How does caffeine work?

Caffeine is believed to have several effects in the body. From a performance point of view the exact mechanism isn’t well understood. Several years ago some studies suggested a role of caffeine during endurance exercise in increasing the ability of the body to burn fat. Increasing the body’s ability to burn fat and conserve carbohydrate LINK HFLC pieces at the same exercise intensity is considered by many to be the holy grail for endurance athletes, allowing the carbohydrate to be tapped into at the end of a race and beat your competition.

But more recent research disputes the role of caffeine in boosting fat utilisation during a race. Instead scientific consensus has swung towards a role for caffeine in the way that the brain regulates pace during exercise. It’s not clear exactly how this works (the science of the brain and exercise performance has barely scratched the surface so far) but it does seem to lower an athlete’s perception of effort for the same amount of actual physical output.

The graph below is from a study conducted by the Australian Institute of Sport and RMIT University in Melbourne, where highly trained cyclists and triathletes completed four cycling tests. They took caffeine in a sports drink either one hour before (PreCaf), in six doses during (DurCaf), a placebo or as Coke 100 minutes through a two hour ride at 82.5% of peak power output.

The athlete’s Rating of Perceived Exertion (RPE) was measured throughout the two hours, and as you can see from the graph the group taking caffeine before the ride started and maintained a lower RPE throughout the ride compared to placebo, despite riding at the same power output.

Likewise the RPE in the DurCaf group reduced below that of placebo from the 80 minute mark onwards. Those that switched to Coke at the 100 minute mark unsurprisingly had the same RPE as placebo, as they didn’t take the caffeine until the last 20 minutes.

Figure 1: Effect of Precaf, Durcaf, or Coke treatments on ratings of perceived exertion during 2-h steady-state cycling at 70% of VO2peak. Values are means ± SE for 12 subjects. All differences are significant at P ? 0.05: * main effect time, all time points different from t= 10 min; a main effect treatment, Precaf  Precaf and Placebo;c Precaf and Durcaf J Appl Physiol 93:990-999, 2002.

Figure 1: Effect of Precaf, Durcaf, or Coke treatments on ratings of perceived exertion during 2-h steady-state cycling at 70% of VO2peak. Values are means ± SE for 12 subjects. All differences are significant at P ? 0.05: * main effect time, all time points different from t= 10 min; a main effect treatment, Precaf < Placebo; b Durcaf > Precaf and Placebo;c Precaf and Durcaf < Placebo;d Durcaf < Placebo and Coke. Source: J Appl Physiol 93:990-999, 2002.

What’s the effect on performance?

In the same AIS study the athletes followed the two hour fixed intensity ride with a time trial of a set amount of work, calculated to take around 30 minutes. This is when athletes were allowed to vary their pace, to measure performance.

This pattern of “steady state” riding followed by a time trial is typical of many performance studies, simulating the typical pattern of work in a cycling road race.

During the time trial, the riders taking caffeine completed the ride around one minute faster compared to when they took the placebo. There was no statistical differences in performance between taking the caffeine before or during (or as Coke towards the end of) the steady state. Overall the average benefit from caffeine consumption on performance was around 3-3.5%.

Figure 2: Results of the 7 kJ/kg Time Trial following 120 min of steady-state cycling at 82.5% of peak power output.  Placebo = Capsules not containing caffeine taken 1 hour prior to the ride. Coke = 2 X 5mLkg body weight of Coca Cola consumed from 100-120minutes of the steady state ride. Total caffeine = 1mg/kg. DurCaf = Capsules containing 1mg/kg caffeine taken every 20 minutes during the steady state ride (total of 6mg/kg). PreCaf = Capsules containing 6mg/kg caffeine taken 1 hour  prior to the steady state ride. Drawn from results in J Appl Physiol 93:990-999, 2002.

Figure 2: Results of the 7 kJ/kg Time Trial following 120 min of steady-state cycling at 82.5% of peak power output.
Placebo = Capsules not containing caffeine taken 1 hour prior to the ride.
Coke = 2 X 5mLkg body weight of Coca Cola consumed from 100-120minutes of the steady state ride. Total caffeine = 1mg/kg.
DurCaf = Capsules containing 1mg/kg caffeine taken every 20 minutes during the steady state ride (total of 6mg/kg).
PreCaf = Capsules containing 6mg/kg caffeine taken 1 hour prior to the steady state ride.
Drawn from results in J Appl Physiol 93:990-999, 2002.

A number of studies have followed on caffeine and performance. The vast majority show an average benefit from caffeine consumption of between 2-5% across a range of protocols and doses.

Who’s average though?

One of the things that you need to be aware of when interpreting study results is that the average tells you nothing about the responses of individuals within the study. In some cases everyone follows the same pattern and is close to the average, but in others the variation between individuals is massive.

And with caffeine there does seem to be a large variation in how people respond. Whilst the average benefit seems to be around 2-5% in most studies, there are some athletes that have shown no improvement or even performed worse when taking caffeine compared to placebo.

Figure 3: Individual TT performances after consumption of placebo, 1.5mg/kg  and 3mg/kg caffeine. Exercise performed - 120min steady state (65%PPO) cycling + TT (time to complete 7kJ/kg/body weight of work).  Green – benefit of 3mg/kg caffeine compared to placebo Orange – Likely minimal or no benefit from 3mg/kg caffeine compared to placebo Red – Caffeine detrimental to performance Source: Desbrow B et. al. The effects of caffeine on endurance cycling performance and exogenous carbohydrate oxidation: a dose-response study. Presentation from the European College of Sports Science Congress, 2008.

Figure 3: Individual TT performances after consumption of placebo, 1.5mg/kg and 3mg/kg caffeine. Exercise performed – 120min steady state (65%PPO) cycling + TT (time to complete 7kJ/kg/body weight of work).
Green – benefit of 3mg/kg caffeine compared to placebo
Orange – Likely minimal or no benefit from 3mg/kg caffeine compared to placebo
Red – Caffeine detrimental to performance
Source: Desbrow B et. al. The effects of caffeine on endurance cycling performance and exogenous carbohydrate oxidation: a dose-response study. Presentation from the European College of Sports Science Congress, 2008.

No-one knows exactly why the performance response to caffeine is so individual (it’s not related to whether you’re a regular coffee drinker or not, see below), but one reason is probably that the rate at which caffeine enters your bloodstream varies greatly between people, as shown in the graph below.

So whilst one athlete may benefit from taking caffeine one hour before a race, others may need to take it two hours before for it to work. Like almost everything in sports nutrition, it highlights the need to try caffeine in training before you load up on it for an important race.

Figure 4: Average (mean) and individual blood caffeine responses after consumption of 3mg/kg caffeine. Note that the time point 0 on the graph represents when the participants started exercise, one hour after ingesting the caffeine capsules. Source: Desbrow B et. al. The effects of caffeine on endurance cycling performance and exogenous carbohydrate oxidation: a dose-response study. Presentation from the European College of Sports Science Congress, 2008.

Figure 4: Average (mean) and individual blood caffeine responses after consumption of 3mg/kg caffeine. Note that the time point 0 on the graph represents when the participants started exercise, one hour after ingesting the caffeine capsules. Source: Desbrow B et. al. The effects of caffeine on endurance cycling performance and exogenous carbohydrate oxidation: a dose-response study. Presentation from the European College of Sports Science Congress, 2008.

It’s also important to note that some people experience side-effects from caffeine at the doses that have been shown to improve performance. These include increased heart rate, anxiety and reduced ability to perform precision type tasks, as well as difficulty sleeping after training or competition.

What’s the optimal dose of caffeine to improve performance?

Most of the earlier studies on caffeine and endurance performance used very high doses, up to 6mg per kilogram of body weight. To put that in perspective, for a 70kg rider that’s 420mg of caffeine or the equivalent of over 5 cans of Red Bull!

More recently however studies have shown the same performance benefits from half that dose, and 3mg per kg is now generally accepted as the highest dose needed to provide a benefit.

Here’s the amount of caffeine in a variety of foods and sports nutrition products:

  • Coca Cola – 1 X 375mL can = 49mg
  • Red Bull – 1 X 250mL can = 80mg
  • No Doz tablet – 1 X tablet = 100mg
  • PowerBar caffeinated gel – 1 X gel = 50mg
  • Torq caffeinated gel – 1 X gel = 89mg
  • Coffee – Varies (see below)
  • Caffeinated gum is available in some countries, but is not sold in Australia

The caffeine content of coffee is difficult to quantify, because it varies so much. In 2007 a team from Griffith University on the Gold Coast published a study where they purchased 97 short blacks from various coffee shops in the area. Their analysis found that the caffeine content varied from 25-214mg per cup, with an average of 106mg. The Nespresso capsules range from 65mg to 120mg depending on the variety (and therefore the type of coffee beans) used.

Do I need to stop drinking coffee in the lead up to a race to get the benefit of caffeine?

The same team at Griffith University conducted a study in 2011 where regular coffee drinkers were asked to either abstain from caffeine for four days before a cycling time trial, or continue to consume 3mg/kg each day from capsules.

They were then given 3mg/kg on the morning of the time trial. The results in both groups were identical – a benefit of 3-3.6% from taking caffeine before the time trial, with no difference between those who abstained from caffeine for four days compared to those who didn’t.

Is caffeine banned in sport?

No. In 2004 the World Anti-Doping Authority removed any restrictions on caffeine, and it no longer appears on the prohibited list.

Summary

In endurance sports, low to moderate doses of caffeine (up to 3mg/kg) appear to improve performance by an average of around 2-5%. Exactly how caffeine works is still not fully understood, but it’s believed to play a role in the way the brain regulates fatigue and pacing during exercise. A variety of sports foods and supplements contain caffeine and the benefits can be obtained by taking it before or during a training session or race.

Be aware though that not everyone benefits from caffeine and some will experience side effects even at these moderate doses. Like all sports nutrition strategies make sure you try caffeine in training before using it in a race situation.