Scientific Opinion on the safety of caffeine

In the paper

“About 6 % of the adult population may exceed 200 mg of caffeine in a single session of”

In the paper

“The proportion of survey days in which the level was exceeded ranged from about 7 to 12 % in adolescents, from 6 to 15 % in children and from 7 to 37 % in toddlers in the aforementioned countries.”

In the paper

“There are no data to characterise the risk of single doses of caffeine consumed by lactating women, and data on habitual caffeine consumption in this population subgroup are scarce.”

In the paper

“As only limited studies are available on the longer-term effects of caffeine on anxiety and behaviour in children and adolescents, there is substantial uncertainty regarding longerterm effects of habitual caffeine consumption in this age group.”

In the paper

“The proportion of subjects exceeding that level of intake in the above-mentioned countries was about 5 to 10 % for adolescents, 6 to 13 % for children and about 6 % for toddlers.”

In the paper

“The question of whether or not p-synephrine modifies the acute cardiovascular effects of single doses of caffeine has not been adequately investigated in humans, particularly if consumed shortly before intense physical exercise, and therefore no conclusions could be drawn.”

• Also stated in: § 5.1.2. Single doses of caffeine in combination with other common constituents of "energy drinks", alcohol and p-synephrine · plain

  “The Panel also considers that the question of whether or not p-synephrine modifies the acute cardiovascular effects of single doses of caffeine has not been adequately investigated in humans, particularly if consumed shortly before intense physical exercise, and therefore no conclusions could be drawn.”

In the paper

“Chocolate beverages were important contributors to total caffeine intakes in children and toddlers in most countries, and the use of a conservative caffeine value for this food category may have led to an overestimation of caffeine intakes in these age groups.”

In the paper

“This conclusion is based on prospective cohort studies showing a dose-dependent positive association between caffeine intakes during pregnancy and the risk of adverse birth weight-related outcomes (i.e.fetal growth retardation, small for gestational age) in the offspring.”

In the paper

“The estimated 95 th percentile of caffeine intake from foods and beverages on a single day exceeded 3 mg/kg bw in 6 out of 16 countries for adolescents (10 to < 18 years), in 9 out of 16 countries for children (3 to < 10 years) and in 3 out of 10 countries for toddlers (12 to < 36 months).”

In the paper

“This information is not available for other sources of caffeine.”

In the paper

“In these countries, the estimated proportion of the adult population exceeding daily intakes of 400 mg ranged from about 6 % to almost one-third (33 %), and coffee was the main source of caffeine.”

In the paper

“The question of whether or not p-synephrine modifies the acute cardiovascular effects of single doses of caffeine or the long-term effects of caffeine on the cardiovascular system has not been adequately investigated in humans and therefore no conclusions could be drawn.”

In the paper

“Sufficient (n ≥ 60) days to obtain statistically robust 95 th percentiles for toddlers regarding caffeine intakes from different sources on consumption days were available only for chocolate and tea, for which the highest values across Member States were estimated to be 5.3 and 9.6 mg/kg bw per day of caffeine, respectively (Appendix C).”

In the paper

“The mean and 95 th percentile of caffeine intakes in a”

What the sources say

• Not checked yetcites Zucconi et al., 2013

In the paper

“The mean and the 95 th percentile of the daily caffeine intake from all sources were 31 mg and 97 mg per day, respectively (Appendix B).”

In the paper

“The highest contribution to total caffeine intakes from”

In the paper

“The highest 95 th percentiles of caffeine intake on a single day in absolute values and on a per kg bw basis were 240 mg and 4.3 mg/kg bw, respectively (Appendices C and D).”

In the paper

“Means and 95 th percentiles ranged from 18 to 70 mg and from 61 to 212 mg, respectively, among countries (Table 2, Appendix B).”

In the paper

“On a per kg bw basis, the highest 95 th percentile of caffeine intake was from coffee (7.1 mg), followed by chocolate (5.4 mg),”

In the paper

“In this survey, 51, 25, 11, 6 and 7 % of”

What the sources say

• Not checked yetcites Zucconi et al., 2013

In the paper

“In most countries,”

In the paper

“The mean and 95 th percentile of caffeine intakes in a”

What the sources say

• Not checked yetcites Zucconi et al., 2013

In the paper

“The highest 95 th percentiles of caffeine intake from”

In the paper

“Means and 95 th percentiles ranged from 37 to 319 mg and from 109 to 742 mg, respectively, among countries (Table 2, Appendix B).”

In the paper

“In this survey, 52, 29, 11, 5 and 3 % of”

What the sources say

• Not checked yetcites Zucconi et al., 2013

In the paper

“Tea or chocolate was the main caffeine source in all surveys except for Belgium, where cola drinks contributed the most to total caffeine intake (58 %; Appendix E).”

In the paper

“For toddlers, 10 surveys were available.”

In the paper

“The highest 95 th percentiles of caffeine intake from”

In the paper

“Chocolate was the main contributor to caffeine intake in six surveys, coffee was the main contributor in four surveys, cola beverages were the main contributors in three surveys and tea was the main contributor in two surveys.”

In the paper

“Considering caffeine intake in consumers of the different caffeine-containing food groups, coffee consumers had the highest 95 th percentile of caffeine consumption per day (up to 776 mg of caffeine from coffee), followed by tea drinkers (up to 264 mg of caffeine from tea),”

In the paper

“When considering only days with consumption of the different food categories, the 95 th percentile of caffeine intake from coffee provided the highest estimate (15 mg/kg bw per day), followed by chocolate (7.7 mg/kg bw per day) (Appendix C).”

In the paper

“The mean and the 95 th percentile of the daily caffeine intake from all sources were 109 mg and 206 mg per day, respectively (Appendix B).”

In the paper

“In only the first two was a specific code for”

In the paper

“Daily intake estimates for the elderly and very elderly are of a similar magnitude, with a tendency for lower 95 th percentiles.”

In the paper

“In these surveys, the prevalence of”

In the paper

“The prevalence of”

In the paper

“According to this survey, 47, 26, 13, 9 and 5 % of this population declared that they consume one, two, three, four and five or more cans of”

In the paper

“The EFSA”

In the paper

“The highest prevalences of”

In the paper

“In this survey, 48, 25, 13, 7 and 7 % of this population declared that they consume one, two, three, four and five or more cans of”

In the paper

“Data are particularly scarce for pregnant and lactating women, and absent regarding the consumption of caffeine-containing supplements.”

In the paper

“Among the individual studies, a positive association between habitual caffeinated coffee consumption and CVD risk has been reported only for CHD (but not for stroke, AF or heart failure) in 8 out of the 56 studies reviewed, only 6 of which had any type of adjustment for confounding variables.”

In the paper

“Of these, only two reported an increased risk of CHD associated with habitual coffee consumption of four or less cups per day (one for three or more cups per day, one for three to four cups per day), corresponding to about 400 mg per day of caffeine, which may be an underestimation of total caffeine intake”

In the paper

“For 6 out of 14 Member States the estimates of the 95 th percentile of daily caffeine intake from all sources exceeds 3 mg/kg bw.”

In the paper

“For this dose, only 1 out of 42 pre-term infants showed tachycardia and jittering, and only tachycardia (with no jittering) was observed in 8 out of 45 infants at doses 100-fold higher (30 mg/kg bw).”

What the sources say

• Not checked yetcites Steer et al., 2003 source

In the paper

“The mean and the 95 th percentile of the daily caffeine intakes from foods and beverages in the only survey available were 109 mg and 206 mg per day, respectively.”

In the paper

“The Panel notes that the surveys considered did not report on the consumption of caffeine from food supplements, and that there are no data on the frequency and quantity of consumption of caffeine from this source in Europe in different subgroups of the population.”

In the paper

“The mean and the 95 th percentile of the daily caffeine intakes from food and beverages in the only survey available were 31 mg and 97 mg per day, respectively, which are well below 200 mg.”

In the paper

“As only limited studies are available on the longer-term effects of caffeine on anxiety and behaviour in children and adolescents, there is substantial uncertainty regarding longer-term effects of habitual caffeine consumption in this age group.”

In the paper

“The Panel considers that the question of whether or not p-synephrine modifies the long-term effects of caffeine on the cardiovascular system has not been adequately investigated in humans, and therefore no conclusions could be drawn.”

In the paper

“The scientific assessment is based on human intervention and observational studies with adequate control for confounding variables, which have been conducted in healthy subjects at recruitment.”

In the paper

“Estimates of the proportion of single days in which caffeine intake exceeds 3 mg/kg bw among all survey days using the EFSA Comprehensive Database were used as a conservative approximation.”

In the paper

“Overall, at EU level to date, caffeine has only been assessed in the context of energy drinks but the safety of overall caffeine intake, from all sources, and acceptable use levels has not yet been assessed.”

In the paper

“Whenever available, human intervention studies and prospective cohort studies have been preferred over case-control and cross-sectional studies because of the lower risk of reverse causality and recall bias.”

In the paper

“Case reports of adverse events have not been considered for the scientific assessment.”

In the paper

“However, caffeine is not an essential nutrient, and thus ULs cannot be derived.”

In the paper

“Other common constituents of”

In the paper

“This approach is rather conservative in relation to the effects of caffeine on the cardiovascular system, but the limited studies available regarding the longer-term effects of caffeine on anxiety and behaviour in children and adolescents support the proposed caffeine intake level of no safety concern.”

In the paper

“In addition, advice on the safe/recommended timing of caffeine consumption prior to physical activity cannot be provided by the derivation of an UL.”

In the paper

“As caffeine clearance in children and adolescents is at least that of adults, the same levels of no safety concern derived for adults (i.e.5.7 mg/kg bw) may also apply to children, unless there are data showing a higher sensitivity to the effects of caffeine in this age group (i.e.difference in pharmacodynamics).”

In the paper

“The 17 surveys from 11 Member States added in 2014 used the FoodEx2 code classification, which allows accurate reporting of “energy drink” consumption (Appendix A).”

In the paper

“(a): The 95 th percentile estimates obtained from dietary surveys and age classes with fewer than 60 subjects may not be statistically robust (EFSA, 2011b) and consequently were not considered in this table (“-”).”

What the sources say

• Not checked yetcites EFSA, 2011b

In the paper

“Owing to the differences in the methods used for data collection, direct country-to-country comparisons must be taken with caution.”

In the paper

“The Panel notes, however, that a “night out”, “study session” or “sport session” could last more than two hours, and it is not possible from the report to precisely define the time frame covered by a “single session”.”

In the paper

“The 95 th percentiles of caffeine intake on a single day have been calculated to gather information on days of particularly high caffeine consumption, but these do not provide information about the proportion of subjects with high caffeine consumption days.”

In the paper

“Systematic reviews and meta-analysis have been used to summarise the scientific evidence whenever available.”

In the paper

“It is also possible that some of them are decaffeinated.”

In the paper

“Therefore, the Panel considered that this study could be used to calculate caffeine intakes from “energy drinks” in a “single session”, either alone or in combination with physical exercise, but not to calculate total caffeine intakes from all sources or from sources other than “energy drinks”.”

In the paper

“Owing to personal choices, individual consumers might therefore have coffee beverages with lower or higher caffeine levels than those used in the present assessment.”

In the paper

“However, it is not possible to exclude that some of the brands of cocoa-beverage preparation powder consumed in particular by children contain lower levels of caffeine than those used in the present assessment.”

In the paper

“The EFSA Comprehensive European Food Consumption Database was used to calculate caffeine intake from food and beverages.It contains data from 39 surveys in 22 different European countries for a total of 66 531 participants.”

In the paper

“The Panel notes that this study provides useful information about the prevalence of “energy drink” consumption in Europe, the number of “energy drinks” (and the amount of their constituents, including caffeine) consumed in a “single session” and the prevalence of “energy drink” consumption in combination with physical activity.”

In the paper

“(iii) the study contains information about the frequency of consumption of “energy drinks” in combination with alcohol, but not on the amount of alcohol consumed in combination with “energy drinks”.”

In the paper

“The EFSA report on energy drinks was used to calculate caffeine intakes from”

In the paper

“Although the time covered by the term”

In the paper

“In humans, caffeine is rapidly (t max 30-120 minutes) and completely absorbed after oral intake (Blanchard and Sawers, 1983).”

What the sources say

• Not checked yetcites Blanchard and Sawers, 1983 source

In the paper

“The Panel notes the high total caffeine intakes used in these studies (500-750 mg per day), that lower repeated doses of caffeine (< 250 mg) have not been tested and that the time between doses (four hours), about one half-life, is likely to induce an increase in plasma caffeine concentrations throughout the day.”

In the paper

“which induced a mean increase in SBP of about 3-8 mmHg and in DBP of about 4-6 mmHg, with high inter-individual variability.”

In the paper

“The EFSA Comprehensive Database is useful to gather data on daily caffeine intakes from all sources from unselected populations and population subgroups, as well as on the contribution of different food groups to total caffeine intakes.”

In the paper

“Compared with placebo, caffeine significantly raised average SBP through the entire day by about 4 mmHg and DBP by about 3 mmHg, whereas heart rate (HR) decreased by 2 bpm, with no significant interaction between treatment and location (at work, at home).”

What the sources say

• Not checked yetcites Lane et al., 2002

In the paper

“Souza et al. (2014) reported a significant decrease in SBP and MABP and a significant increase in HR after exercise with caffeine and placebo.DPB and HR (but not SPB) were significantly higher at rest with caffeine than with placebo, as was peripheral vascular resistance (SBP, DBP, HR, stroke volume or cardiac output) in the nine hours post exercise.”

What the sources say

• Not checked yetcites Souza et al., 2014 source

In the paper

“A significant increase in SBP (by about 7-8 mmHg) was reported following consumption of the sports drink plus caffeine in both preliminary tests, whereas consumption of the other beverages did not increase SBP significantly.No significant changes in DBP were noted.”

What the sources say

• Not checked yetcites Bischoff, 2013

In the paper

“All studies available in healthy individuals (Mahmud and Feely, 2001;Vlachopoulos et al., 2003Vlachopoulos et al., , 2006;;Hartley et al., 2004;Karatzis et al., 2005;Swampillai et al., 2006) reported a significant adverse effect of caffeine (in supplements, coffee or tea) at doses of 100-250 mg on one or more measures of arterial compliance”

What the sources say

• Not checked yetcites Mahmud and Feely, 2001 source
• Not checked yetcites Vlachopoulos et al., 2003 source
• Not checked yetcites Hartley et al., 2004 source

In the paper

“Paraxanthine, theophylline and theobromine are further metabolised and then excreted in the urine.”

In the paper

“Significant increases in SBP and DBP were also reported for a combination of 5.5 mg p-synephrine and 5.7 mg octopamine and caffeine (239.2 mg), and a significant increase in DBP, but not in SPB, was observed with 21 mg p-synephrine and 304 mg caffeine.”

What the sources say

• Not checked yetcites Stohs et al., 2012 source

In the paper

“The Panel considers that, although these studies have small sample sizes and are difficult to compare, they suggest an additive effect of caffeine and resistance training on BP during exercise, and that caffeine could attenuate the decrease in BP observed after resistance training.”

In the paper

“These authors also found a higher CYP1A2 activity in smokers and men than in non-smokers and women, respectively.”

What the sources say

• Not checked yetcites Rasmussen et al., 2002 source

In the paper

“The Panel notes that the prevalence of”

In the paper

“In the study by Astorino et al. (2007), SBP, HR and ratepressure product (but not DBP) significantly increased during exercise with caffeine and placebo and were significantly higher before and during exercise with caffeine than with placebo, but no time by treatment interaction was observed, suggesting an additive (but not synergistic) effect of caffeine and exercise on BP.”

What the sources say

• Not checked yetcites Astorino et al., 2007 source

In the paper

“In the study by Astorino et al. (2013), in which four resistance exercise protocols were tested, SBP significantly increased during all exercise protocols and significantly decreased post exercise with caffeine and placebo.SBP was significantly higher before and two hours after exercise with caffeine than with placebo, whereas HR and DBP were not different between treatments.”

What the sources say

• Not checked yetcites Astorino et al., 2013 source

In the paper

“The effect was observed at single doses of caffeine ranging from 80 to 300 mg, although most studies tested doses of about 200-300 mg”

In the paper

“Such differences do not allow direct between-country comparisons, and thus ranges of means and 95 th percentiles across surveys should be interpreted with caution.”

In the paper

“compared with decaffeinated coffee, caffeinated coffee has been reported to significantly increase SBP (130 mg caffeine) and DBP (80 and 130 mg caffeine), as well as to decrease endothelium-dependent flow-mediated dilatation (FMD) in habitual moderate coffee consumers after 12-24 hours' caffeine withdrawal”

What the sources say

• Not checked yetcites Papamichael et al., 2005 source
• Not checked yetcites Buscemi et al., 2010 source

In the paper

“These studies suggest that repeated doses of 250 mg caffeine taken four hours apart may induce a significant increase in daytime BP, that BP remains significantly elevated up to 9-12 hours after consumption of the last dose and that the effect, which depends on habitual caffeine consumption, is mostly observed after caffeine withdrawal.”

In the paper

“As indicated in the background provided by the European Commission and in Section 2.4 of this opinion, the health concerns expressed by national and international risk assessment bodies in relation to caffeine mostly refer to its effects on pregnancy outcomes, the cardiovascular system and the CNS.”

In the paper

“As in previous risk assessments by, for example, the SCF and EU Member State Committees, the present opinion considers primarily human data and addresses specific subgroups of the population, such as pregnant and lactating women, children and subjects performing physical exercise.”

In the paper

“Concerns were also raised with respect to caffeine in so-called”

In the paper

“Other primary metabolites are theophylline and theobromine.”

In the paper

“The homozygote AA genotype has been considered by some authors as a”

In the paper

“Caffeine significantly raised both SBP and DBP in all groups.However, the strongest response to caffeine was observed among diagnosed men (mean increase in SBP of 10 mmHg), followed by the stage 1 and high-normal groups and then by the normal and optimal groups, with a change 1.5 times greater in diagnosed hypertensive individuals than in the group with optimal BP.”

What the sources say

• Not checked yetcites Hartley et al., 2000 source

In the paper

“Caffeine significantly increased resting MBF (1.71 ± 0.41 mL/min per g tissue versus 2.22 ± 0.49 mL/min per g tissue; p < 0.001) and significantly decreased exercise-induced hyperaemic MBF (5.15 ± 0.79 mL/min per g tissue versus 3.98 ± 0.83 mL/min per g tissue; p < 0.005), leading to a decrease in MFR of 39 % (3.13 ± 0.60 to 1.87 ± 0.45; p < 0.005).”

What the sources say

• Not checked yetcites Namdar et al., 2006 source

In the paper

“There was no apparent increase in risk in the hour following consumption of caffeinated tea (RR = 0.9; 95 % CI = 0.4-2.0;p = 0.85) or cola (RR = 1.0; 95 % CI = 0.4-2.4;p = 0.95)”

What the sources say

• Not checked yetcites Mostofsky et al., 2010a, b

In the paper

“Compared with placebo, caffeine significantly increased core (rectal) temperature during exercise by up to about 0.5 C, whereas it had no significant effect on skin temperature, heart rate, loss of body mass or sweat rate.”

What the sources say

• Not checked yetcites Roelands et al., 2011 source

In the paper

“Although not formally tested, the Panel considers that changes in BP and MBF induced by repeated intakes of caffeine at doses and time intervals which would not exceed the maximum plasma concentrations achieved with a single dose of 200 mg of caffeine would also be of low clinical relevance for healthy individuals in the general population under normal environmental conditions.”

In the paper

“Similar results were obtained in two studies with a similar design conduced in healthy males, in which 5 mg/kg bw of caffeine were consumed daily for four consecutive days by low habitual caffeine consumers (< 100 mg/day) (Silva et al., 2013)”

What the sources say

• Not checked yetcites Silva et al., 2013 source

In the paper

“After caffeine consumption, sweating rate was significantly higher and the onset of sweating was significantly delayed during exercise.”

What the sources say

• Not checked yetcites Kim et al., 2011 source

In the paper

“patients with light/occasional intake of coffee (up to one cup per day; n = 66) had a RR of MI in the hour after taking coffee of 4.14 (95 % CI = 2.03-8.42),those with moderate consumption of coffee (two to three cups per day; n = 280) had a RR of 1.60 (95 % CI = 1.16-2.21),and heavy drinkers (four or more cups per day; n = 120) had a RR of 1.06 (95 % CI = 0.69-1.63;p = 0.006, test for interaction”

What the sources say

• Not checked yetcites Baylin et al., 2006 source

In the paper

“The effect may be more pronounced in subjects with high BP and after caffeine withdrawal.”

In the paper

“No significant differences in core temperature were observed between the caffeine and placebo conditions.”

What the sources say

• Not checked yetcites Ping et al., 2010

In the paper

“The Panel notes that caffeine antagonises the vasodilator effect of adenosine and other A 2A receptor agonists in the coronary arteries in a dose-dependent manner and this effect leads to a reduction of MBF and MFR during intense physical exercise, primarily in subjects with CAD, but also in healthy subjects to some degree.”

In the paper

“Resting MBF was not significantly affected by caffeine consumption in either group.Exercise-induced hyperaemic MBF and MFR significantly decreased 50 minutes after caffeine ingestion compared with baseline measurements without caffeine.MFR significantly decreased in healthy subjects by 14 % (p < 0.05) and in CAD patients by 18 % (p < 0.05) in remote segments and by 25 % (p < 0.01) in stenotic”

What the sources say

• Not checked yetcites Namdar et al., 2009 source

In the paper

“The risk of stroke onset was 2.3-fold higher (95 % CI = 1.4-4.0;p = 0.002) within one hour of stroke and RR was 1.6 (95 % CI = 1.0-2.5;p = 0.05) in the second hour after alcohol use compared with periods of non-use.The RR returned to baseline thereafter.”

What the sources say

• Not checked yetcites Mostofsky et al., 2010a, b

In the paper

“The estimated RR of dying within one hour after coffee consumption was 1.73 (95 % CI = 1.13-2.65)and was 3.00 (95 % CI = 1.61-5.68)within two hours after alcohol consumption.”

What the sources say

• Not checked yetcites Selb Semerl and Selb, 2004

In the paper

“The available data suggest that BP generally increases 30 minutes after caffeine consumption, reaches a peak after 60-90 minutes and returns to baseline after about two to four hours, which is consistent with the pharmacokinetics of caffeine.”

In the paper

“This was the only variable that significantly modified the risk; age, sex or physical activity, history of diabetes, hypercholesterolaemia, hypertension, smoking status or having at least three (n = 101) or less than three of these risk factors for disease did not significantly modify the risk.”

What the sources say

• Not checked yetcites Baylin et al., 2006 source

In the paper

“The MBF (mean ± standard error of the mean (SEM)) was not significantly different between caffeine and placebo at rest (1.13 ± 0.04 mL/min per g versus 1.06 ± 0.05 mL/min per g) or under stress (2.98 ± 0.14 mL/min per g versus 3.05 ± 0.14 mL/min per g).There was no difference in MFR between caffeine and placebo (2.75 ± 0.16 mL/min per g versus 2.97 ± 0.16 mL/min per g).The data show with one-sided”

What the sources say

• Not checked yetcites Gaemperli et al., 2008 source

In the paper

“Compared with placebo, single doses of 9 mg/kg bw caffeine consumed after four days of caffeine abstinence also increased core (rectal) body temperature (by 0.20-0.30C), but not skin temperature, in 10 healthy men”

What the sources say

• Not checked yetcites Cheuvront et al., 2009 source

In the paper

“Rectal temperature in the CAF + CE trials was significantly higher than in the placebo trials (by 0.19 to 0.29 C) after 60 minutes, but was significantly higher than in the CE trials only in the last 15 minutes”

What the sources say

• Not checked yetcites Millard-Stafford et al., 2007

In the paper

“a similar study using regadenoson stress-single-photon emission computed tomography (SPECT) MPI in patients with CAD showed that doses of caffeine of 200 mg and 400 mg significantly decreased the number of cardiac segments with reversible vascularisation defects, suggesting that caffeine at doses of 200 mg or greater could significantly counteract the vasodilator effect of regadenoson in the”

What the sources say

• Not checked yetcites Tejani et al., 2014 source

In the paper

“Single doses of caffeine of about 100-600 mg consumed prior to exercise, however, did not lead to a significant increase in urinary volume or a significantly different water retention during dehydration, and did not induce adverse water-electrolyte imbalances compared with water or placebo, particularly in habitual caffeine consumers (Armstrong, 2002).”

What the sources say

• Not checked yetcites Armstrong, 2002 source

In the paper

“The RR after coffee intake was not significantly increased when two or three hours were selected as the hazard period, suggesting that the risk dropped to basal conditions between one and two hours after drinking coffee.”

What the sources say

• Not checked yetcites Baylin et al., 2006 source

In the paper

“The Panel also notes that repeated doses of caffeine (250 mg) taken four hours apart also induce an increase in BP (of about 3-4 mmHg) which may last up to 9-12 hours, particularly after caffeine withdrawal.”

In the paper

“The Panel notes that the effect of higher doses of caffeine has not been tested.”

In the paper

“The association between ischaemic stroke in the hour after coffee consumption was apparent only among those consuming at most one cup per day but not for patients who consumed coffee more regularly (p for trend = 0.002).”

What the sources say

• Not checked yetcites Mostofsky et al., 2010a, b

In the paper

“Conversely, in another study using the same caffeine dose, mean body temperature was significantly higher (by 0.27 C) one hour after caffeine consumption than with placebo only at the beginning of the exercise (cycling for 30 minutes at 50 % of VO 2peak in a 40 °C and 25 % relative humidity environment), whereas no differences between caffeine and placebo were observed with respect to core, skin”

What the sources say

• Not checked yetcites Ely et al., 2011 source

In the paper

“Children appear to develop tolerance to the central effects of caffeine at high habitual intakes (> 300 mg per day) and show withdrawal symptoms.”

In the paper

“The overall pooled estimates of treatment effect associated with coffee drinking were 2.4 mmHg (95 % CI = 1.0-3.7 mmHg) for SBP and 1.2 mmHg for DBP (95 % CI = 0.4-2.1 mmHg).”

What the sources say

• Not checked yetcites Jee et al., 1999 source

In the paper

“No significant differences in self-rated anxiety or investigatorrated items of behaviour were found between placebo and caffeine at any dose.”

What the sources say

• Not checked yetcites Elkins et al., 1981 source
• Not checked yetcites Rapoport et al., 1981b

In the paper

“Caffeine significantly increased urinary output and natriuresis, whereas taurine had no effect on either of these outcomes and did not appear to modify the diuretic effects of caffeine when administered simultaneously.”

What the sources say

• Not checked yetcites Riesenhuber et al., 2006 source

In the paper

“Caffeine significantly increased SBP and DBP in studies with a majority of women (n = 8), but not in those with a majority of men (n = 17).”

What the sources say

• Not checked yetcites Noordzij et al., 2005 source

In the paper

“Scores for this side effect were significantly higher for the 3 mg/kg bw dose than for placebo, and for the 10 mg/kg bw dose than for the 3 mg/kg bw dose.”

What the sources say

• Not checked yetcites Elkins et al., 1981 source
• Not checked yetcites Rapoport et al., 1981b

In the paper

“Caffeine intake was not significantly associated with any of these measures of self-reported anxiety, whereas a significant linear association was reported between caffeine concentrations in saliva and the anxiety state item in the VAA-R scale only.”

What the sources say

• Not checked yetcites Bernstein et al., 1994 source

In the paper

“whereas higher intakes (10 mg/kg bw per day) may increase anxiety and adversely affect behaviour and sleep in habitual low caffeine consumers.”

In the paper

“The Panel notes that this metaanalysis shows a sustained BP-raising effect of continuous (≥ 7 days) caffeine consumption at doses of about 400 mg per day, but not at lower doses.”

What the sources say

• Not checked yetcites Noordzij et al., 2005 source

In the paper

“However, the Panel also notes the scarcity of studies available and the low number of children enrolled, as well as the heterogeneity of the studies regarding the testing conditions (e.g. with respect to the tools used to assess anxiety and behavioural changes, and to the caffeine status before testing), which hampers the comparability of the results.”

In the paper

“After excluding nine coffee trials with an open design, changes in SBP were 2.81 mmHg (95 % CI = 1.08-4.53mmHg) and in DBP were 1.17 mmHg (95 % CI = 0.54-1.81mmHg).HR did not change significantly.”

What the sources say

• Not checked yetcites Noordzij et al., 2005 source

In the paper

“Duration of run-in and coffee dose significantly modified the effect of coffee on BP.Studies with shorter run-ins and the highest doses of coffee showed the biggest effect on BP.”

What the sources say

• Not checked yetcites Jee et al., 1999 source

In the paper

“the type of intervention, sex and caffeine dose significantly affected changes in BP, whereas age, baseline BP, baseline caffeine intake and study duration (< 6 weeks or ≥ 6 weeks) did not.”

What the sources say

• Not checked yetcites Noordzij et al., 2005 source

In the paper

“The Panel considers that this route of administration is not appropriate to investigate the effects of oral caffeine consumption on anxiety.”

What the sources say

• Not checked yetcites Nickell and Uhde (1994)

In the paper

“Combining all caffeine and coffee studies, SBP significantly increased by 2.04 mmHg (95 % CI = 1.10-2.99mmHg) and DBP significantly increased by 0.73 mmHg (95 % CI = 0.14-1.31mmHg).”

What the sources say

• Not checked yetcites Noordzij et al., 2005 source

In the paper

“In children, similar single doses of caffeine on a weight basis (3 mg/kg bw) do not appear to induce anxiety or behavioural changes, although inter-individual variability in relation to habitual caffeine intakes has not been studied.”

In the paper

“Heterogeneity in the pooled SBP (I 2 = 72 %) and DBP (I 2 = 41 %) analysis was explored by considering type of coffee, sex and prestudy BP in subgroup analyses, but no significant interactions were found.”

What the sources say

• Not checked yetcites Steffen et al., 2012 source

In the paper

“Previous meta-analyses have consistently reported positive associations between habitual consumption of caffeine-containing foods and beverages (mostly coffee) and the risk of CVD from case-control studies, but not from prospective cohort studies (Greenland, 1993;Kawachi et al., 1994;Sofi et al., 2007).”

What the sources say

• Not checked yetcites Greenland, 1993 source
• Not checked yetcites Kawachi et al., 1994 source
• Not checked yetcites Sofi et al., 2007 source

In the paper

“Caffeine from any source at doses ≥ 410 mg per day significantly increased SBP (mean 2.98, 95 % CI = 2.15-3.80)and DBP (mean 1.96, 95 % CI = 1.19-2.73),whereas caffeine doses < 410 mg per day did not (change in SBP = mean 0.72, 95 % CI = -0.35 to 1.78; change in DBP = mean -0.52, 95 % CI = -1.62 to 0.57).”

What the sources say

• Not checked yetcites Noordzij et al., 2005 source

In the paper

“The Panel also notes that the above-mentioned tools to assess self-reported anxiety have been developed to discriminate between children with high and low anxiety levels, rather than to assess changes in anxiety.”

In the paper

“The Panel considers that results obtained in children with ADHD cannot be extrapolated to children in the general population and will not be further considered in this opinion”

In the paper

“Changes (mean and 95 % CI) in SBP and DBP for coffee (18 strata) were 1.22 mmHg (0.52-1.92 mmHg) and 0.49 mmHg (-0.06 to 1.04 mmHg), and for caffeine (seven strata) were 4.16 mmHg (2.13-6.20 mmHg) and 2.41 mmHg (0.98-3.84 mmHg).”

What the sources say

• Not checked yetcites Noordzij et al., 2005 source

In the paper

“The Panel notes that caffeine intakes were not significantly associated with anxiety, and that caffeine concentrations in saliva were generally not associated with self-reported measures of anxiety in this study.”

What the sources say

• Not checked yetcites Bernstein et al., 1994 source

In the paper

“The Panel also notes that the studies available at doses of ≤ 3 mg/kg bw have small sample sizes and are heterogeneous in design, and that doses between 3 and 10 mg/kg bw per day have not been investigated.”

In the paper

“The pooled weighted difference in mean change of SBP was -0.55 mmHg (95 % CI = -2.46 to 1.36 mmHg) and of DBP was -0.45 mmHg (95 % CI = -1.52 to 0.61 mmHg).”

What the sources say

• Not checked yetcites Steffen et al., 2012 source

In the paper

“In the context of this opinion, however, a reduction in the perceived exertion/effort during exercise can be considered a potential adverse health effect for the general population (not limited to adults performing endurance exercise willing to obtain such an effect) under the assumption that the perception of fatigue is a physiological mechanism leading to the spontaneous ending of physical”

In the paper

“Higher doses of alcohol have not been systematically investigated.”

In the paper

“the study by Palatini et al. (2007) reported an increased risk of developing sustained hypertension associated with caffeinated coffee consumption compared with no consumption, with no significant difference in risk among categories of coffee intake”

What the sources say

• Not checked yetcites Palatini et al., 2007

In the paper

“The Panel notes that, although this case-control study suggests that polymorphisms of the CYP1A2 gene may affect the risk of MI in relation to caffeine consumption, which may be further affected by polymorphisms of the serotonin receptor gene HTR2A, the results have not yet been replicated and the hypothesis has not been tested in prospective cohort studies.”

What the sources say

• Not checked yetcites Cornelis et al., 2006 source

In the paper

“Compared with the lowest category of coffee consumption (median zero cups per day), the RRs of CHD were 0.89 (95 % CI = 0.85-0.94)for the first category (median 1.5 cups per day), 0.90 (95 % CI = 0.84-0.97)for the second category (median 3.5 cups per day) and 0.93 (95 % CI = 0.84-1.02)for the third category of coffee consumption (median five cups per day).”

What the sources say

• Not checked yetcites Ding et al., 2014 source

In the paper

“Coffee consumption was inversely associated with the risk of CVD up to eight cups per day.No association was observed between coffee consumption and CHD risk at higher intake.”

What the sources say

• Not checked yetcites Ding et al., 2014 source

In the paper

“The cumulative RR for all cohort studies was 1.04 (95 % CI = 0.90-1.19)for the first category of coffee consumption (one to two cups per day), 1.05 (95 % CI = 0.90-1.22)for the second category (three to four cups per day) and 1.16 (95 % CI = 0.95-1.41)for the third category (four or more cups per day) compared with the reference category (none or less than one cup per day).”

What the sources say

• Not checked yetcites Sofi et al., 2007 source

In the paper

“In multivariate models including beverage type, rather than actual caffeine intake, significant inverse associations between intake of caffeinated coffee (p for trend = 0.02 and 0.03 in NHS I and NHS II, respectively) and the risk of hypertension was observed in both cohorts, but an association was not found with decaffeinated coffee (p for trend = 0.08 and 0.67 in NHS I and NHS II, respectively).”

What the sources say

• Not checked yetcites Winkelmayer et al., 2005 source

In the paper

“Changes in 24-hour ambulatory BP did not differ significantly among interventions with respect to baseline.”

What the sources say

• Not checked yetcites Hodgson et al., 1999 source

In the paper

“However, these two studies exclusively enrolled subjects with never-treated stage 1 hypertension at baseline (incidence of hypertension requiring medication of 50.7 % in the six-year follow-up) and do not provide information about the general (unselected) adult population.”

What the sources say

• Not checked yetcites Palatini et al., 2007
• Not checked yetcites Palatini et al., 2009 source

In the paper

“In total, 55 % of cases (n = 1 114) and 54 % of controls (n = 1 082) were carriers of the slow *1F allele, for which the multivariate-adjusted odds ratios (ORs) and 95 % CIs for non-fatal MI associated with the consumption of less than one, one, two to three and four or more cups of coffee per day were 1.00 (reference), 0.99 (0.69-1.44), 1.36 (1.01-1.83)and 1.64 (1.14-2.34),respectively.Cor”

What the sources say

• Not checked yetcites Cornelis et al., 2006 source

In the paper

“The pooled RRs of CHD for all studies combined were 0.96 (95 % CI = 0.87-1.06)for the moderate, 1.04 (95 % CI = 0.92-1.17)for the heavy and 1.07 (95 % CI = 0.87-1.32)for the very heavy categories of coffee consumption.”

What the sources say

• Not checked yetcites Wu et al., 2009

In the paper

“The Panel notes the low number of coffee abstainers included in the study and the low number of incident cases of hypertension in this group.”

What the sources say

• Not checked yetcites Uiterwaal et al., 2007 source

In the paper

“The risk of CHD (total, non-fatal MI and fatal CHD), coronary artery bypass grafting (CABG) and percutaneous transluminal coronary angioplasty (PTCA), stroke or total CVD did not increase significantly across categories of caffeinated or decaffeinated coffee consumption (one or less cup, two to three cups and four or more cups per day compared with none) or of total caffeine intake (quintiles: 0-”

What the sources say

• Not checked yetcites Grobbee et al., 1990 source

In the paper

“In the study by Uiterwaal et al. (2007), where the category of more than zero to three cups per day was taken as the reference, higher coffee intake was not associated with an increased risk of hypertension, whereas coffee abstainers showed a decreased risk compared with the reference category.”

What the sources say

• Not checked yetcites Uiterwaal et al., 2007 source

In the paper

“A significant association between caffeinated tea intake and incident hypertension was found in the cohort of younger women in NHS II (p for trend = 0.01), but not in NHS I (p for trend = 0.79).”

What the sources say

• Not checked yetcites Winkelmayer et al., 2005 source

In the paper

“The risk of CHD, stroke and total CVD did not increase significantly across categories of total caffeine intake (> 30, 30-100, 100-350 and ≥ 350 mg per day) or of caffeinated beverages consumed (< 0.5, 0.5 to < 2, 2 to < 4 and ≥ 4 servings per day).The risk of CVD and CHD mortality significantly decreased across categories of caffeine and caffeinated beverages consumption in subjects ≥ 65 years of”

What the sources say

• Not checked yetcites Greenberg et al., 2008 source

In the paper

“The Panel also notes that almost no study reported an increased risk of CHD associated with habitual coffee consumption of four or less cups per day, corresponding to about 400 mg per day of caffeine, which may be an underestimation of total caffeine intake considering that other sources of caffeine were not taken into account in the majority of the studies”

In the paper

“Using categorical analysis, an inverse U-shaped association between caffeine consumption and incident hypertension was found.Compared with participants in the lowest quintile of caffeine consumption (mean intakes 14.8 and 19.6 mg per day), those in the third quintile (209.3 and 174.7 mg per day) had a significant 14 % and 15 % increased risk of hypertension in NHS I and NHS II, respectively,”

What the sources say

• Not checked yetcites Winkelmayer et al., 2005 source

In the paper

“Caffeinated coffee drinking was also positively associated with initiation of antihypertensive treatment (p for trend = 0.024 across categories of coffee intake: zero to one cup per day, two to three cups per day, four to five cups per day, six to seven cups per day and eight or more cups per day) in the study by Hu et al. (2007), which enrolled never-treated hypertensive and normotensive subjects”

What the sources say

• Not checked yetcites Hu et al., 2007

In the paper

“The Panel also notes that the interaction between coffee intake and sex on the incidence of hypertension was not statistically significant, and thus subgroup analyses by sex were not statistically justified.”

In the paper

“Although it has been suggested that polymorphisms of the CYP1A2 gene may affect the risk of hypertension associated with caffeine consumption and explain in part the different findings among studies, this hypothesis has not been tested prospectively in unselected populations.”

In the paper

“In a dose-response meta-analysis, the best-fitting model showed an inverse “J-shaped” curve (p for quadratic term < 0.001), with the risk of hypertension increasing up to three cups per day (RR for the comparison of three cups with zero cups per day: 1.07, 95 % CI = 0.97-1.20)and decreasing with higher intake (RR for the comparison of six cups with zero cups per day: 0.99, 95 % CI = 0.89-1.10).”

What the sources say

• Not checked yetcites Zhang et al., 2011 source

In the paper

“A significant effect of caffeinated coffee on BP was observed only by pooling studies using five or more cups of coffee, but not in studies using ≤ 4.5 cups.”

What the sources say

• Not checked yetcites Jee et al., 1999 source

In the paper

“The Panel also notes that different types of caffeinated beverages were differently associated with the risk of incident hypertension.Whereas caffeinated (but not decaffeinated) coffee was inversely associated with the risk of hypertension, caffeinated tea (in one cohort) and cola beverages (in both cohorts) were directly associated with that risk.”

What the sources say

• Not checked yetcites Winkelmayer et al., 2005 source

In the paper

“Caffeine consumption of > 200 mg per day during pregnancy was associated with a reduction in birth weight of about 60-70 g, with a significant trend for greater reduction in birth weight with higher caffeine intake (p = 0.004).”

What the sources say

• Not checked yetcites CARE Study Group, 2008 source

In the paper

“Prospective cohort studies on the relationship between habitual caffeine intake and long-term changes in BP and on the risk of incident hypertension are conflicting and difficult to interpret.”

In the paper

“Overall, the results from these studies support a positive association between caffeine intake and the risk of adverse birth weight-related outcomes, whereas the relationship between caffeine consumption during pregnancy and other pregnancy outcomes (e.g. in relation to length of gestation or fetal death) was less consistent (COT, 2008)”

What the sources say

• Not checked yetcites COT, 2008