点击显示 收起
Carolina Population Center
University of North Carolina
123 West Franklin Street
Chapel Hill, NC 27516
E-mail: popkin{at}unc.edu
University of Connecticut
Human Performance Laboratory
Storrs, CT
Louisiana State University Medical Center
and
Pennington Biomedical Research Center
Baton Rouge, LA
Johns Hopkins University
Baltimore, MD
Linus Pauling Institute
Oregon State University
Corvallis, OR
Harvard School of Public Health
Boston, MA
Dear Sir:
We want to thank Weaver et al for their letter, which provides us with a chance to clarify a few issues that they raised. Our article (1) identified the conceptual and practical issues surrounding the role of beverages in the US diet. Except for alcohol, the 2005 Dietary Guidelines Committee did not discuss in detail the health effects of different beverages, concentrating instead on their nutrient content. Thus, we regard our effort as complementary to that of the Dietary Guidelines Committee. Unlike the broad goals of the Dietary Guidelines Committee and the Institute of Medicine (IOM) Dietary Reference Intakes committees, our beverage guidance panel's effort is focused on the overall role of beverages in the diet. We are concerned that 21% of calories in the American diet are derived from beverages, and we believe that this is an important factor in excessive energy intake and the obesity epidemic. Both the Dietary Guidelines Committee and our panel agree that obesity is one of the major public health problems in the United States, and both emphasize the need to control caloric intake. The 2005 Dietary Guidelines aimed to insure "adequate nutrient intake within caloric needs." We believe that reducing the consumption of caloric beverages is a key step in controlling energy intake, because caloric beverages represent almost 50% of the increase in caloric intake over the past 10 y. Because evidence suggests that energy consumed in fluid form is, at best, only weakly related to satiety and elicit only a small compensatory response, it is important to provide more specific guidance to the public regarding beverage intake to reduce total energy intake. This point has not been addressed in detail in the Dietary Guidelines Committee report and is not mentioned in Weaver's letter. The 3 key issues raised in the letter are addressed below. Guidance about beverage intake should consider both water and energy needs and ways to balance the critical tradeoffs between nutrients and calories contained in beverages and foods.
Water
Although recognizing that fluid needs vary greatly from day to day among individuals, the IOM Panel on Water and Electrolytes defined an adequate intake (AI) for water of 3.7 L/d for men and 2.7 L/d for women (2). Although the AI is not an indicator of actual requirement, it is used to provide guidance regarding recommended intakes. The IOM Dietary Reference Intakes reports presented important nutrient requirements, including calcium, in terms of AI instead of the Estimated Average Requirement. And although it is true that thirst can regulate hydration in healthy people, this relation becomes complex when fluids contain energy (calories), caffeine, or a sweet taste. The IOM Water and Electrolytes report suggests that "automatic" fluid intake can increase significantly when sweetener or flavor is added to plain water. Thus, considering the problem of obesity and the dramatic increase in caloric beverage use, we believe that US citizens need guidance regarding the amount and type of beverages to consume. And, contrary to the assertion by Weaver et al that our advice regarding water consumption has no scientific basis, our advice is entirely based on data from the IOM Water and Electrolytes report. The data we used in Figure 2 is based on data presented in Table 4.14 of the IOM Water and Electrolytes report, which presents a menu plan that fulfills all nutrient needs and the AI for water for an adult who consumes 2200 kcal/d (2). Our example does not recommend 20–50 fl oz of water, but states that water may range from 100% of fluid needs to 20–50 fl oz when other beverages are consumed.
The IOM report did not measure or examine the issues of adequacy of water intake or the distribution of water consumption in the United States. The IOM report examined data regarding median water intake and noted that this median level was adequate. Because limited studies have focused on hydration and tonicity in large populations, and because minimal agreement exists regarding how to measure hydration status, the IOM explicitly did not discuss these matters. More data on hydration levels are needed, particularly in the elderly (3, 4).
Tea and coffee
The 2005 Dietary Guidelines Committee did not address the health effects of tea and coffee consumption. Research on the health effects of tea and coffee is still evolving, and more evidence is needed to make categorical recommendations. We feel our measured conclusions reflect this. We stated that "the current data suggest that consumption of 3 cups of black tea/d may modestly decrease the risk of myocardial infarction" and "recent evidence suggests that tea consumption improves endothelium-dependent vasodilation." We also concluded that "despite these intriguing results, the potential health benefits of flavonoids in tea ... remain to be fully explored." We additionally discussed several prospective cohort studies that showed "significant inverse associations between regular coffee consumption and the risk of type 2 diabetes." On the other hand, we found no evidence of adverse effects of tea or coffee consumption within the limits of moderate daily caffeine intake, and thus we recommended consumption of unsweetened coffee or tea, next to water. Note that the suggested intake range for tea and coffee begins at zero, indicating that no consumption at all is also acceptable.
Milk
The contribution of milk to the intake of several essential nutrients in the United States is well recognized, as acknowledged in our article. But we would like to point out that neither the 2005 Dietary Guidelines Committee nor the Water and Electrolytes Panel consider milk indispensable. The 2005 Dietary Guidelines accept the possibility of a milk-free diet, stating "those who avoid all milk products need to choose rich sources of the nutrients provided by milk," and refer the reader to an appendix where such alternative sources are listed. The Water and Electrolytes Panel, after mentioning the multiple sources of fluids in the human diet (including milk), states that "while all these can contribute to meeting the Adequate Intake [for water], no one source is essential for normal physiologic function or health." The current US reliance on milk products to fulfill human needs for several key nutrients is not the result of a biological imperative, but rather the result of behavioral, cultural, and economical factors. Weaver considers the fulfillment of potassium needs from a diet without milk "particularly problematic." But the contribution of milk to potassium intake has more to do with the poor intake of other potassium-rich foods than to the inherent properties of milk. For example, although 1 cup of milk provides 382 mg potassium, one-half cup of cooked beet greens provides 655 mg, the same amount of spinach yields 419 mg, and white beans 595 mg (all figures taken from the 2005 Dietary Guidelines Committee report). In many cases, consuming the same amount of potassium from these sources results in a lower caloric intake than if consumed from nonfat milk. In addition, some studies have suggested possible adverse effects of long-term milk consumption in the adult population, such as the risk of prostate cancer. The role of milk on adults' diet should continue to be studied in an unbiased and rigorous manner.
Finally, we wish to stress that our contribution sought to highlight the key role that beverages play in the relation between the current US diet and health. Future Dietary Guidelines committees should consider the implications of using liquid energy and nutrient sources to fulfill dietary requirements. There are important differences in the human physiologic response to solid and fluid calories. First, evidence suggests that fluid calories are not as easily "recognized" by systems regulating energy balance as are solid calories, and thus they are more likely to result in overconsumption. Second, day-to-day fluid needs vary widely in a person, whereas nutrient requirements are relatively stable on a day-to-day basis. Thus, an optimal diet should fulfill the relatively stable daily nutrient and energy needs while allowing for variations in fluid consumption. This can be best achieved by providing much of the nutrient needs from solid foods and fulfilling fluid needs with water or other noncaloric fluids. Of course, variety and taste are very important, and it is advantageous to allow for a variety of noncaloric fluids in addition to water.
A primary purpose of our panel was to highlight the potential role of calories from beverages in the current epidemic of obesity and the difficulty of compensating for these calories by reducing energy from solid foods when energy from beverages increases. These are important issues for the Dietary Guidelines Committee to consider when the next version of the Dietary Guidelines is prepared, and we welcome this dialogue with the nutrition community to help develop this focus. We hope that this exchange of ideas will contribute to the clarification of these important issues.
ACKNOWLEDGMENTS
BMP, LEA, GMB, BC, and WCW do not have any advisory board affiliations or financial or personal interests in Unilever. BF is a member of a scientific advisory board of Unilever.
REFERENCES