• Decrease font size
  • Return font size to normal
  • Increase font size
U.S. Department of Health and Human Services

Database of Select Committee on GRAS Substances (SCOGS) Reviews

  • Print
  • Share
  • E-mail


Return to Listing
Report No.:  89
Type of Conclusion:  3,4
ID Code:  58-08-2
Year:  1978
CFR Section:  182.1180
SCOGS Opinion: 

Caffeine has been consumed by man for centuries in coffee and tea. This opinion is not concerned with such naturally ocurring caffeine in foods. It is addressed solely to caffeine as commercially added to food commodities. Cola beverages comprise the largest and only significant source of caffeine in this latter group. Such beverages have been in use for decades.

The Select Committe's opinion on the health aspects of caffeine as a commercially-added food ingredients rests primarily upon the integrated interpretation of a cluster of eight factors. These are:

1. Levels of consumption of caffeine as a commercially-added food ingredient: Three points of reference have been used: (a) the total amount of caffeine added to foods in the United States supplies about 0.2 mg per kg per day for the population as a whole; (b) most cola drinkers consume about 0.3 mg of caffeine per kg per day from this source with a few in the 1 to 5 years range consuming as much as 1.8 mg per kg per day; (c) the consumption of a 12-ounce container of cola beverage containing 0.01 percent caffeine represents a dose of about 0.9 mg per kg for a 40kg child, for example, or about 0.6 mg per kg for an adult. It is to be noted that these figures represent the amount of caffeine solely form cola drinks. They, therefore, represent the minimum levels of caffeine consumption on the part of individuals consuming a range of caffeine-containing foods and beverages.

2. History of cola consumption: Despite widespread human consumption of cola beverages over many years, the literature contains no definitive studies of possible long-term effects.

3. Mutagenicity: Caffeine causes chromosomal damage in certain microbial and other non-mammalian test systems, and caffeine has similar effects at high concentrations on mammalian cells in culture in several in vitro tests. However, in vivo tests utilizing mice and rats have failed to demonstrate mutagenic effects of caffeine.

4. Teratogenicity: Many animal tests showed that teratogenic effects are generally absent at caffeine doses up to 50 mg per kg body weight. At doses up to 75 mg per kg of body weight, teratogenic effects of caffeine are neither striking nor consistently demonstrated. At bolus doses greater than 75 mg per kg terratogenic effects are apparent. Two retrospective studies of more than 14,000 mothers on whom caffeine consumption histories were obtained, revealed no association between caffeine intakes and abnormalities in offspring.

5. Carcinogenicity: A very recent and as yet unpublished report has indicated that rats given caffeine orally at daily doses of 150 to 250 mg per kg for 15 months develop cancer of several organs. Epidemiological studies suggest that there is no casual relationship between drinking coffee or the caffeine contained in it and cancer.

6. Long term feeding studies

Two to four generation studies with rats intubated daily with up to 30 mg per kg body weight of caffeine or supplied the same dosage in drinking water, showed no consistent dose-related effects. Mice given 250 mg caffeine per kg body weight daily in drinking water throughout life, continued to breed satisfactorily. Fertility of eggs from pullets fed 80mg caffeine per kg per day was not impaired but embryonic losses were about 16 percent compared to 5 percent in controls. Marked but reversible reduction in sperm output and concentration was found in roosters after 3 weeks feeding of 160mg caffeine per kg per day.

7.Dose effects in humans. The pharmacological dose of caffeine used to stimulate central nervous system activity in humans is about 3 mg per kg and is observable at about 2 mg per kg. The acute human fatal dose of caffeine appears to be greater than 170 mg per kg. Oral administration of caffeine (4 mg per kg) has been found to increase blood pressure in fasted individuals.

8. Behavioral effects on children: Concerns with respect to behavioral effects are less for adults, particularly when the amount of caffeine consumed as cola beverages is compared to that consumed in coffee, tea or other natural sources of caffeine, than for children where there can be chronic consumption of caffeine in colatype beverages during the period of brain growth and development. It is during this period of plasticity that the developing central nervous system is most sensitive to the effects of all aspects of the environment. The estimated levels of caffeine intake at these ages are near those levels that are known to cause central nervous system effects in adults. A few short-term human studies suggest that infants (treated neonatally with caffeine for apnea) were not more sensitive than adults to caffeine, that neonatal caffeine treatment did not predispose to neurological or intellectual residua when the children were evaluated 2 years post-treatment, and that caffeine had no observable behavioral effects on either normal or hyperactive school aged children. Other studies indicate that children may be more sensitive than adults to the stimulant effect of caffeine. In mice, rats and monkeys noticeable behavioral effects have been elicited at dose levels as low as 1 to 2 mg of caffeine per kg, with definite effects apparent at doses upward of 10 mg per kg.

In addition to considering these eight factors, the Select Committee views with concern the continued addition of caffeine to cola-type beverages, representing as it does a unique addition to food of a pharmacologically active central nervous system stimulant. The amount of caffeine consumed as cola type beverages borders on the dose known to produce central nervous system stimulation in animals and man. Whether such stimulation constitutes an adverse effect or whether a potential hazard may exist for the segment of the population, particularly children, that is exposed chronically to stimulating doses of caffeine, cannot be answered on the basis of the evidence now available. Despite the long history of use and absence of definitive evidence of toxicity in mammalian in vivo test systems, there is a possibility that behavioral effects in children from the consumption of caffeine from infancy through adolescence exist, even though these potential effects are neither adequately documents nor are their consequences clear. The Select Committe is well aware that caffeine in cola-type beverages is not the only dietary source of this substance. However, based on the evidence available in May 1978, the Committee believes it is not appropriate to continue to consider caffeine as a generally recognized as safe substance for addition to cola-type beverages. A series of rigorously controlled chronic studies in appropriate species, including fetal, neonatal, and growing animals, of the immediate and ultimate behavioral and cardiovascular effects of caffeine added to the diet and given in cola-type beverages, would reduce areas of speculation in this regard. Such animal studies should include doses equivalent to the present estimated daily intakes of caffeine due to consumption of cola-type beverages containing 0.01 percent caffeine, and multiples of that dosage. In addition, appropriate pharmacokinetic studies (acute and chronic) in human subjects appear to be essential in order to establish patterns of absortion, biotransformation and excretion of caffeine. Such studies should consider caffeine as it occurs in cola-type beverages, as it occurs in natural sources, and in combinations of the two in common use by the general population, in order to study the effects of ranges of total body loads. Such data would be useful in providing a sounder basis for interpreting the possible long term health implications of caffeine consumption. In light of these considerations, the Select Committee concludes that: A. While no evidence in the available information on caffeine demosntrates a hazard to the public when it is used in cola type beverages at levels that are now current and in the manner now practiced, uncertainties exist requiring that additional studies be conducted. B. It is inappropriate to include caffeine among the substances generally recognized as safe (GRAS). At current levels of consumption of cola-type beverages, the dose of caffeine can approximate that known to induce such pharmacological effects as central nervous system stimulation. A minority report of Select Committee member R.G.H. Siu appears as an appendix to this report.