April 4, 2006
Robert E. Brackett, Ph.D.
Director, Center for Food Safety and Applied Nutrition
U.S. Food and Drug Administration
5100 Paint Branch Parkway
College Park, MD 20740-3835
Dear Dr. Brackett:
Thank you for your March 21, 2006 response to our letter regarding benzene in soft drinks. While your letter has helped us better understand the FDA's actions and position on this issue, it has mainly served to heighten our concerns about potential health risks from benzene in beverages and the approach FDA has taken to investigate those potential risks and share them with the public.
Your letter asserts that we have not provided a rationale for our concern that benzene in beverages could pose a clear health threat and we appreciate the chance to summarize that rationale. Perhaps the World Health Organization expresses our concerns most aptly when stating, "Benzene is carcinogenic to humans and no safe level of exposure can be recommended." Benzene is directly linked to leukemia and it also crosses the placenta to the fetus at levels greater than or equal to the amount in the mother's blood. Benzene is a commonly detected industrial pollutant in the food supply, making additional and avoidable exposures of even greater concern. And in most beverages, benzene is completely unnecessary and readily avoidable, which only raises our concerns about the FDA's unwillingness to pursue this objective.
Because of these widely recognized health risks, food safety officials generally consider benzene contamination of beverages a serious health issue. In Great Britain just this weekend, four products were immediately removed from store shelves. According to the Times of London (4/1/06), "The Food Standards Agency (FSA) said last night that it wanted urgent talks with the soft drinks industry to ensure that all products meet the legal level for tap water of one part per billion." In contrast to the US FDA, which has withheld all of its current test results, "The Food Standards Agency (FSA) rushed out results yesterday of tests on 149 drinks including a range of fruit juice, iced tea, squash, fizzy and low-sugar drinks."
Our concern about FDA's failure to disclose information to the public is not allayed by your observation that agency scientists published results from its laboratory testing in a technical, scientific journal three years after the problem of benzene in beverages came to the government's attention. While the article presents, in the authors words, "limited data" on benzene levels in soft drinks, it is primarily about the development of analytical methods to detect benzene in food, as suggested by its title, "Survey of Benzene in Foods by Using Headspace Concentration Techniques And Capillary Gas Concentrations."
The paper does report the results of seven samples of soft drinks, and one each of iced tea and fruit punch, with levels of benzene ranging from less than one to two parts per billion. This tiny sample, however, is hardly sufficient to support your conclusion that benzene in the food supply does "not suggest a safety concern." Of perhaps greater concern to anyone who might have read the paper are the results of tests done by the agency where 266 parts per billion of benzene were reported in solutions containing the amounts of ascorbic acid and benzoate preservatives "typical of those used in beverage formulations" that sat for eight days at room temperature in a dark cabinet. The authors do not say how long the drinks they tested were stored or under what conditions, a critical piece of missing information.
You further indicate that FDA has concluded that benzene in beverages is not a health concern because the "vast majority" of soft drinks tested recently were contaminated with less than 5 parts per billion of benzene, the U.S. federal drinking water standard. But the agency's own Total Diet Study (TDS) does not support this claim when individual foods are examined. For example, between 1995 and 2001, FDA tested 24 samples of diet soda for benzene; 19 (79 percent) were contaminated with benzene above 5 parts per billion. The average benzene level was 19 ppb, nearly four times the 5 ppb tap water. The maximum detection was 55 ppb, 11 times the tap water limit.
|Composite Food||Number of Composite Samples||Number of Detections||Number of samples > 5 ppb||Average Concentration (ppb)||Highest Concentration (ppb)|
|fruit drink, from powder||2||1||
|low-calorie cola carbonated beverage||24||20||19
|grapefruit juice, from frozen concentrate||2||1||1
|cola carbonated beverage||24||5||3
|fruit-flavored carbonated beverage||24||9||4
|orange juice, from frozen concentrate||24||5||5
Source: EWG compiled from FDA Total Diet Study, 1995-2001
Test results for other drinks revealed the presence of very elevated benzene levels in some other beverages. One cola sample, which like all samples in the TDS was made from three different bottles of cola, was contaminated at 138 parts per billion, 27 times the 5 ppb tap water limit. One powdered fruit drink sample had 95 parts per billion. Orange and grapefruit juice from concentrate were also contaminated with benzene at levels above FDA's 5 ppb level of concern.
It is important to note that levels above 5 ppb in tap water would be illegal and would not be allowed for consumption under federal law. In fact, bottled water with greater than 5 ppb benzene could not be sold, but diet soda, for example, with the same amount of benzene is perfectly legal, although it is clearly no safer.
Several states have adopted tougher tap water standards for benzene than the federal government. New Jersey allows just 1 ppb of benzene in tap water, and California has a drinking water standard that is 33 times more protective than the US EPA at 0.15 ppb.
FDA's data strongly suggest that some, if not many, beverages on the market today have benzene levels well above the 5 parts per billion federal tap water standard. The majority may have levels above the 1 ppb safety standard used as a legal limit for tap water by several states and national governments. Yet for some reason the agency refuses to disclose information to the public on how often, in what products, and under what condition these elevated levels of benzene occur.
Managing this information in a manner that leaves basic questions unanswered for 15 years serves mainly to insulate beverage companies and FDA from consumer concerns. Obviously, few consumers can be expected to have read a scientific article published in a relatively obscure technical journal. And even if they had, they would not have come away with information that they could use to minimize their benzene exposure. The TDS results analyzed above are buried deep within an obscure database on the FDA's website that requires programming skills to access and analyze. It cannot be considered information that is useful or accessible to the average consumer.
What consumers most want to know is whether or not elevated benzene levels are a problem in any brands of beverages they routinely drink or give to their children. They want you to tell them whether they can avoid exposures to elevated levels of benzene by avoiding consumption of beverages that have been stored under certain conditions (beverages with the relevant reactant chemicals that may have been stored for extended periods at elevated temperatures), and they want to know whether there are certain types of drinks—perhaps diet drinks or fruit drinks—that they can avoid in order to minimize benzene exposure.
Although we would much prefer that the FDA simply make public the complete results of its testing programs, Environmental Working Group will file a Freedom of Information Act request seeking all of the agency's test results for benzene in food and beverages. We look forward to the release of this information.
Sr. Vice President