our database of sweat test data from thousands of athletes from a long list of pro sports was a bit of a treasure trove ready to be explored from a scientific point of view.
When they looked at our spreadsheets in detail, Mayur and Ben eventually identified that we had data from almost 700 athletes across 5 key sports that met the strict quality criteria needed for inclusion in an academic paper.
- Baseball (Major League Baseball players)
- Basketball (NBA players)
- American Footballers - (NFL players)
- Soccer/Football (Premier League and Football League players)
- Rugby - Aviva Premiership and International team players
We were very happy to let them take a look at the anonymous data to see what they would find out by doing some hardcore number crunching, so we handed it over and let them get cracking.
As well as the athlete’s sweat sodium concentration as measured in our Advanced Sweat Test, we gave the guys access to the subjective questionnaire-based data on things like the athlete's perception of their own sweat rate, cramping history and how much sodium they thought they lost in their sweat. These are all data points we capture as standard during each and every Sweat Test we conduct.
Having recruited some further specialist assistance for the statistical analysis from Dr Nicholas Tiller and Girish Ramchandani from Sheffield Hallam University here in the U.K., the guys fed the data into their computer (my imagination wants this to look like some kind of 1970s NASA super computer, taking up a whole room at the university, but it was probably just a small IBM laptop…), put the kettle on and waited for the results to come back.
Once the computer had done it’s thing, it turned out their were some pretty interesting findings. So, after the not-so-insignificant job of deciphering the numbers and writing it all up into an acceptable format, the Journal of the International Society of Sports Nutrition accepted the paper for submission and published it online here in October. It’s got the typically academic title of “Normative data on regional sweat-sodium concentrations of professional male team-sport athletes”. As our Marketing Director Dave said, not exactly clickbait but we'll take it!
As with most academic journal articles, it’s not exactly ‘easy' reading if you’re not from a scientific background. However if you’re interested in the main things we found out, I’ve tried to summarise them…
1) The average sweat sodium concentration of the male pro athletes in our sample was 46.4 mmol/l (~950 mg/l).
This is very similar to the average figure for our entire athlete database, which numbers in the single-digit thousands. It’s also remarkably similar to the average sweat sodium numbers commonly reported in all sorts of historical literature on sweat composition of normal (healthy) human beings.
This is useful to know as it gives us confidence in our numbers and data collection methods. The results we’re getting are very much in the ballpark of what other independent researchers in this area have found in the past, even when they’ve collected sweat during exercise itself (our sweat test is done at rest).
2) The range of sweat sodium concentrations in our pro athlete sample was 17mmol/l (~350 mg/l) to 92 mmol/l (~1,900 mg/l).
Again, this is very similar to the range seen across our full database and in the wider literature on the topic of human sweat sodium losses. This is another tick in the box that our methods are on par with other ways scientist are measuring sweat sodium loss in athletes across the world.
These findings point to the idea that, when taken as an overall group, professional male athletes probably don’t differ dramatically from the population at large when it comes to how much sodium they lose in their sweat. In that small area of physiology at least, they’re basically just like the rest of us.
3) The athletes tested in North America (i.e. the baseball, basketball and american football players) had slightly higher than average sweat sodium losses than the U.K. based soccer and rugby players.
The difference was not massive. US athletes lost ~50 mmol/l on average, whereas in the U.K. it was ~43 mmol/l. But it was a statistically significant finding, so it’s not likely to have been caused purely by chance.
It’s hard to speculate exactly why that difference might have been observed, so it’s maybe an area where more research might be directed in the future…
4) There was a strong correlation between how much sodium the athletes’ thought they lose in their sweat and their actual sweat sodium concentration - in this group of professional athletes at least.
This is probably the most interesting and important finding from the exercise. As part of the subjective data gathered before measuring their sweat, the athletes we test are all asked to comment on whether they thought they lost a small, moderate, large or very large amount of sodium (salt) in their sweat.
We’ve always been very interested in this because we have long had a hunch that, due to living with the symptoms of it on a daily basis, athletes with large sweat sodium losses might ‘know’ how much sodium they were losing in their sweat intuitively. That’s because they’re likely to have seen salt crusts build up on their skin/clothing after training and competition, or to get muscle cramps or cravings for salty foods after exercise.
In other words the ‘salty ones’ knew that they were salty even without needing to be tested. This lends weight to the idea that questionnaire-based methods (such as our free Online Sweat Test) are a very viable alternative for helping athletes figure out if they might benefit from replacing more sodium using sports drinks and supplements.
Of course, it’s always going to be preferable to actually measure how much sodium you lose in your sweat if you want to truly personalise your hydration strategy, but getting an educated ‘guesstimate’ based on some key clues might be a very decent alternative in situations where cost or logistics make that difficult.
Overall, we feel like we learned a hell of a lot by being involved in the process of publishing a paper in a scientific journal. It’s pretty rewarding to think that the data we’ve worked hard to collect over the last few years might be of use to the wider scientific community, as well as to the individual athletes we’ve tested.
We definitely want to say a huge thank you to Dr’s Ranchordas, Tiller and Jutley, as well as Ben Drury and Girish Ramchandani for doing the heavy lifting when it came to analysing the data and writing up the paper. We’d also like to thank all of the athletes who contributed data (anonymously, of course) when they came to us for a Sweat Test. Every test we do furthers our understanding of the topic of hydration.
Plans are currently afoot to do more research with the guys (and with some other collaborators in the US) to see what else we can learn to keep pushing knowledge forward in this interesting area of sports physiology. So, watch this space for future updates as that works gets underway!