点击显示 收起
Canadian Sugar Institute
10 Bay Street
Suite 620
Toronto
Ontario M5J 2R8
Canada
E-mail: rkaplan{at}sugar.ca
Dear Sir:
A "beverage guidance system" for the United States was recently proposed that ranks beverages based on caloric and nutrient content and an assessment of health benefits and risks (1). Water was ranked as the preferred beverage, followed, in order, by tea and coffee, low-fat and skim milk and soy beverages, noncalorically sweetened beverages, beverages with "some nutritional benefits" (juices, whole milk, alcohol, and sports drinks), and calorically sweetened, nutrient-poor beverages. Unfortunately, the usefulness of the ranking is questionable because it is based on selective evidence.
Ranking beverages from the lowest to highest "value" is based on the unfounded belief that beverages are inherently more or less healthy regardless of their context within an overall diet. This approach is analogous to ranking foods such as cheese, strawberries and chicken from healthiest to least healthy. Ranking these foods is nonsensical because each provides a range of nutrients that can contribute to a healthy diet, and their importance to overall eating is dependent on individual requirements and the content of the rest of the diet. The approach is not consistent with the fact that overall patterns of eating over time, rather than individual foods or beverages, determine whether a diet is relatively healthy or unhealthy.
The major factor used to rank beverages was caloric content, with higher-calorie beverages generally ranked lower than lower-calorie beverages, with the intention that this would help reduce energy intake and obesity. Data in support of this hypothesis was presented; however, evidence suggesting that this may not be the case or that data are equivocal was ignored.
The authors suggest that higher-calorie beverages, particular calorically sweetened soft drinks, contribute directly to obesity because intakes have increased concurrently with obesity rates and calories from liquids are less satiating than are those from solids; however, other evidence suggests the putative relation between soft drinks and obesity may be unfounded. The authors noted that soft drink intake increased between 1977 and 1996 concurrently with obesity rates; however, they failed to point out that there has actually been no increase in per capita consumption (disappearance) of sweetened soft drinks over the past decade (1995–2004) (2), a time when obesity rates have risen dramatically (3). During this time, disappearance of other beverages is also not consistent with the supposed link between caloric content of beverages and obesity. Disappearance of bottled water, diet soft drinks, coffee, and wine increased, that of milk and tea decreased, and that of fruit and vegetable juices and drinks, beer, and distilled spirits remained stable. The authors also ignored comprehensive reviews, including the 2005 Dietary Guidelines Advisory Committee Report, that have concluded that data suggesting that calories of liquids are less satiating than those of solids are conflicting and inconclusive (4-7). A recent systematic review of 6 proposed mechanisms that could explain a link between soft drinks and obesity also concluded that current evidence is inconclusive (6). These authors noted that "even though some evidence supported a relation between sweetened beverages and obesity, other studies supported no such relation, and a few indicated a negative relation." Six studies provided strong evidence for no relation between soft drinks and obesity.
In addition to caloric content, nutrient content was used to rank beverages; however, this approach is flawed because it suggests that it is always preferable to consume more nutrients. The fact that one beverage contains more nutrients than another does not mean that it is inherently healthier. Rather, the importance of the nutrients in a beverage will depend on whether an individual has obtained his requirement from other sources, because there is no known benefit to consume nutrients in amounts exceeding requirements (8). A beverage containing vitamin C, for instance, may be important for an individual who does not consume vitamin C from other sources but unimportant for an individual who meets his requirement from other foods. Thus, the premise that more micronutrient dense beverages are preferable for all individuals, regardless of the total diet and energy needs, is misguided.
The authors contend that calorically sweetened beverages provide "no nutritional benefit" and that there is no specific need to consume these beverages. In fact, calorically sweetened beverages contribute water and carbohydrate, which are both essential nutrients. Although true that these nutrients can be obtained from other foods, the same can be said of any food or beverage. Milk, eggs, broccoli, or any other food is not required to meet nutritional needs. A diet that meets energy and nutrient needs and reduces the risk of chronic disease can be consumed without any one or all these foods. A wide variety of foods and beverages can make up a healthy diet. Thus, ranking calorically sweetened beverages low because they are not "needed" is a meaningless argument, one which the authors also use to suggest that fruit juices should be ranked low.
Another claim by the authors is based on selective evidence. Non-calorically sweetened beverages were ranked lower than water, tea, and coffee partly on the basis of what the authors refer to as a "new literature" in rodents suggesting that high sweetness in beverages may contribute to conditioning for a higher preference for sweetness. However, evidence in humans suggests that this is not the case: children have an innate preference for sweet tastes that naturally declines in adolescents and adults (9), and restricting sweet tasting foods increases, rather than decreases, preference for sweet tastes (10). Moreover, most studies show no difference in preference for sweet tastes between obese and nonobese adults, suggesting that this trait may not affect body weight (9).
Another concern is the ideal pattern of beverage consumption proposed. No evidence is provided to justify the recommended ranges of intakes for each beverage category, resulting in ranges that appear to be arbitrary. Why, for instance, is the maximum intake level for low-fat and skim milk and soy beverages 16 fl oz/d as opposed to 15 or 17 fl oz/d? Furthermore, the authors do not consider different needs for different age or sex groups or for varying energy requirements. The importance of fluids, calories, and nutrients depend on these factors.
In conclusion, the proposed beverage guidance system is based on selective evidence and the belief that some beverages are inherently healthier than others regardless of overall diet. An evidence-based approach and more research are required to assess the influence of beverages on health.
ACKNOWLEDGMENTS
The author is an employee of the Canadian Sugar Institute.
REFERENCES