What does the evidence show?
The evidence in support of carbohydrate drinks is flimsier than you might think—which is why the results of a new study are worth paying attention to
The hottest topic in the sports drink world these days is whether hydrogels or isotonics are better for helping you rapidly absorb carbohydrate into your system. This debate is just the most recent in a long line of breathlessly hyped breakthroughs: slow-burning “superstarch,” potassium-packed coconut water, brain-boosting theanine, and heck, the whole concept of electrolytes as something you need to drink during exercise.
It’s entirely possible that some of these things really do give you an infinitesimal edge—but it would be a big stretch to say that any of them are solidly backed by research at this point. And that’s a problem that hasn’t escaped the industry’s critics. Back in 2012, the British Medical Journal published a scathing takedown of sports drink research, pointing out the general lack of evidence behind the industry’s performance-boosting claims, and the numerous problems riddling the evidence that did exist: shoddy study design, unrealistic outcome measures, potential bias from industry funding, and so on.
In general, I’ve cheered these sorts of takedowns, and sometimes participated myself in poking holes in overhyped or unjustified claims. But the BMJ piece left me feeling uneasy, because the overall impression it conveyed was that sports drinks probably don’t work at all—not just the latest fancy innovations, but the whole concept of taking in easily digested carbohydrate during prolonged exercise. And that struck me as a little overzealous.
Stripped of frills, the basic claim of sports drinks is that during prolonged exercise—longer than about 90 minutes, say—ingesting some carbohydrates will provide your body with fuel to sustain a higher level of performance for longer. Based on what we’ve learned since the 1960s about how the body functions during exercise, this claim makes perfect sense. In fact, it’s so seemingly obvious that scientists rarely bother to test it. And, as the BMJ piece pointed out, when you look closely at the old studies that everyone relies on, it’s easy to poke holes in their methodologies.
That’s why I think it’s worth taking a look at a newly published study in the Scandinavian Journal of Medicine & Science in Sports that explores one of the common criticisms of sports drink research. I’ll reveal the conclusions right away: the results suggests that sports drinks enhance endurance performance. This is not a bombshell, but it might help reassure skeptics that, regardless of what they think of hydrogels and superstarches, the underlying principle of sports drinks is sound.
The question the study set out to investigate was whether your pre-exercise meal determines sports drink effectiveness. Scientists generally like to control as many variables as possible in their studies, so in sports drink research they often ask their subjects to show up in the lab after an overnight fast, without eating breakfast. This puts all the subjects on an even playing field, nutritionally speaking, but it’s not a very realistic scenario. If sports drinks only help you when you’re half-starved, then they’re not really that useful.
Researchers at the Federal University of Pernambuco in Brazil put nine cyclists through a series of four identical exercise protocols: 105 minutes of moderately hard riding at a constant pace (around lactate threshold), followed by a 10K time trial. That design roughly mimics the dynamics of a long road race, and gives the researchers an opportunity to compare metabolic responses to a predetermined pace, as well as assessing actual race performance. The four conditions were:
- No breakfast, placebo (artificially sweetened) sports drink
- No breakfast, real (8 percent maltodextrin added) sport drink
- Breakfast, placebo sports drink
- Breakfast, real sports drink
The breakfast contained 824 calories, two-thirds of which was carbohydrate, and was consumed three hours before exercise. The sports drink was consumed in doses of 2 ml per kg of bodyweight (about 5 ounces for a 160-pound person) every 15 minutes during the tempo ride and at the halfway mark of the 10K race.
A few details on the drinks: the placebo was a mix of water, orange flavor, yellow dye, and artificial sweetener. The “real” drink was the same thing with added maltodextrin, a mostly tasteless carbohydrate. The researchers did pilot testing to ensure that people found the two drinks “indistinguishable in smell, flavor, sweetness, color and viscosity”—and sure enough, the subjects in the actual study were unable to guess when they got the real drink versus the placebo. They used maltodextrin because it’s ideal for secretly adding to drinks without changing the taste, but it works pretty much exactly like glucose, so there’s nothing special about the drink recipe.
The results weren’t quite what the scientists expected. Previous research (and logic) suggested that the sports drink might give a bigger boost when the cyclists had fasted. Instead, there didn’t seem to be much difference: taking a sports drink helped with or without breakfast. Without the sports drink, in contrast, skipping breakfast definitely had a negative effect. Here’s what the power outputs during the 10K race looked like; FAST means no breakfast, FED means breakfast, PLA means placebo drink, and CHO means carbohydrate drink:
The average power with the sports drink was 198 watts with breakfast, 197 watts without—so no real difference. With breakfast but no sports drink it was 173 watts; with no carbs at all it was 154 watts.
The pattern through the race is also interesting. At the start of the 10K race, the breakfast-plus-placebo group is right with the sports drink groups. But as the race progresses—and fuel stores presumably get used up—they fall farther and farther behind as their fake drink fails to replenish them.
Of course, by the time the 10K starts, they’ve already been cycling for 105 minutes. The data from that steady ride shows interesting (and predictable) patterns in things like plasma glucose levels and fat oxidation. Most tellingly, here’s the data on their subjective rating of perceived exertion (RPE):
As you’d expect, everyone starts around the same level. But pretty soon the lines split up. The fully breakfasted, fully refueled group (the black circles) consistently reports the lowest effort: the ride feels easier to them. The no-breakfast, placebo sports drink group (the white squares) has by far the highest effort: it feels hard. Interestingly, the two intermediate conditions: breakfast but no sports drink, or sports drink but no breakfast, sit almost on top of each other. For a 105-minute submaximal ride, there doesn’t seem to be much difference between taking your carbs before or during the ride—though neither option is as good as doing both.
We could parse these results in more depth, but it’s not very useful to do so. This is a small study, and the most detailed conclusions about the pros and cons of each protocol inevitably depend on the characteristics of the riders in the study (how well trained are they? what do they usually eat?) and the specific exercise protocol (what happens if you’re racing right from the start? or going twice as long?).
Instead, the reason I think the study is worth highlighting is for the simple big-picture conclusion: consuming carbohydrate during prolonged exercise makes a difference. This study has some notable strengths, like the use of appropriate blinding and placebos so that subjects didn’t know if they were getting the real drink; a race-based exercise test; a realistic pre-race fueling scenario; and a lack of industry funding. (The paper’s senior author confirmed to me that neither he nor any of his colleagues has any funding ties to the sports drink industry.)
This doesn’t mean commercial sports drinks are the only option. Maybe you’ll do just as well eating a banana or drinking flat Coke, a la Frank Shorter. It certainly doesn’t prove you need a hydrogel or some other form of engineered sports drink. And if you’re just out for a long training run or ride rather than racing, there’s no rule that says you have to worry about maximizing your performance by lugging along extra calories: there is something to be said for simplicity. But leaving aside all the caveats and acknowledging the imperfections of the existing evidence, if you’re out there for a few hours or more and want to truly maximize your performance, you should be fueling on the go.
My new book, Endure: Mind, Body, and the Curiously Elastic Limits of Human Performance, with a foreword by Malcolm Gladwell, is now available. For more, join me on Twitter and Facebook, and sign up for the Sweat Science email newsletter.