G.F. Combs, Jr.
Division of Nutritional Sciences and
Department of Animal Science
Cornell University

The desire to reduce the numbers of preventable deaths in the U.S. are driving efforts to alter American diets. Two types of disease are most important in this regard: heart disease and cancer. These constitute the leading causes of death in the U.S. The etiologies of each are now seen as being effected by diet and/or diet-related factors. The National Academy of Sciences has estimated that at least one-fifth of heart disease could be prevented by improving the American diet (1); the National Cancer Institute (NCI) has estimated that more than one-third of all cancer deaths could be similarly prevented. Most recommendations for such dietary improvements have involved increasing the consumption of fruits and vegetables. Such recommendations involve reductions in the intakes of saturated fats, which have been identified as risk factors to heart disease. They are also based on a substantial body of accrued evidence of inverse associations between intake of fruits and vegetables and risk to various cancers (2-4). The most visible effort in this regard has been the 5 A Day for Better Health Program that was instituted in 1991 by the NCI, with joint support from food industry groups, to implement the national health promotion and disease prevention objectives (5) of encouraging Americans to eat 5 or more servings of fruits and vegetables each day. This program has been implemented at the state level under various names (e.g., "5-A-Day", "High Five", "Gimme 5").

The concept underlying the 5-A-Day programs departs from the traditional paradigm within which dietary guidance has been developed, in as much as it focuses on the promotion of overall health and well-being rather than on only the assurance of quantitative nutrient intakes. A marker of this change was the development nearly 20 years ago of the Dietary Goals for the United States by the U.S. Senate Select Committee on Nutrition and Human Needs (6). Those quantitative goals, although greeted with controversy, were proposed as measures to promote health. The ensuing discussion and the research it stimulated revealed that the typical diets of Americans were very different from the kinds of diets that would be necessary to achieve the Dietary Goals (7,8). Analyses based on estimates of nutrient contents of common foods indicate that the Dietary Goals can be achieved with diets that provide daily averages of at least 5 servings of vegetables and fruits (9). That this level of fruit and vegetable intake may be important in reducing risk to chronic disease is suggested not only by findings in cancer epidemiology (2-4), but also by research that has shown the antioxidant nutrients, some of which these foods tend to provide, to play important roles in support of general health. Indeed, the antioxidant nutrients vitamin E, vitamin C, selenium (Se) and, perhaps, -carotene are know to function in the metabolic protection from oxidative stress by scavenging and minimizing the formation of very highly reactive oxygen-derived free radicals (10). Thus, they are thought to be important in promoting health by reducing the cumulative oxidative damage of cells. Evidence shows that oxidative damage is involved in the etiologies of atherosclerosis, cancer, aging, rheumatoid arthritis, inflammatory bowel diseases and neurodegenerative diseases. Therefore, the emphasis on fruits and vegetables has support from both empirical observations as well as basic science.

Despite this emerging picture of health benefits of diets richer in fruits and vegetables and the antioxidants that they can provide, surveys have shown that the regular intakes of fruits and vegetables of many Americans continue to fall far short of the 5-A-Day goal. In 1991 (at the beginning of the "5-A-Day" program), only 23% of the general adult population were consuming that level of fruits and vegetables (11). Further, several demographic groups were found to consume particularly low amounts of these foods: men tended to eat one fewer serving than women (3 vs. 3.7 servings per day, respectively); Hispanics ate fewer servings than either whites or blacks (3 vs. 3.4 vs. 3.4 servings per day, respectively); younger people ate fewer servings than older people (3 vs. 3.4 vs. 3.6 vs. 4.1 servings per day for 18-34 yrs., 35-49 yrs., 50-64 yrs. and 65+ yrs., respectively). These trends would appear to be related to awareness of the health value of fruits and vegetables, as the lowest rates of consumption were observed among lower socio-economic groups and among those persons who were not aware of the health benefits attached to fruits and vegetables.

That the list of most frequently consumed fruits and vegetables (Table 3) is short shows that few Americans tend to consume servings of these foods more than once per week. In general, one-fourth of Americans surveyed regularly eat vegetables cooked in some type of fat. This rate is clearly higher among individuals of lower socio-economic status (33%) and among blacks (37% vs. 26% for Hispanics and 22% for whites). In addition, more than one-fifth reported adding cheese or cream sauces, butter, margarine or oil to cooked vegetables; and most (85%) added to green salads (the most frequently consumed vegetable dish) some sort of salad dressing only 61% of which is of a reduced-fat type. These data show significant increases (an estimated 28.5% for vegetables and 38% for non-citrus fruits) over the levels reported in the 1970's, which may reflect such trends as the emergence of salad bars in restaurants, the development of new varieties, increases in imports and improvements in consumer selection year-round. Despite these positive trends, Americans tend to eat relatively few servings of fruits and vegetables, particularly those of low-fat content.

New Yorkers appear to be following national trends in that they, too, tend to consume relatively few fruits and vegetables on regular bases. The 1992 New York State Nutrition Surveillance Program (12) found that three-quarters of adults reported eating fewer than the recommended numbers of servings of foods
in several categories (Table 4). The underconsumption of healthful foods was particularly great among
older adults and teens. Data of this sort show the need to educate Americans of all ages regarding the
health significance of their diet choices.
Current dietary guidance is based on the dietary guidelines developed in 1980 jointly by the US Departments of Agriculture (USDA) and of Health and Human Services (DHHS) (13), and the revisions that have been made as recently as 1990 (14). The most useful instrument for this purpose has been the Food Guide Pyramid, a graphic representation of those dietary guidelines (15). This graphic was developed to communicate the concepts of variety (in food choice), proportionality (in selection from various food groups) and moderation (in use of fats and sugars). Extensive testing showed the final version to be informative in this regard to most groups of Americans.

The Food Guide Pyramid has been useful in communicating the likely health benefits for many Americans associated with increasing their regular intakes of fruits and vegetables. Unfortunately, this emphasis has been misinterpreted by some as a suggestion to eliminate foods derived from animals. In fact, that suggestion is contrary to the rationale upon which the Food Guide Pyramid was based, i.e., it relates to the variety of foods currently consumed by Americans. It's planners acknowledged that limiting meat choices, for instance, may result in diets providing less than recommended amounts of iron and zinc and that eliminating dairy products will lead to special difficulties in obtaining calcium.

While it is generally accepted that vegetarian, but not fruitarian, diets can be nutritionally adequate if sensibly selected, it is clear that problems can arise in either vegetarian or non-vegetarian diets if the variety of foods making up the diet is restricted and, particularly, if the consumption of dairy products is low. Animal products, in fact, provide important amounts of some key nutrients in the typical non-vegetarian American diet (e.g., Table 5): vitamins A, B6 and B12, thiamin, niacin, calcium, iron, zinc, iodine and selenium in addition to taurine and docosahexanoic acid (22:6n-3). Therefore, it is possible that over-emphasis on what McDonald has called the "trendy admonition" (19) to avoid eating red meat (in order to avoid saturated fat) may have negative health consequences. New York survey data (12) suggest that substantial numbers of people already consume less than RDA-amounts of vitamins and minerals provided in their low-fruit and vegetable diets (Table 6). That study further showed that, with the exception of vitamin C for women, the prevalences of low nutrient intakes were greater for older (ø65 yrs.) than younger (50-64 yrs.) adults.

Data are not as complete for the antioxidant nutrients. Clearly, there is no evidence for frank deficiencies of vitamin E, vitamin C or Se at any appreciable level in the U.S. today. However, these nutrients may have beneficial health roles that are discharged in ways that are not directly related to their physiologic functions which are usually manifest at critically limiting levels. Therefore, it is probable that these nutrients may be beneficial at levels of intake substantially greater than those minima required to prevent their respective specific clinical deficiency syndromes. The "antioxidant hypothesis" for disease would predict that, because high tissue concentrations of such nutrients as vitamins E and C are be beneficial, supranutritional intakes sufficient to achieve such increased tissue levels of either or both nutrients would be desirable.

In the case of Se, a dietary change involving reduce animal products might have the potential to reduce intakes that have hitherto been thought adequate. Selenium is known to be required for several enzymes (e.g., glutathione-peroxidases, iodothyronine 5'-deiodinases) for which intakes of 40-50 µg/day (adults) appear to be sufficient. Intakes below this level have been associated with juvenile cardiomyopathy and exocrine pancreatic disease. Although such very low intakes level are known in other parts of the world (e.g., parts of central China) few Americans appear to consume less than ca. 75 µg Se per day (adults). In fact, many Americans appear to consume >150 µg Se/day (20). Although Se deficiency is not seen as a problem in this country, the nutrient has attracted considerable interest with the emergence of evidence that it may be a risk factor to cancer (21). In fact, result of numerous epidemiological and experimental animal tumor model investigations over the last three decades have indicated cancer-preventative roles for Se-compounds. Evidence of effects of Se-supplementation on cancer incidence in humans is presently emerging. An intervention trial in China suggested a modest protective effect of Se against cancer mortality (22), and our decade-long clinical trial (23) may show stronger protective effects. Americans typically consume about one-half of their total dietary Se by way of meats, beef, pork and chicken (20, 24), most of the balance coming from wheat products. Foods in each of these categories typically contain 20-35 µg Se/100 g, which is 2-3 fold that of beans or other cereal-based foods and 20-30 fold that of fruits and vegetables. Therefore, reductions in the intakes of meats, even if balanced by increases in the consumption of legumes, is likely to result in reductions in total Se intakes unless wheat products are also increased. Clearly, emphasis on fruits and vegetables will not increase and, depending upon how such changes are implemented, may reduce Se intakes.

This raises important questions concerning the impact on overall nutrient intake of programs that emphasizing fruits and vegetables, particularly in the context of reduced intakes of meats (important sources of Se) and replacement of vegetable oils (important source of vitamin E) with reduced fat substitutes. While many fruits and vegetables contain vitamin C, -carotene and/or other carotenoids; few contain appreciable amounts of either Se or vitamin E. Therefore, diet changes involving increases in fruits and vegetables can be expected to increase intakes of only some antioxidants, namely, vitamin C and carotenoids. If such changes also involve reductions in the intakes of meats and vegetable oils, it is likely that they will result in reduced consumption of Se and vitamin E.

With emerging data showing beneficial health effects of supranutritional levels of Se and vitamin E, strategies for diet improvement will bear close examination in order that improved health may be realized. Accordingly, this will mean making even more clear that, in order to attain the health benefits associated with increasing fruits and vegetable consumption and reducing total fat intake, the other food groups remain important as sources of essential nutrients and fiber. In regard to the antioxidant nutrients, it is important to emphasize the importance cereal-based foods, legumes and low-fat diary products and meats as sources of Se and vitamin E.


1. Committee on Diet and Health. 1989. Diet and Health. Implications for Reducing Chronic Disease Risk. National Academy of Sciences, Washington, DC, 749 pp.

2. Steinmatz, K.A. and J.D. Potter. 1991. Cancer Causes Control 2:825.

3. Steinmatz, K.A. and J.D. Potter. 1991. Cancer Causes Control 2:427.

4. Block, G., B. Patterson and A. Subar. 1992. Nutr. Cancer 18:1.

5. U.S. Dept. of Health and Human Services. 1990. Healthy People 2000: National Health Promotion and Disease Prevention Objectives. DHHS publication (PHS) 91-50212, Washington, DC.

6. U.S. Senate Select Committee on Nutrition and Human Needs. 1977. Dietary Goals for the United States, 2nd. ed., U.S. Government Printing Office, Washington, DC, 88 pp.

7. Peterkin, B.B., R.L. Kerr and C.J. Shore. 1978. J. Nutr. Ed. 10:15.

8. Peterkin, B.B., C.J. Shore and R.L. Kerr. 1979. J. Amer. Diet. Assoc. 4:423.

9. Welsh, S.O., C. Davis and A. Shaw. 1993. USDA's Food Guide. Background and Development. USDA misc. publ. no. 1514, 38 pp.

10. Halliwell, B. 1994. Nutr. Rev. 52:253.

11. Subar, A.F., J. Heimendinger, S.M. Krebbs-Smith, B.H. Patterson, R. Kessler and E. Pivonka. 5 A Day for Better Health: A Baseline Study of American's Fruit and Vegetable Consumption. NCI, Rockville, MD, 36 pp.

12. Campbell, C. 1992. New York State Nutrition, Cornell University, Ithaca, NY, 176 pp.

13. U.S. Department of Agriculture and U.S. Department of Health and Human Services. 1980. Nutrition and Your Health: Dietary Guidelines for Americans. USDA Home Garden Bull. No. 232, 20 pp.

14. U.S. Department of Agriculture and U.S. Department of Health and Human Services. 1990. Nutrition and Your Health: Dietary Guidelines for Americans. USDA Home Garden Bull. No. 232, 28 pp.

15. U.S. Department of Agriculture, Human Nutrition Information Service. 1992. The Food Guide Pyramid. Home and Garden Bull. No. 252, 32 pp.

16. Sauberlich, H.E., M.J. Kretsch, H.L. Johnson and R.A. Nelson. 1982. Animal products as a source of vitamins, Chapter 17 in Animal Products in Human Nutrition (D.C. Beitz and R.G. Hansen, eds.), Academic Press, New York, p. 339.

17. Murphy, S.P., A.F. Subar and G. Block. 1990. Vitamin E intakes and sources in the United States. Am. J. Clin. Nutr. 52:361.

18. Subar, A.F., G. Block and L.D. James. 1989. Folate intakes and food sources in the US population. Am. J. Clin. Nutr. 50:508.

19. McDonald, H.B. 1991. Can. J. Publ. Health 82:331.

20. Combs, Jr., G.F. 1988. Adv. Food Res. 32:85.

21. Combs, G.F., Jr. 1995. Selenium and Cancer, in Biochemistry and Molecular Biology of Selenium (Z. Wang, ed.), Chin. Acad. Sci., Beijing

22. Blot, W.J., J.Y. Li, P.R. Taylor, W. Guo, S. Dawsey, G.Q. Wang, C.S. Yang, S.F. Zheng, M. Gail, G.Y. Li, B.Q. Liu, J. Tangrea, Y.H. Sun, F. Liu, F. Fraumeni, Jr., Y.H. Zhang & B. Li. 1993. J. Nat. Cancer Inst. 85:1483.

23. Clark, L.C., G.F. Combs, Jr. & B.W. Turnbull. 1995. unpublished observations.

24. Schubert, A., J.M. Holden and W.R. Wolf. 1987. J. Am. Diet. Assoc. 87:285.