2.1. Zwiększenie wydolności fizycznej podczas krótkotrwałych ćwiczeń o wysokiej intensywności (ID 737, 1486, 1489)

The claimed effects are “physical performance (short term and endurance activities)” and “endurance during short term high intensity exercise”. The Panel assumes that the target population is active individuals in the general population.
In the context of the proposed wordings, the Panel assumes that the claimed effects refer to increase in physical performance during short-term high-intensity (>80 % maximum O2 consumption) exercise. Performance relates to the ability of completing a certain task (e.g. running a certain distance) as fast as possible.
The Panel considers that an increase in physical performance during short-term high-intensity exercise is a beneficial physiological effect.

2.2. Wzrost wydolności fizycznej (ID 737, 1486)

The claimed effect is “physical performance (short term and endurance activities)”. The Panel assumes that the target population is adults performing endurance exercise.
In the context of the proposed wordings, the Panel assumes that the claimed effect refers to increase in endurance performance (i.e. during longer-term exercise generally at intensity <80 % maximum O2 consumption). Performance relates to the ability of completing a certain task (e.g. running a certain distance) as fast as possible.
The Panel considers that an increase in endurance performance is a beneficial physiological effect.

2.3. Wzrost wytrzymałości (ID 1488)

The claimed effect is “supports exercise performance (reduction in perceived exertion, improve time to exhaustion and exercise capacity)”. The Panel assumes that the target population is adults performing endurance exercise.
In the context of the proposed wordings, the Panel assumes that the claimed effect refers to an increase in endurance capacity. Endurance capacity refers to the exercise time to self-reported fatigue when exercising at a constant workload or speed, generally at intensity <80 % maximum O2 consumption.
The Panel considers that an increase in endurance capacity is a beneficial physiological effect.

2.4. Zmniejszenie stopnia odczuwania wysiłku podczas ćwiczeń (ID 1488, 1490)

The claimed effects are “supports exercise performance (reduction in perceived exertion, improve time to exhaustion and exercise capacity)” and “reduces perception of effort”. The Panel assumes that the target population is adults performing endurance exercise.
In the context of the proposed wordings and the clarifications provided by Member States, the Panel assumes that the claimed effects refer to reduction in the rated perceived exertion/effort during exercise.
The Panel considers that a reduction in the rated perceived exertion/effort during exercise is a beneficial physiological effect.

3.1. Zwiększenie wydolności fizycznej podczas krótkotrwałych ćwiczeń o wysokiej intensywności (ID 737, 1486, 1489)

In the meta-analysis by Doherty and Smith (2004), the evaluation of a potential ergogenic effect of caffeine on short-term, high-intensity exercise (ST) included 12 studies evaluating 26 outcomes. Of these, 22 outcome measures addressed non-Tlim protocols, and therefore exercise performance (Anderson et al., 2000; Anselme et al., 1992; Bruce et al., 2000; Collomp et al., 2002; Doherty et al., 2004; Greer et al., 1998; Vanakoski et al., 1998; Williams et al., 1988). No significant effect of caffeine consumption on exercise performance was observed (ES=0.00±0.33, 95 % CI -0.02 to 0.02). In these studies, subjects were low to moderate caffeine users (in most studies usual caffeine intake was not reported), caffeine was withdrawn 24-168 hours before testing (24-72 h in most studies), and the caffeine dose ranged from 3.5 to 9 mg/kg body weight (5 to 7 mg/kg body weight in most studies) and was administered between 30 and 120 min (mostly between 60 and 120 min) before exercise testing.
Three individual studies which investigated the effect of caffeine consumption on physical performance during short-term exercise tests were provided in the consolidated list (Collomp et al., 1992; Paton et al., 2001; Wiles et al., 2006), one of which had been already considered in the meta-analysis above (Collomp et al., 1992).
In the study by Wiles et al. (2006), eight trained cyclists cycled 1 km in the laboratory after consumption of 5 mg of caffeine per kg body weight, or the same amount of placebo, in a control situation (no intervention) following a randomised, double-blind (for caffeine and placebo) cross-over design. Performance time was significantly reduced (by 3.1 %) after consumption of caffeine compared to the placebo and control tests. This decrease was accompanied by significant increases in mean speed, mean power, and peak power output after caffeine consumption compared to placebo and control. On the other hand, in the study by Paton et al. (2001), physical performance during a repeated sprint test comprising 20-metre sprints repeated 10 times during 100 seconds was not affected in 16 male team-sport athletes, 60 minutes after ingesting caffeine (6 mg/kg body weight) compared to placebo.
In weighing the evidence, the Panel took into account that the evidence for an effect of caffeine consumption on physical performance during short-term high-intensity exercise is inconsistent, and
that one meta-analysis of RCTs including 8 studies and 22 outcome measures did not show an effect of caffeine consumption on physical performance during short-term high-intensity exercise.
The Panel concludes that a cause and effect relationship has not been established between the consumption of caffeine and an increase in physical performance during short-term high-intensity exercise.

3.2. Wzrost wydolności fizycznej (ID 737, 1486)

In the meta-analysis by Doherty and Smith (2004), the evaluation of a potential ergogenic effect of caffeine on endurance exercise included 23 studies evaluating 39 outcomes. Of these, five studies (with five outcome measures) addressed non-Tlim protocols, and therefore endurance performance (Bell and McLellan, 2002; Cole et al., 1996; Conway et al., 2003; Cox et al., 2002; MacIntosh and Wright, 1995). A statistically significant effect of caffeine consumption on endurance performance was observed (ES=0.28±0.160, 95 % CI 0.08 to 0.47). In these studies, subjects were low to moderate caffeine users, caffeine was withdrawn 24-48 hours before testing, and the caffeine dose ranged from 4 to 6 mg/kg body weight and was administered between 60 and 150 min before the exercise testing. The Panel notes that the meta-analysis did not address whether the observed positive effects on endurance performance following caffeine intake were due to an ergogenic effect of caffeine intake per se, or to a detrimental effect of previous caffeine withdrawal.
A number of individual, randomised, placebo-controlled intervention studies which investigated the effects of caffeine on endurance performance were provided in the consolidated list (Berglund and Hemmingsson, 1982; Bridge and Jones, 2006; Cox et al., 2002; Kovacs et al., 1998; MacIntosh and Wright, 1995). The studies by McIntosh and Wright (1995) and Cox et al. (2002) were included in the meta-analysis by Doherty and Smith (2004) and are not discussed further.
Berglund and Hemmingsson (1982) found that performance in well trained cross-country skiers during a 21 km time trial was enhanced at sea level and at high altitude after caffeine consumption (6 mg/kg body weight 60 min prior to testing) compared with placebo. Subjects were habitual caffeine consumers and caffeine was withdrawn 12 hours before testing. A 24-sec reduction (-1.2 %) in 8-km running time was found by Bridge et al. (2006) after ingestion of caffeine (3 mg/kg body weight 60 min prior to the race) compared with placebo in 8 trained male distance runners. Subjects were habitual caffeine consumers and caffeine was withdrawn 48 hours before testing. In a study by Kovacs et al. (1998) 15 well-trained men (triathletes and cyclists) were given 1.2 mg, 1.8 mg or 2.56 mg caffeine per kg body weight during a 20-minute cycling warm-up protocol, and an additional 0.9 mg, 1.35 mg or 5.1 mg caffeine/kg body weight during a subsequent one-hour time trial. All doses led to significant improvements in performance compared with placebo.
In weighing the evidence, the Panel took into account that most of the human intervention studies provided (including a meta-analysis of five RCTs and three individual RCTs) showed an effect of caffeine consumption on endurance performance at doses of at least 3 mg/kg body weight administered at least one hour prior to exercise, and after at least one day of caffeine withdrawal in habitual caffeine consumers.
The Panel concludes that a cause and effect relationship has been established between the consumption of caffeine and an increase in endurance performance.

3.3. Wzrost wytrzymałości (ID 1488)

In the meta-analysis by Doherty and Smith (2004), the evaluation of a potential ergogenic effect of caffeine on endurance exercise included 23 studies evaluating 39 outcomes. Of these, 34 outcome measures addressed Tlim protocols, and therefore endurance capacity (Alves et al., 1995; Bell et al.,
1998; Bell and McLellan, 2002, 2003; Butts and Crowell, 1985; Cha et al., 2001; Cole et al., 1996; Conway et al., 2003; Denadai and Denadai, 1998; French et al., 1991; Fulco et al., 1994; Graham and Spriet, 1991, 1995; Graham et al., 1998; Greer et al., 2000; Lindinger et al., 1993; Pasman et al., 1995; Trice and Haymes, 1995; Van Soeren and Graham, 1998). A statistically significant effect of caffeine consumption on endurance capacity was observed (ES=0.68±0.42, 95 % CI 0.53 to 0.84). In these studies, subjects were low to moderate caffeine users, caffeine was withdrawn 12-96 hours before testing, and the caffeine dose ranged from 3 to 13 mg/kg (mostly between 5 and 6 mg/kg) body weight and was administered between 60 and 360 min (mostly up to 120 min) before exercise testing. The Panel notes that the meta-analysis did not address whether the observed positive effects on endurance performance following caffeine intake were due to an ergogenic effect of caffeine intake per se, or to a detrimental effect of previous caffeine withdrawal.
A number of individual intervention studies which investigated the effects of caffeine on endurance capacity have been provided with the consolidated list, some of which were already included in the meta-analysis by Doherty and Smith (2004) and which will not be discussed further (Bell and McLellan, 2002, 2003; Denadai and Denadai, 1998; French et al., 1991; Graham and Spriet, 1991, 1995; Pasman et al., 1995; Trice and Haymes, 1995; Van Soeren and Graham, 1998). Two additional studies assessed physical capacity during short-term high intensity exercise, and are therefore not pertinent to the claim (Flinn et al., 1990; Greer et al., 2000). Another study addressed the effects of caffeine in combination with glucose, which did not allow any conclusions to be drawn on the effects of caffeine alone (Sasaki et al., 1987).
Three human intervention studies on the effects of caffeine intake on endurance performance were not included in the meta-analysis (Costill et al., 1978; Jackman et al., 1996; Spriet et al., 1992).
In the study by Spriet et al. (1992), eight recreational cyclists (7 males) cycled to exhaustion at 80 % of VO2max on two different days after consuming either caffeine or placebo in a single-blind manner. All subjects consumed placebo in the first test and caffeine in the second. The Panel notes the uncontrolled nature of the study and considers that no conclusions can be drawn from this study for the scientific substantiation of the claim.
In a double-blind, cross-over, placebo-controlled intervention by Costill et al. (1978), nine competitive cyclists (seven males) consumed either decaffeinated coffee (5 g in 200 mL of hot water) or caffeinated coffee (same amount plus 330 mg caffeine) 60 minutes prior to cycling on a bicycle ergometer at 80 % of VO2max. Total exercise time to exhaustion was significantly increased after
consumption of caffeinated coffee (19.5 % SEM 2.3; p<0.05). Also in a double-blind, cross-over, placebo-controlled intervention by Jackman et al. (1996), using an intense intermittent exercise protocol (2 min of cycling at an intensity corresponding to maximal VO2, interrupted by 6 min rest periods), consumption of caffeine (6 mg/kg body weight) increased time to exhaustion significantly compared to placebo (dextrose) in 14 young adults who were active recreational or university of endurance activity athletes (11 males).
A plausible mechanism by which caffeine could exert the claimed effect is through the reduction of perceived exertion (RPE) during exercise (Doherty and Smith, 2005) (see section 3.4).
In weighing the evidence, the Panel took into account that most of the human intervention studies provided (including a meta-analysis of 23 RCTs evaluating 39 outcomes and two individual RCTs) showed an effect of caffeine consumption on endurance capacity at doses of at least 3 mg/kg body weight administered at least one hour prior to exercise, and after at least 12 hours of caffeine withdrawal in habitual caffeine consumers, and that evidence on a plausible mechanism by which caffeine could exert the claimed effect has been provided.
The Panel concludes that a cause and effect relationship has been established between the consumption of caffeine and an increase in endurance capacity.

3.4. Zmniejszenie stopnia odczuwania wysiłku podczas ćwiczeń (ID 1488, 1490)

In the meta-analysis by Doherty and Smith (2005) based on 22 laboratory-based, double-blind, fully randomised (and mostly cross-over), placebo-controlled intervention studies, the effects of caffeine ingestion on ratings of perceived exertion (RPE) during exercise were examined, and it was found that caffeine compared to placebo significantly reduced RPE during exercise (in 20 out of the 22 studies) by 5.6 % (95 % CI -4.5 to -6.7), and that RPE could account for 29 % of the variance in the improved exercise performance (based on 16 studies where changes in exercise performance were tested). This analysis comprised studies from 1975 to 2004 representing over 200 subjects (74 % men). The typical subject can be characterised as being 20 to 35 years of age, physically active individuals to extremely well trained elite athletes, and both habitual caffeine users and non-users (half of the studies do not give information on coffee use). The protocols varied, including work intensities from 50 % to 125 % (mean=80 %) of VO2max. The caffeine doses ranged from 4 to 10 mg per kg (median 6 mg/kg) body weight and were typically given one hour before the start of the exercise test, and the caffeine abstinence of the subjects varied from 12 to 240 hours (median = 24 hours).
In a double-blind, cross-over, placebo-controlled intervention study published after the meta-analysis by Doherty and Smith (2005), nine competitive male rugby players ingested either caffeine (6 mg/kg body weight) or placebo (dextrose) 70 min before performing a rugby test consisting of seven circuits in each of two 40-min halves with a 10-min half-time rest (Stuart et al., 2005). The development of fatigue during the test was significantly reduced after caffeine consumption compared to placebo.
In weighing the evidence, the Panel took into account that most of the human intervention studies provided (including a meta-analysis of 22 RCTs and one individual RCT) showed an effect of caffeine consumption on RPE during exercise at doses of at least 4 mg/kg body weight administered at least one hour prior to exercise, and after at least 12 hours of caffeine withdrawal in habitual caffeine consumers.
The Panel concludes that a cause and effect relationship has been established between the consumption of caffeine and a reduction in the rated perceived exertion/effort during exercise.

4.1. Wzrost wydolności fizycznej (ID 737, 1486)

The Panel considers that the following wording reflects the scientific evidence: “Caffeine contributes to an increase in endurance performance”.

4.2. Wzrost wytrzymałości (ID 1488)

The Panel considers that the following wording reflects the scientific evidence: “Caffeine contributes to an increase in endurance capacity”.

4.3. Zmniejszenie stopnia odczuwania wysiłku podczas ćwiczeń (ID 1488, 1490)

The Panel considers that the following wording reflects the scientific evidence: “Caffeine contributes to a reduction in the rated perceived exertion/effort during exercise”.

5.1. Wzrost wydolności fizycznej (ID 737, 1486)

The Panel considers that in order to obtain the claimed effect, caffeine should be consumed at doses of 3 mg/kg body weight one hour prior to exercise. The target population is adults performing endurance exercise.

5.2. Wzrost wytrzymałości (ID 1488)

The Panel considers that in order to obtain the claimed effect, caffeine should be consumed at doses of 3 mg/kg body weight one hour prior to exercise. The target population is adults performing endurance exercise.

5.3. Zmniejszenie stopnia odczuwania wysiłku podczas ćwiczeń (ID 1488, 1490)

The Panel considers that in order to obtain the claimed effect, caffeine should be consumed at doses of 4 mg/kg body weight one hour prior to exercise. The target population is adults performing endurance exercise.