Caffeine is the most widely used psychoactive drug in the world. In the US, an estimated 87% of the population regularly consume beverages containing caffeine. Mean caffeine consumption among all users is 193 mg per day, with the highest intake among men aged 35–54 who consume an average of 336 mg of caffeine per day.
Caffeine is a natural constituent of more than 60 species of plants, including coffee, tea, cola nut, cacao, yerba mate, and guarana. Caffeine (1,3,7-trimethylxanthine) is a member of the methylxanthine class of alkaloids that include theobromine and theophylline. In its free base form, caffeine is a bitter white powder that is moderately soluble in water (21.7 mg ml1). Worldwide, caffeine is most commonly consumed as coffee and tea. Consumption of beverages with added caffeine (i.e., soft drinks) has markedly increased over the past half century, with consumption volume of soft drinks now being approximately twice that for coffee in the US. A notable recent trend has been the increasing popularity of ‘‘energy drinks’’ which vary considerably in the amount of caffeine (from 50 mg to over 375 mg of caffeine per can or bottle). Hundreds of such products are now marketed in the US.
Caffeine is rapidly and completely absorbed from the gastrointestinal tract after oral administration. Caffeine is readily distributed throughout the body and is found in all body fluids. Peak plasma concentrations are typically reached 30–45 min following oral ingestion. The fraction of caffeine bound to plasma protein is 10–35%.
More than 25 caffeine metabolites have been identified in humans. The primary metabolic pathways involve the cytochrome P-450 liver enzyme system which produces three demethylated active metabolites: paraxanthine, theobromine, and theophylline, accounting for 80%, 10%, and 4% of caffeine metabolism, respectively. The average half-life of caffeine is 4–6 h, with elimination rates varying by more than 10-fold across individuals. Caffeine half-life is prolonged in individuals with liver disease and during the end of pregnancy. Caffeine metabolism is inhibited by numerous compounds including oral contraceptive steroids, cimetidine, some quinoline antibiotics, fluvoxamine, mexiletine, and high doses of caffeine itself. Neonates have a markedly increased caffeine half-life (80–100 h) due to immature liver enzyme systems which are fully developed at approximately 6 months of age. Tobacco smoking increases caffeine metabolism by stimulating the cytochrome P-450 1A2 enzyme (CYP1A2), with smokers metabolizing caffeine about twice as fast as non-smokers. Genetic variations in CYP1A2 activity are a significant determinant of caffeine metabolism. For more information on individual differences in genetics, see the Section on Caffeine Genetics.
Mechanisms of Action
The primary cellular site of action of caffeine is the adenosine receptor. Among the adenosine receptor subtypes that have been identified, A1 and A2A receptors are the preferential targets of caffeine. A1 receptors are widely expressed throughout the brain, whereas A2A receptors are concentrated in dopaminergic- rich areas, such as the striatum. Adenosine is an endogenous purine nucleoside that generally exerts inhibitory effects throughout the central and peripheral nervous system (e.g., excitatory neurotransmitter inhibition, suppression of motor activity, and inhibition of gastric secretion). Caffeine is a nonselective competitive A1 and A2A receptor antagonist. Thus, caffeine produces a variety of central and peripheral effects that are opposite to the effects of adenosine.
Of most relevance to caffeine’s central nervous system (CNS) stimulating effects, caffeine enhances dopamine activity indirectly by competitive antagonism of adenosine receptors that are co-localized and functionally interact with dopamine. Adenosine receptors can form functional receptor heteromers with dopamine receptors (i.e., A1-D1 and A2A-D2). There is some evidence that the motor stimulant and reinforcing effects of caffeine are mediated by dopamine. Preclinical studies demonstrate that caffeine produces behavioral effects similar to the dopamine-mediated effects of classic stimulants, such as cocaine and amphetamine. Moreover, dopamine depletion or blockade of dopamine receptors significantly impairs the motor stimulant and discriminative stimulus effects of caffeine.
Although caffeine can inhibit phosphodiesterase and increase intracellular calcium concentration, typical dietary doses of caffeine are believed to be too low to be significantly influenced by these nonadenosine mechanisms. Thus, caffeine’s effects appear to be primarily mediated by direct antagonism of adenosine and indirect enhancement of brain dopamine activity. For more information on ergogenic mechanisms of action, see the Section on Caffeine and Exercise.
Physiological and Health Effects of Caffeine
Caffeine modestly increases blood pressure but appears to have no effects on or to reduce heart rate. Hypertensive and hypertensive-prone caffeine users appear to be particularly sensitive to the pressor effects of caffeine. Caffeine produces increases in gastric acid secretion, colonic stimulation, diuresis (30% or more increased volume), respiratory stimulation, and bronchodilation. Caffeine also increases plasma epinephrine, norepinephrine, adrenocorticotropic hormone, cortisol, renin, and free fatty acids. Acute caffeine administration produces increased cerebral blood flow velocity and electroencephalography (EEG) beta power activity. In addition, caffeine has prominent sleep-disrupting effects. For more information on the sleep disrupting effects of caffeine see the Section on Caffeine-Induced Sleep Disorder.
Although studies of the association between caffeine consumption and coronary heart disease have yielded inconsistent findings, one recent investigation demonstrated an association between slow caffeine metabolism and the incidence of coronary heart disease in moderate and heavy caffeine consumers. For more information on caffeine genetics, see the Section on Caffeine Genetics. Several studies have found that moderate coffee intake is associated with decreased risk for coronary heart disease, possibly because of protective effects of antioxidant and other protective compounds in coffee.
Over the last few decades, studies have yielded inconsistent findings regarding the effects of caffeine on reproductive and perinatal outcomes. Although some investigations have not found evidence of a significant association between caffeine exposure and adverse birth outcomes, several recent studies did show a relationship. Although equivocal findings preclude definitive conclusions regarding the effects of caffeine on pregnancy, some governmental health agencies have taken a prudent stance and issued health warnings to limit the use of caffeine during pregnancy. Health Canada recommends that women of reproductive age consume no more than 300 mg of caffeine per day. The Food Standards Agency of the UK advises that pregnant women keep their daily intake of caffeine below 200 mg.
Caffeine and Performance
Numerous investigations have examined the effects of caffeine on human performance. The most consistent finding to emerge is that caffeine restores performance that has been degraded by sleep deprivation, fatigue, or prolonged vigilance. At normal dietary doses, caffeine may improve tapping speed, reaction time, and sustained attention (vigilance), although results have been variable and sizes of the effects are often modest. A large number of experimental studies have examined the effects of caffeine on memory, but evidence is insufficient to conclude that caffeine produces acute improvements in memory.
A significant limitation of the majority of studies that have found performance-enhancing effects of caffeine is that subjects in these studies have been regular caffeine users who were required to abstain from caffeine before testing (e.g., overnight abstinence). Thus, the observed performance-enhancing effects of caffeine in these studies may reflect restoration of deficits that are produced by caffeine withdrawal. For more information on the caffeine withdrawal syndrome, see the Section on Caffeine Withdrawal. It is important to note, however, that a few studies have found performance-enhancing effects of caffeine in non-dependent caffeine users and non-users. A few studies have also demonstrated performance increases in caffeine users who were not required to abstain from usual caffeine use, suggesting that complete tolerance to the performance-enhancing effects of caffeine does not occur at usual dietary doses. However, in high-dose caffeine consumers, performance enhancement beyond withdrawal reversal is likely to be modest.
Caffeine and Exercise
A large body of research has examined the effects of caffeine on exercise performance. Numerous well-controlled studies have found that relative to placebo, caffeine enhances performance during endurance exercise. Studies have also generally found that caffeine reduces ratings of perceived exhaustion or effort during exercise. Ergogenic effects of caffeine are typically demonstrated at doses of 3–6 mg per kg; higher doses of caffeine (e.g., 9 mg per kg) appear to exert little or no additional benefit on endurance exercise. There is some evidence that caffeine produces greater endurance exercise benefits in caffeine nonusers and in athletes who abstained from caffeine for several days before dosing. Findings from studies examining the effects of caffeine on short-term, highintensity exercise performance have generally been equivocal, however a recent review suggested that caffeine can improve performance in team-sports exercise and power-based sports, with this effect more common in elite athletes who do not regularly consume caffeine. Although not rigorously studied, findings are suggestive that tolerance occurs to the ergogenic effects of caffeine. Several non-independent mechanisms have been proposed for caffeine’s effects on exercise performance, including increased fatty acid oxidation, increased availability of muscle glycogen, mobilization of intracellular calcium, increased muscle contractile force, and direct CNS effects via adenosine antagonism.
Much of the variability in caffeine consumption and individual differences in response to caffeine can be accounted for by genetic factors. Findings from twin studies indicate that there may be common genetic factors underlying the use of caffeine, nicotine, and alcohol. Moreover, twins studies indicate that genetic factors may influence total caffeine consumption, heavy caffeine consumption, caffeine tolerance, caffeine withdrawal, caffeine intoxication, and caffeine-related sleep disturbances.
The CYP1A2 gene codes for the isoenzyme P-450 1A2, which is responsible for the demethylation of caffeine to paraxanthine, theobromine, and theophylline. For more information on caffeine metabolism, see the Section on Caffeine Pharmacokinetics. More than 150 CYP1A2 single-nucleotide polymorphisms have been identified. Individual variability in the pharmacokinetics of caffeine can be in large part accounted for by variations in CYP1A2 activity. Recent evidence suggests that individuals homozygous for the allele associated with slow metabolism (CYP1A2 1F) are at increased risk for non-fatal myocardial infarction associated with caffeinated coffee intake. Thus, caffeine consumption may increase risk for myocardial infarction in individuals with slow caffeine metabolism.
Genetic differences in adenosine A2A receptors have been implicated in individual differences in human caffeine responses. Variations in A2A receptor polymorphisms have been associated with caffeine sensitivity, caffeine-induced anxiety, caffeine-related sleep impairment, and caffeine consumption. One study reported evidence that a polymorphism in dopamine DRD2 receptors is associated with caffeine-induced anxiety.
Caffeine Subjective Effects
The qualitative subjective effects of caffeine depend on caffeine dose, individual differences in sensitivity, and degree of tolerance to caffeine. Low to moderate doses of caffeine typically produce positive subjective effects, including increased well-being, arousal, energy, alertness, concentration, motivation to work, and sociability, and decreased feelings of sleepiness or tiredness. Positive subjective effects are more likely to be reported in individuals who have undergone overnight caffeine abstinence.
At higher acute doses of caffeine (e.g., 400–800 mg), negative subjective effects of caffeine typically emerge. Negative subjective effects include increased anxiety, nervousness, jitteriness, tense negative mood, and upset stomach. Anxiogenic subjective effects are more likely to be reported in individuals with panic disorder or generalized anxiety disorder, and in non-clinical populations who endorse high levels of anxiety sensitivity (i.e., fear of anxiety). For more information about high dose caffeine effects, see the Section on Caffeine Intoxication.
The efficacy of a substance in establishing or maintaining self administration behavior reflects the reinforcing effects of the substance. The circumstantial evidence indicating that caffeine functions as a reinforcer is compelling. Caffeine is the most widely used mood-altering drug in the world. Regular daily consumption of pharmacologically active doses occurs in widely varying cultural and social contexts. Historically, efforts to restrict or eliminate consumption of caffeine-containing foods and beverages have been unsuccessful. Caffeine consumption occurs in a wide variety of vehicles (e.g., coffee, tea, mate´ soft drinks, energy drinks; chewing kola nuts). Finally, caffeine-containing beverages tend to be more popular than their caffeine-free counterparts. As an example, in 2009, in the US, the top six selling carbonated soft drink brands, and eight of the top 10 selling brands, contained added caffeine.
Numerous well-controlled experimental studies have demonstrated caffeine reinforcement in various subject populations. Across studies, approximately 40% of normal caffeine users demonstrate caffeine reinforcement. Higher rates of reinforcement have been observed among individuals with high levels of caffeine consumption or a history of drug or alcohol abuse. Caffeine can function as a reinforcer at very low doses (i.e., 25 mg per cup of coffee), but may produce avoidance at higher doses (e.g., 400 or 600 mg).
In habitual caffeine consumers, avoidance of caffeine withdrawal plays an important role in the reinforcing effects of caffeine. This relationship has been shown in retrospective questionnaire studies and in experimental studies that have used direct behavioral measures of reinforcement. For example, in one experimental study, moderate caffeine consumers who reported withdrawal symptoms (i.e., headaches and drowsiness) were more than twice as likely to show caffeine reinforcement. Other studies that have prospectively manipulated caffeine physical dependence have demonstrated that subjects were more than twice as likely to exhibit caffeine reinforcement when they were caffeine physically dependent (and thus prone to experiencing withdrawal symptoms when they abstain).
Studies using a conditioned flavor preference paradigm have provided indirect evidence of caffeine reinforcement. In these studies, caffeine abstinent subjects develop a liking and preference for caffeine-paired flavored beverages, relative to beverages paired with placebo. Further studies showed that, in subjects who were repeatedly exposed to a caffeine-paired flavored beverage, the development of liking and preference for the beverage was determined by alleviation of withdrawal symptoms. These studies suggest that conditioned flavor preferences (driven at least in part by alleviation of unpleasant withdrawal symptoms) likely play an important role in consumer preferences for caffeine-containing beverages.
Cessation or reduction of daily caffeine consumption results in withdrawal symptoms in many caffeine users. Caffeine withdrawal has been well characterized in numerous rigorous experimental studies and in survey studies. Caffeine withdrawal headache, which is the hallmark feature of the caffeine withdrawal syndrome, has been the most frequently assessed withdrawal symptom. Approximately half of regular caffeine users report headache when abstaining from caffeine. The caffeine withdrawal headache, which develops gradually, is described as diffuse, throbbing, severe, and phenomenologically distinct from migraine headache. In addition to headache, other withdrawal symptoms that have been reliably observed across experimental and survey studies include: fatigue, decreased energy/activeness, decreased alertness, drowsiness, decreased contentedness, depressed mood, difficulty concentrating, irritability, foggy/not clear-headed, nausea/ vomiting, and muscle stiffness/pain.
Based on these observed symptoms, five primary clusters of withdrawal symptoms have been proposed:
- fatigue and drowsiness,
- dysphoric mood, depressed mood, or irritability,
- difficulty concentrating, and
- flu-like somatic symptoms, nausea, vomiting, or muscle pain/stiffness.
Onset of withdrawal typically occurs 12–24 h after abrupt caffeine cessation, although onset has been observed as early as 6 h and as late as 43 h after abstinence in some individuals. Typically, peak intensity of symptoms occurs 1 to 2 days after abstinence. The duration of caffeine withdrawal symptoms is generally 2–9 days. Re-administration of caffeine (usually within 30–60 min of onset) rapidly and often completely reverses caffeine withdrawal. Figure 1 shows the time-course of caffeine withdrawal from an illustrative double-blind experimental study.
The severity of caffeine withdrawal symptoms can range from mild to extreme and depends on several factors. Some caffeine users report clinically significant functional impairment associated with caffeine withdrawal (e.g., interference with work or child care activities). Studies show that clinically significant distress occurs in approximately 13% of caffeine users. A much higher rate (73%) of withdrawal-related clinically significant distress occurs in individuals meeting criteria for caffeine dependence. Severity of caffeine withdrawal is positively associated with increases in caffeine maintenance dose, such that greater withdrawal is experienced following cessation of higher maintenance doses. Caffeine withdrawal can occur after daily doses of caffeine as low as 100 mg per day. Caffeine withdrawal may also occur when lower doses of caffeine are substituted for the maintained caffeine dose. As the substituted dose of caffeine decreases, withdrawal severity increases. Nevertheless, even a small amount of substituted caffeine (e.g., 25 mg) can mitigate severity of caffeine withdrawal symptoms.
Caffeine tolerance may occur in response to daily caffeine consumption. Tolerance can occur to the subjective, sleep disrupting, and physiological effects of caffeine. The degree of caffeine tolerance depends on several factors including the challenge and maintenance doses, frequency of administration and individual differences in elimination rate. The prevalence of self-reported tolerance among current caffeine users varies from 8% to 50%. Rates as high as 92% have been reported among caffeine-dependent individuals.
Regular caffeine users may acquire complete tolerance (i.e., no difference between placebo and caffeine after prolonged exposure to caffeine) to some, but not all, of the subjective effects of caffeine. For example, experimental studies showed that volunteers who received moderate to high doses of caffeine (400 mg to 900 mg per day) for at least two weeks developed complete tolerance to ratings of subjective stimulant effects. Other studies indicate that complete tolerance to caffeine subjective effects does not occur at lower caffeine doses and over shorter periods of dosing.
Tolerance development may differ across different outcome measures. One study showed that tolerance or complete tolerance developed to subjective ratings but not to measures of cerebral blood flow or EEG in volunteers receiving 400 mg per day.
With regard to other physiological effects, tolerance may develop to the effects of caffeine on diuresis, parotid gland salivation, metabolic rate, plasma norepinephrine and epinephrine levels, and plasma renin activity. Findings suggest that regular caffeine users develop partial, but not complete tolerance to the effects of caffeine on cerebral blood flow and EEG measures. Two studies, which tested a small number of volunteers demonstrated the development of complete tolerance to the pressor effects of high doses of caffeine (e.g., 600 to 850 mg per day). However, several more recent studies examining a larger number of volunteers showed that tolerance to the pressor effects of caffeine is variable across individuals, with some subjects showing complete tolerance, whereas others show only incomplete tolerance.
Caffeine intoxication, which is a DSM-IV-TR recognized disorder, is defined by the development of symptoms and clinical features in response to acute caffeine consumption that cause clinically significant distress or impairment. Although the DSM-IV-TR definition specifies that the diagnosis depends on recent consumption of at least 250 mg of caffeine, symptoms typically emerge at doses greater than 500 mg. Symptoms of caffeine intoxication include restlessness, nervousness, insomnia, flushed face, diuresis, gastrointestinal disturbance, muscle twitching, rambling flow of thought and speech, tachycardia or cardiac arrhythmia, periods of inexhaustibility, and psychomotor agitation. Although children and caffeine intolerant individuals may be particularly sensitive to the acute adverse effects of caffeine, habitual caffeine users may also experience episodes of caffeine intoxication. Several case reports and experimental studies suggest that caffeine consumption may produce hallucinations in some individuals, particularly under conditions of stress.
Caffeine-Induced Sleep Disorder
Caffeine reduces total sleep time and limits latency to sleep onset, most probably by blocking the sleep promoting effects of adenosine. In addition, caffeine decreases stage 3–4 sleep and suppresses EEG slow wave activity during sleep. The sleep disrupting effects of caffeine are well documented even at low doses (e.g., one cup of coffee). Surveys have found associations between daily dietary caffeine intake and sleep problems in both adults and adolescents. Some caffeine users may develop caffeine-induced sleep disorder, which is a DSM-IV-TR recognized disorder typically characterized by insomnia. Some caffeine users may present with caffeine-induced hypersomnia with daytime sleepiness due to withdrawal symptoms. Sleep disturbances secondary to caffeine may increase in severity as caffeine dose increases and proximity to caffeine administration at bedtime decreases. Individuals who are not regular caffeine users and are not tolerant, or have only partial tolerance to the sleep-disrupting effects of caffeine are more likely to experience caffeine-related sleep disruption.
Caffeine-Induced Anxiety Disorder
In addition to the symptom of anxiety that can be a component of caffeine intoxication, caffeine can also produce caffeine-induced anxiety disorder, a DSM-IV-TR disorder. Presentation of a caffeine-induced anxiety disorder may include symptoms of generalized anxiety, panic attacks, obsessive–compulsive disorder, or phobic disorder. Individuals who have an existing anxiety disorder, or who endorse symptoms of anxiety sensitivity, are at increased risk of experiencing anxiety symptoms in response to caffeine.
Substance dependence is characterized by a cluster of cognitive, behavioral, and physiological symptoms indicating that an individual is continuing to use a substance despite experiencing clinically significant substance-related problems. Caffeine dependence is recognized as a diagnosis in ICD-10, the official diagnostic system of the World Health Organization. In contrast, the DSM-IV-TR currently excludes caffeine from a diagnosis of substance dependence despite using very similar diagnostic criteria to ICD-10. A growing literature from experimental studies, clinical interviews, and survey studies indicates that some caffeine users manifest a pattern of symptoms consistent with a DSM-IV-TR diagnosis of substance dependence as applied to caffeine.
One population-based survey study of 162 randomly selected caffeine users found that 9% of the sample endorsed three or more of four DSM-IV-TR substance dependence criteria that are thought to be most relevant of a meaningful diagnosis of caffeine dependence. The criteria and past-year incidence were:
- Persistent desire or unsuccessful efforts to cut down or control use (56%);
- Characteristic withdrawal syndrome or substance taken or relieve or avoid withdrawal (18%);
- Use is continued despite a physical or psychological problem likely caused or exacerbated by the substance (14%); and
- Tolerance defined by either a need for markedly increased amounts to achieve desired effect or markedly diminished effect with continued use of the same amount (8%).
Individuals meeting criteria for caffeine dependence vary considerably in the amount of caffeine consumed per day and in the types of caffeine-containing products that they regularly consume (e.g., coffee, soft drinks, tea). Importantly, the problems associated with caffeine dependence are not trivial. These include, but are not limited to anxiety, insomnia, stomach problems, and cardiovascular problems. One survey found that 13% of caffeine users had been advised by a physician or counsellor to reduce or cut down caffeine in the last year. Fifteen percent of caffeine consumers were particularly resistant to modifying their use, indicating that they would not change when or how much caffeine they used, no matter what they were doing or where they were.
Caffeine and Food
In addition to being consumed in its natural plant forms (e.g., coffee and tea), caffeine is also frequently consumed as an added ingredient in many popular sugar-sweetened soft drinks and energy drinks. The bitter taste profile of caffeine is often masked or obscured by the addition of sugar, fat, and other flavors in caffeine-containing foods and beverages. Beverage manufacturers have made the claim that caffeine is added to beverages in order to enhance flavor, but most individuals are unable to detect flavor differences between sugar-sweetened soft drinks with and without caffeine at the caffeine concentration found in most soft drink beverages. It is likely that caffeine-containing beverages are widely consumed because caffeine can function as a reinforcer, increase flavor preferences for caffeine-containing beverages, and produce physical dependence, which results in a substance dependence syndrome.
As described above, caffeine-dependent users may be unable to cut down or control caffeine use despite a persistent desire to do so, and may also continue to use caffeine despite medical problems associated with caffeine consumption. Thus, caffeine dependence may exacerbate adverse health outcomes associated with the consumption of caffeine-containing sugar-sweetened beverages. Sugary drinks have been associated with weight gain, obesity, and type-2 diabetes even after controlling for other factors. Of particular concern is that sugar-sweetened beverage consumption is associated with weight gain and obesity in children and may displace milk and other important nutrients in the diets of children and adolescents.
Caffeine is an added ingredient in many over-the-counter weight loss medications. Experimental studies have generally found that acute caffeine consumption is associated with increased energy expenditure, decreased food intake, and reduced ratings of hunger. There is also some evidence that caffeine increases fat oxidation. It is not clear whether any or all of these effects are due to acute caffeine effects per se versus reversal of caffeine withdrawal (e.g., the observed increased energy expenditure may be due to the reversal of suppressed energy expenditure in the caffeine-deprived comparison condition). Prospective longitudinal studies have shown that caffeine consumption is negatively associated with weight gain, but few well-controlled studies have examined the long-term efficacy of caffeine alone as an intervention for weight loss and weight loss maintenance.
Experimental studies have shown that the co-ingestion of caffeine and catechins enhances energy expenditure and fat oxidation more than an equivalent amount of caffeine without added catechins. Numerous investigations have examined the combined effects of caffeine and catechins, most commonly co-ingested in green tea, on weight loss outcomes. A recent meta-analysis of studies comparing a caffeine/catechin condition with either a placebo condition or a low-dose caffeine/catechin condition found that the caffeine-catechin combination had small positive effects on weight loss and maintenance of weight loss after a period of negative energy balance. There is some evidence that effects are attenuated in habitual caffeine consumers with high daily caffeine intake (e.g., 4300 mg per day). This effect may be mediated through caffeine tolerance. Several studies have found that caffeine in combination with ephedrine is efficacious for weight loss. However, ephedrine alkaloid supplements have been associated with adverse events and have been banned by the Food and Drug Administration (FDA) in the US.
Astorino TA and Roberson DW (2010) Efficacy of acute caffeine ingestion for short term high-intensity exercise performance: A systematic review. Journal of Strength and Conditioning Research 24 (1): 257–265.
Cornelis MC and El-Sohemy A (2007) Coffee, caffeine, and coronary heart disease. Current Opinion in Clinical Nutrition and Metabolic Care 10: 745–751.
Ferre´ S (2008) An update of the mechanisms of the psychostimulant effects of caffeine. Journal of Neurochemistry 105: 1067–1079.
Ganio MS, Klau JF, Casa DJ, Armstrong LE, and Maresh CM (2009) Effect of caffeine on sport-specific endurance performance: A systematic review. Journal of Strength & Conditioning Research 23 (1): 315–324.
Griffiths RR and Reissig CJ (2008) Substance abuse: Caffeine use disorders. In: Tasman A, Kay J, Lieberman J, First MB, and Maj M (eds.) Psychiatry, 3rd edn, vol. 1, pp. 1019–1040. Chichester, UK: John Wiley & Sons.
Juliano LM, Ferre´ S, and Griffiths RR (2009) Caffeine: Pharmacology and clinical effects. In: Ries RK, Fiellin DA, Miller SC, and Saitz R (eds.) Principles of Addiction Medicine, Fourth Edition, pp. 159–178. Philadelphia: Lippincott Williams & Wilkins.
Juliano LM and Griffiths RR (2004) A critical review of caffeine withdrawal: Empirical validation of symptoms and signs, incidence, severity, and associated features. Psychopharmacology 176(1): 1–29.
Malik VS, Schulze MB, and Hu FB (2006) Intake of sugar-sweetened beverages and weight gain: A systematic review. American Journal of Clinical Nutrition 84: 274–288.
Yang A, Palmer AA, and de Wit H (2010) Genetics of caffeine consumption and responses to caffeine. Psychopharmacology 211(3): 245–257.