skip page navigationOregon State University

News Articles

The New Recommendations for Dietary Antioxidants:
A Response and Position Statement
by the Linus Pauling Institute

The dietary reference intakes (DRIs) for nutrients are values established by the Food and Nutrition Board (FNB) of the National Academy of Sciences that encompass the estimated average requirement (EAR), the recommended dietary allowance (RDA), the adequate intake, and the tolerable upper intake level. The RDA is defined as "the average daily dietary intake level that is sufficient to meet the nutrient requirement of nearly all (97 to 98 percent) healthy individuals in a particular life stage and gender group."

Since the last publication of the RDAs in 1989, the FNB has changed the criteria for establishing RDAs from prevention of deficiency diseases to prevention of chronic diseases. Nevertheless, the revised RDAs recently published by the Panel on Dietary Antioxidants and Related Compounds of the FNB were based primarily on the prevention of deficiency symptoms, not the prevention of chronic diseases. Not surprisingly, therefore, the new RDAs for dietary antioxidants differ little from the RDAs published in 1989. The new RDAs for adults (19 yrs and older) are: vitamin C, 90 milligrams (mg)/day for men and 75 mg/day for women (old RDA: 60 mg/day for both men and women); vitamin E (natural or d-alpha-tocopherol), 15 mg/day (22 International Units) for men and women (old RDA: 10 and 8 mg/day alpha-tocopherol equivalents, respectively, for men and women); selenium, 55 micrograms (mcg)/day for men and women (old RDA: 70 and 55 mcg/day, respectively, for men and women). The new RDA for vitamin C is set lower for women than for men because of their lower lean body mass, but interestingly the same argument was not applied to the RDAs for vitamin E and selenium. No recommendations were made for beta-carotene and other carotenoids.

The Panel considered the possibility that higher intakes than the RDAs might prevent chronic diseases, but concluded that there was insufficient data to prove that increased intakes of antioxidants exert beneficial health effects beyond the prevention of deficiency symptoms.

The Panel also established values for the tolerable upper intake level (UL) for dietary antioxidants. The UL is defined as "the highest level of daily nutrient intake that is likely to pose no risk of adverse health effects to almost all individuals in the general population." The ULs for adult men and women are: 2 grams/day for vitamin C; 1,000 mg/day for vitamin E (any form of supplemental alpha-tocopherol); 400 mcg/day for selenium; and no recommendations for beta-carotene and other carotenoids.

In this article, the Linus Pauling Institute presents its own summary of the current evidence for health benefits and recommendations for antioxidant vitamins.

Vitamin C

Vitamin C (ascorbic acid) is required for the biosynthesis of collagen, carnitine, and catecholamines. A lack of vitamin C in the diet causes the deficiency disease scurvy, which is prevented by as little as 10 mg/day of vitamin C. The 1989 RDA for vitamin C was 60 mg/day, an intake level that prevents the development of scurvy for about one month on a diet lacking vitamin C. Remarkably, approximately 20% of U.S. adults consume less than 60 mg of vitamin C daily and about 10% consume less than 30 mg.

In addition to and distinct from its known metabolic functions, vitamin C is an important dietary antioxidant, which, according to the Panel, is "a substance in foods that significantly decreases the adverse effects of reactive species, such as reactive oxygen and nitrogen species, on normal physiological function in humans." The adverse effects of these reactive species are oxidative damage to biomolecules, such as lipids, DNA, and proteins. Such oxidative damage has been implicated in chronic diseases, including heart disease, stroke, cancer, and cataract.

In a rigorous review of over 200 research articles on the health benefits of vitamin C, we concluded that the totality of evidence from the available human studies strongly suggests that a dietary intake of about 100 mg/day of vitamin C is associated with a reduced incidence of and mortality from heart disease, stroke, and cancer. Even higher intakes of vitamin C, and possibly supplementation, may be required to reduce the risk of cataract, although the evidence is less compelling due to the limited number of studies. Furthermore, over twenty clinical studies have shown that daily doses of 500 mg of vitamin C or acute doses of 1-3 grams significantly improve the relaxation of blood vessels in humans. Impairment of blood vessel relaxation is a significant risk factor for angina pectoris, heart attacks, and strokes. In addition, evidence indicates that a daily dose of 500 mg of vitamin C can reduce blood pressure in patients with mild to moderate hypertension, another important cardiovascular risk factor. Since heart disease, stroke, and cancer are the three top killers in the U.S., causing about 1.3 million deaths per year, the potential of an adequate vitamin C nutriture, as well as vitamin E and selenium (see below), to benefit public health and reduce the economic and medical costs and concomitant suffering associated with these chronic diseases is substantial.

The level of 100 mg/day of vitamin C required for optimum risk reduction of chronic diseases is derived primarily from epidemiological studies that either assessed vitamin C intake by diet history or measured plasma vitamin C levels. Interestingly, a well-controlled study of vitamin C metabolism in seven healthy men found that saturation of white blood cells (neutrophils), and thus presumably tissues, occurs at a vitamin C intake of 100 mg/day. Therefore, the daily dose of vitamin C that results in cell saturation is also associated with a reduced incidence of chronic diseases.

Using the criteria established by the FNB to calculate RDAs (RDA = 1.2 x EAR), our above proposed average requirement of 100 mg/day of vitamin C corresponds to an RDA of 120 mg/day. This is greater than the new RDA for men and women of 90 and 75 mg/day, respectively. The Panel also used neutrophil saturation as a criterion to estimate the average requirement, and thus the RDA. The Panel concurred that in healthy young men neutrophils are saturated with vitamin C at a daily intake of 100 mg. However, the Panel used urinary excretion of vitamin C as an additional criterion to determine the EAR. In doing so, it considered the findings that 25 mg of a 100 mg dose of vitamin C is excreted in the urine, but essentially none of a 60 mg dose is excreted. The Panel concluded that 80% of maximal neutrophil concentration with minimal urinary loss is achieved by a vitamin C intake of 75 mg/day. We believe this conclusion is questionable, because there are no published data for an intake or dose of 75 mg/day of vitamin C, either for neutrophil levels or urinary excretion. By setting the EAR for men at 75 mg/day and, consequently, the RDA at 90 mg/day, the Panel used only lack of urinary excretion as a criterion, not neutrophil saturation. If neutrophil saturation is to be the criterion, the RDA should be 120 mg/day.

With respect to the UL, we agree with the Panel that there is no scientific evidence that even very large amounts of vitamin C are toxic or exert adverse health effects. Specifically, in healthy individuals vitamin C does not cause mutations, cancer, birth defects, hardening of the arteries (atherosclerosis), kidney stones, pro-oxidant effects, "rebound scurvy," excess iron absorption, vitamin B12 deficiency, allergic response, or erosion of dental enamel. The Panel used osmotic diarrhea and gastrointestinal disturbances as criteria to determine the UL for vitamin C and arrived at a level of 2 grams/day. We disagree with this conclusion because it is primarily based on data from uncontrolled case reports. Some studies have reported no gastrointestinal disturbances or diarrhea at up to 6 grams/day of vitamin C, and gastrointestinal disturbances have been observed at widely differing threshold levels (from 3 grams/day up to 10 grams/day). More importantly, we believe that diarrhea and gastrointestinal disturbances are not toxic or severe enough effects to justify a UL based on these criteria. Thus, the side effects of vitamin C are generally not serious and individuals experiencing these effects may easily eliminate them by reducing vitamin C intakes.

Based on our review of the literature, we conclude that the RDA for vitamin C should be 120 mg/day for optimum risk reduction of heart disease, stroke, and cancer in healthy individuals. Special populations, such as older adults and individuals with disease, may require substantially larger amounts of vitamin C to achieve optimum body levels and derive therapeutic benefits. Furthermore, we conclude that there is currently no consistent and compelling data for serious adverse effects of vitamin C in humans, and a UL can therefore not be established.

Vitamin E

There are three important aspects of the Panelís recommendation for vitamin E: form, dietary sources, and amount. The new recommendation distinguishes between natural and synthetic forms of alpha-tocopherol because the chemically different natural and synthetic vitamin E are handled differently by the body. Synthetic alpha-tocopherol acetate is only about half as efficacious as the International Units (IU) listed on the label of vitamin E supplements. Synthetic (all rac or dl) alpha-tocopherol contains 8 different forms of alpha-tocopherol (so-called "stereoisomers": RRR, RSR, RRS, RSS, SRR, SSR, SRS, SSS). The naturally occurring stereoisomer of alpha-tocopherol is called RRR-alpha-tocopherol or d-alpha-tocopherol. The body only uses the RRR-, RSR-, RRS-, and RSS-alpha-tocopherol forms and not SRR, SSR, SRS, and SSS, or other naturally occurring vitamin E forms, such as gamma- or delta-tocopherol.

Most of the dietary vitamin E is found in fats and oils. Alpha-tocopherol is found predominantly in canola, olive, and sunflower oils. If one consumes a low-fat diet, it is advisable to closely evaluate vitamin E intake because servings of most foods contain about 1% or less of the vitamin E required daily, except for nuts and fortified foods.

As mentioned above, the DRIs are based on prevention of vitamin deficiency symptoms. To determine the RDA for vitamin E, the Panel used data published in the 1960s on the amounts of vitamin E required to prevent hydrogen peroxide-induced disintegration (lysis) of red blood cells in the test tube. It is questionable whether these test tube experiments are relevant to reactions occurring in our body, and astonishing that no useful data have been published in the last three decades that could be used as the basis for the determination of the RDA for vitamin E.

The question arises whether the new RDA of 15 mg RRR-alpha-tocopherol is sufficient to meet the requirements of all healthy individuals. The amount of vitamin E consumed by most U.S. adults is sufficient to prevent overt symptoms of deficiency like peripheral neuropathy. However, the actual quantities consumed by American adults are closer to 8 mg as assessed from surveys. These low intakes may be real, or they may result from under-reporting of fat intakes. Nevertheless, it is quite possible that many people do not consume sufficient alpha-tocopherol. Moreover, it is not clear whether chronic diseases such as heart disease and cancer are a hallmark of these long-term suboptimal vitamin E intakes.

It is especially noteworthy that studies at the molecular level demonstrate that vitamin E modulates pathways important in heart disease. While studies in animals demonstrate that vitamin E decreases atherosclerosis, supplementation studies in humans investigating the effects on heart attacks and strokes have resulted in conflicting outcomes. It should be pointed out, however, that most of these supplementation studies investigated the treatment effects of vitamin E in patients with existing heart disease, not the prevention of heart disease in healthy individuals. It also has been suggested that in the human studies in which vitamin E had the greatest benefit, the subjects had certain genetic defects. However, most individuals do not know whether or not they have these genetic defects that increase susceptibility to disease. Thus, the concept of orthomolecular nutrition--the right amount of a nutrient in the right tissue at the right time--suggests that higher intakes of alpha-tocopherol may be beneficial if chronic diseases, such as heart disease, stroke, cancer, diabetes, and Alzheimerís disease result, in part, from suboptimal protection by antioxidants. Importantly, the amounts of vitamin E that exerted beneficial effects in intervention studies are not achievable by dietary means.

Because it has not yet been proven that vitamin E prevents or ameliorates chronic diseases, the Panel concluded that a universal recommendation to consume vitamin E supplements is not warranted.

What about possible risks of high alpha-tocopherol intakes? The ability of vitamin E to interfere with platelet aggregation is likely a major contributor to the beneficial effects of vitamin E supplements in reducing heart disease risk. While this may be important in preventing blood clot formation that leads to heart attacks and strokes, it also highlights a potential risk: supplemental vitamin E may increase bleeding tendencies. Additionally, vitamin E may potentiate the inhibitory effects of aspirin on blood-clotting. Therefore, vitamin E supplement users should inform their physicians that they are taking vitamin E. The UL established by the FNB of 1,000 mg/day of alpha-tocopherol should not be exceeded.

For those who would like to take vitamin E supplements because they may protect against heart disease, a daily dose of 200 mg of natural, RRR-alpha-tocopherol or d-alpha-tocopherol will saturate plasma levels and will likely double tissue concentrations of alpha-tocopherol in two years. This level of vitamin E intake is probably sufficient to provide significant benefit, and limited enough to avoid potential risks.


Selenium is an essential component of at least 11 selenoenzymes or selenoproteins. There are two major families of selenoenzymes-glutathione peroxidases and deiodinases. The metabolic function of the glutathione peroxidases is to convert oxidized fat (lipid hydroperoxides), which are generated as the result of normal metabolism and contribute to heart disease and stroke, to less harmful compounds. This activity is similar to the antioxidant activity of vitamin E. The deiodinase enzymes regulate the metabolism of thyroid hormones. Interestingly, the recently discovered selenoenzyme thioredoxin reductase has been suggested to play a role in vitamin C metabolism. A human disease known to be caused by selenium deficiency and found in various regions of China is Keshan disease, a cardiomyopathy (disease of the heart muscle) in children.

Extensive data from studies in China have been used to establish the required and safe levels of selenium for humans. These studies revealed significant correlations between daily selenium intake and the selenium content of blood, breast milk, and urine. Significant correlations were also observed between urinary, plasma, hair, fingernail, and toenail selenium levels.

Fingernail brittleness and hair loss were used by the Chinese scientists, as well as the Panel, as the main criteria for chronic selenium toxicity, or selenosis, which occurs at an intake of about 5 mg (5,000 mcg) of selenium daily. Adverse effects were observed at daily dietary selenium intakes between about 600 and 1,600 mcg. The maximum safe dietary selenium intake was calculated to be about 800 mcg/day, but may be as low as 600 mcg in some individuals. The Panel set the UL for selenium at 400 mcg/day, which was selected to protect sensitive individuals. The Chinese scientists suggested a level of about 40 mcg daily as the minimum requirement, which is similar to the new RDA of 55 mcg/day. This RDA established by the Panel is based on the saturation of plasma glutathione peroxidase. An intake of less than 11 mcg daily of selenium will definitely put one at risk of deficiency.

Early epidemiological studies suggested a possible inverse relationship between selenium intake in humans and the incidence of certain cancers. More than 100 relevant experiments with animals exposed to various chemical and viral carcinogens have been carried out. The majority of these studies showed anticancer effects of selenium. Three human trials on selenium and cancer have been completed, and all of them showed positive results. In one trial, the addition of selenium to table salt significantly reduced the incidence of liver cancer in a Chinese population. After 5 years of supplementation with selenium, vitamin E, and beta-carotene, the incidence of stomach and esophageal cancer in another Chinese population was significantly reduced. However, it is not clear which supplement was mainly responsible for this effect. A study in the U.S. showed that 970 men supplemented with 200 mcg of selenium daily (as selenium-enriched yeast) for 4.5 years had a 63% reduction in the incidence of prostate cancer, as well as a significantly reduced incidence of colorectal, lung, and total cancers. These supplementation studies are consistent with a recent study showing one-half to two-thirds reduction in the risk of prostate cancer among men with the highest selenium status, as assessed by toenail levels of selenium that indicate long-term selenium intake. Overall, the evidence that selenium can lower the risk of prostate and possibly other human cancers was considered very promising by the Panel, but it concluded that there is currently no proof for an anticancer effect of selenium.

It is estimated that Americans consume about 100 mcg/day of dietary selenium. In the aforementioned prostate cancer study, subjects were given 200 mcg supplements daily, which boosted their estimated daily intake to about 300 mcg. To prevent selenium deficiency symptoms, a daily intake of 55 mcg is required. For maximal protection against certain cancers, a total daily intake of 200-300 mcg is probably necessary.


Beta-carotene, a yellow-orange compound found in various fruits and vegetables, is a precursor of vitamin A (retinol). It has been suggested to act as an antioxidant in humans. More than 30 human studies have found a relationship between dietary intake of beta-carotene, either determined by diet history or blood levels, and lower rates of cancer, especially lung, oral cavity, throat, and cervical cancers. The studies also suggest a 30-60% reduced risk of heart disease in individuals consuming diets high in beta-carotene. These relationships and the antioxidant properties of beta-carotene were used as the rationale to conduct beta-carotene supplementation studies in humans. The results from these studies, however, do not support the hypothesis that beta-carotene is the major dietary component in fruits and vegetables responsible for protection against chronic diseases.

In a skin cancer study, 1,800 individuals were supplemented daily with 50 mg beta-carotene for five years, but no protective effect was observed. In another study, patients with previous cancer of the large intestine were given vitamin C and vitamin E with or without beta-carotene. The rate of polyp reoccurrence was not affected by beta-carotene. A similar colon polyp prevention trial in Australia that combined beta-carotene with low fat or high wheat bran diets also found no protection by beta-carotene.

Evidence that beta-carotene supplementation is ineffective against cancer and might actually be deleterious was provided by the Alpha-Tocopherol Beta-Carotene (ATBC) Cancer Prevention Study in Finland. In this study of over 29,000 men, mostly heavy smokers, supplementation with 20 mg/day of beta-carotene for seven years significantly increased the risk of developing lung cancer. The CARET (Beta-Carotene and Retinol Efficacy Trial) was conducted in the U.S. with over 18,000 men and women at high risk for lung cancer. This study was terminated early when it became evident that daily supplementation with 30 mg beta-carotene and 25,000 IU vitamin A was increasing the rate of lung cancer. Finally, in the Physiciansí Health Study, supplementation of 22,000 physicians with 50 mg beta-carotene every other day had no effect on the incidence of cancer and heart disease, both in smokers and non-smokers.

The negative results of these intervention studies may have been influenced, however, by the type of beta-carotene used. Natural beta-carotene (a mixture of all-trans and cis) may be a more effective antioxidant than synthetic beta-carotene (all-trans), which is more commonly found in supplements. While people at high risk for lung cancer, such as smokers, should refrain from beta-carotene supplementation, we donít yet know if supplementation in people not at risk for lung cancer is prudent.

We agree with the Panel that, based on these studies, it is not possible to make recommendations in support of beta-carotene supplementation. Other carotenoid components, such as alpha-carotene or lycopene, perhaps in combination with phytochemicals and vitamins, may be responsible for the protection afforded by diets rich in fruits and vegetables.

In summary, the best and most consistent evidence for optimum human health is to consume 5-9 daily servings of fruits and vegetables, which will provide adequate amounts of vitamin C (more than 200 mg) and carotenoids. Additionally, abstinence from tobacco use and regular exercise are two critical components of a healthy lifestyle. We also suggest daily supplements of 200 mg of natural vitamin E and 100-200 mcg of selenium, especially for men at increased risk of prostate cancer. Vitamin E and selenium supplements should not exceed 1,000 mg/day and 400 mcg/day, respectively. Beta-carotene supplementation is not advisable. Even very large doses of vitamin C do not exert any adverse health effects. While 120 mg/day of vitamin C appears sufficient to help prevent heart disease and cancer in young, healthy adults, considerably larger doses may be required to maintain optimum levels in older adults and to treat disease.